Paper End Detection Apparatus for Image Reading Apparatus and Image Reading Apparatus With Paper Leading End and Trailing End Detection Apparatus

1. Field of the Invention

This invention relates to a paper end detection apparatus for an image reading apparatus suitable for use with an image scanner and an image reading apparatus with a paper end detection apparatus.

2. Description of the Related Art

In recent years, image reading apparatuses such as image scanners have been and are continually being developed in order to input image information to a computer (host computer) apparatus.

In an image reading apparatus of the type mentioned, an optical image reading unit optical reads image information on a paper sheet. In order to transfer data of an image signal obtained by the optical image reading unit, an end of a paper sheet must be detected.

Conventionally, an end of a paper sheet is developed using a photo-interrupter, and thereafter, taking a predetermined margin into consideration, data of the image signal are transferred.

Further, for detection of the trailing end of a paper sheet, a number of pulse signals corresponding to a size of the paper sheet is counted to detect it.

With the conventional detection method, however, since a margin after detection of an end of a paper sheet must be estimated, it is difficult to obtain an accurate data transfer timing. Further, since, for detection of the trailing end of a paper sheet, a number of pulse signals corresponding to a size of the paper sheet is counted to detect it, there is another problem to be solved in that the trailing end of a paper sheet cannot be detected where the paper sheet is of a free size.

It is an object of the present invention to provide a paper end detection apparatus for an image reading apparatus wherein an end of a paper sheet can be accurately detected.

In order to attain the object described above, according to an aspect of the present invention, there is provided a paper end detection apparatus for an image reading apparatus wherein paper image information is read from a paper sheet being transported along a paper transport path by an optical image reading unit at a fixed location along the paper transport path, comprising one-line peak value detection means for detecting, from an image signal from the optical image reading unit, a peak value in one line along a direction perpendicular to the direction in which the paper sheet is transported, storage means for storing a past peak value in one line detected prior to the current peak value detected by the one-line peak value detection means, and comparison means for comparing the current peak value detected by the one-line peak value detection means and the past peak value from the storage means and outputting a result of the comparison as a paper end detection signal.

With the paper end detection apparatus for an image reading apparatus, the leading end and the trailing end of a paper sheet can be detected with certainty with a simple structure, and consequently, reading timings of or a changing over timing of reading between front face data and rear face data of a paper sheet can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

The paper end detection apparatus for an image reading apparatus may further comprise a shift register for storing both of the current peak value detected by the one-line peak value detection means and the past peak value in one line detected prior to the current peak value. Due to such storage by the shift register, a comparison calculation by the comparison means can be performed readily with certainty.

Or, the paper end detection apparatus for an image reading apparatus may be constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means, and the comparison means compares the current peak value detected by the one-line peak value detection means and an addition correction past peak value from the addition means and outputs a result of the comparison as a paper end detection signal. Where the addition correction past peak value is used for comparison in this manner, the leading end and the trailing end of a paper sheet can both be detected with certainty with a simple structure, and consequently, reading timings or a reading changing over timing can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is similarly comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

Alternatively, the paper end detection apparatus for an image reading apparatus may be constructed such that it further comprises subtraction means for subtracting a predetermined value from the past peak value from the storage means, and the comparison means compares the current peak value detected by the one-line peak value detection means and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal. Also where the subtraction correction past peak value is used for comparison in this manner, the leading end and the trailing end of a paper sheet can be detected with certainty with a simple structure, and consequently, reading timings or a reading changing over timing can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is similarly comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

Alternately, the paper end detection apparatus for an image reading apparatus may be constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means, and subtraction means for subtracting a predetermined value from the past peak value from the storage means, and the comparison means compares the current peak value detected by the one-line peak value detection means with an addition correction past peak value from the addition means and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal. The construction allows employment of a common circuit construction for a paper leading end detection circuit and a paper trailing end detection circuit.

In this instance, the paper end detection apparatus for an image reading apparatus may be constructed further such that the comparison means includes first comparison means for comparing the current peak value detected by the one-line peak value detection means and the addition correction past peak value from the addition means with each other, and second comparison means for comparing the current peak value detected by the one-line peak value detection means and the subtraction correction past peak value from the subtraction means with each other, and it further comprises outputting means for outputting, when a paper end detection signal is outputted from at least one of the first comparison means and the second comparison means, the paper end detection signal as a paper end detection signal of the paper end detection apparatus. Also this construction allows employment of a common circuit construction for a paper leading end detection circuit and a paper trailing end detection circuit.

In any case, the paper end detection apparatus for an image reading apparatus may further comprise latching means for latching an output of the comparison means. Provision of the latching means is effective to prevent noise or chattering with certainty.

According to another aspect of the present invention, there is provided an image reading apparatus with a paper end detection apparatus, comprising a paper transport path for transporting therealong a paper sheet from which an image is to be read, an optical image reading unit for optically reading paper image information from the paper sheet being transported along the paper transport path at a fixed location along the paper transport path, a paper end detection apparatus for detecting an end of the paper sheet from an image signal obtained by the optical image reading unit, and image signal processing means for processing the image signal obtained by the optical image reading unit in response to a result of detection of the paper end detection apparatus, the paper end detection apparatus including one-line peak value detection means for detecting, from an image signal from the optical image reading unit, a peak value in one line along a direction perpendicular to the direction in which the paper sheet is transported, storage means for storing a past peak value in one line detected prior to the current peak value detected by the one-line peak value detection means, and comparison means for comparing the current peak value detected by the one-line peak value detection means and the past peak value from the storage means and outputting a result of the comparison as a paper end detection signal.

With the image reading apparatus with a paper end detection apparatus, the leading end and the trailing end of a paper sheet can be detected with certainty with a simple structure, and consequently, reading timings of or a changing over timing of reading between front face data and rear face data of a paper sheet can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

According to a further aspect of the present invention, there is provided a paper end detection apparatus for an image reading apparatus wherein paper image information is optically read from a paper sheet using an optical image reading unit of the type which scans the paper sheet in two perpendicular directions, comprising a first start mark and a second start mark provided on the paper sheet and extending in two perpendicular directions, first start mark detection means for detecting the first start mark from an image signal from the optical image reading unit, second start mark detection means for detecting the second start mark from the image signal from the optical image reading unit, and paper leading end detection signal outputting means for outputting a paper leading end detection signal when both of the first start mark detection means and the second start mark detection means detect the start marks.

According to a still further aspect of the present invention, there is provided an image reading apparatus with a paper end detection apparatus, comprising an optical image reading unit for scanning a paper sheet in two perpendicular scanning directions to optically read paper image information from the paper sheet, a paper end detection apparatus for detecting an end of the paper sheet from an image signal obtained by the optical image reading unit, and image signal processing means for processing the image signal obtained by the optical image reading unit in response to a result of detection of the paper end detection apparatus, the paper end detection apparatus including a first start mark and a second start mark provided on the paper sheet and extending in two perpendicular directions, first start mark detection means for detecting the first start mark from an image signal from the optical image reading unit, second start mark detection means for detecting the second start mark from the image signal from the optical image reading unit, and paper leading end detection signal outputting means for outputting a paper leading end detection signal when both of the first start mark detection means and the second start mark detection means detect the start marks.

With the paper end detection apparatus for an image reading apparatus and the image reading apparatus with a paper end detection apparatus, an end of a paper sheet can be detected accurately from image signals from the two different start marks obtained by the optical image reading unit. Consequently, an accurate data transfer timing can be obtained readily.

Further objects, features and advantages of the present invention will become apparent from the following detailed description when read in conjunction with the accompanying drawings in which like parts or elements are denoted by like reference characters.

FIG. 1 is a block diagram illustrating one aspect of the present invention;

FIG. 2 is a block diagram illustrating another aspect of the present invention;

FIG. 3 is a block diagram of an image data processing system showing a preferred embodiment of the present invention;

FIG. 4 is a schematic side elevational sectional view of an image reading apparatus to which the image data processing system is incorporated;

FIG. 5 is a perspective view showing an outer profile of the image reading apparatus of FIG. 4;

FIG. 6 is a schematic side elevational view showing an outer profile of the image reading apparatus of FIG. 4;

FIG. 7 is a diagrammatic view schematically showing, in side elevation, an arrangement of principal components of the image reading apparatus of FIG. 4;

FIG. 8 is an exploded perspective view schematically showing a driving system of the image reading apparatus of FIG. 4;

FIG. 9 is a diagrammatic view schematically showing, in side elevation, the driving system shown in FIG. 8;

FIG. 10 is a diagrammatic view schematically showing, in plan, the driving system shown in FIG. 8;

FIGS. 11(A) and 11(B) are a schematic plan view and a schematic side elevational view, respectively, showing a paper transport system of the image reading apparatus of FIG. 4;

FIG. 12 is a schematic side elevational view showing the construction of an image reading mechanism of the image reading apparatus of FIG. 4;

FIG. 13 is a diagrammatic view schematically showing the construction of the optical image reading mechanism shown in FIG. 12;

FIG. 14 is a block diagram showing the construction of a white level information correction apparatus to which the present invention is applied together with several associated elements;

FIG. 15 is a block diagram showing a white level algorithm circuit of the white level information correction apparatus shown in FIG. 14;

FIG. 16 is a block diagram showing another form of the white level algorithm circuit of the white level information correction apparatus shown in FIG. 14;

FIG. 17 is a block diagram showing a further form of the white level algorithm circuit of the white level information correction apparatus shown in FIG. 14;

FIG. 18 is a time chart illustrating operation of the while level information correction apparatus shown in FIG. 14;

FIG. 19 is a block diagram showing the construction of another white level information correction apparatus to which the present invention is applied together with several associated elements;

FIG. 20 is a block diagram showing the construction of a further white level information correction apparatus to which the present invention is applied together with several associated elements;

FIG. 21 is a block diagram showing the construction of an outputting section and an output control circuit of the image data processing system shown in FIG. 2;

FIG. 22 is a flow chart illustrating operation of the outputting section and the output control section shown in FIG. 21;

FIG. 23 is a block diagram showing the construction of an original end detection circuit of the image data processing system shown in FIG. 2;

FIG. 24 is a time chart illustrating operation of the original end detection circuit shown in FIG. 23;

FIG. 25 is a sequence diagram illustrating initialization operation of a hopper system of the image reading apparatus of FIG. 4;

FIG. 26 is a sequence diagram illustrating operation of the hopper system in an automatic reading mode;

FIG. 27 is a sequence diagram illustrating operation of the hopper system in a manual insertion mode;

FIG. 28 is a sequence diagram illustrating operation of a transport system of the image reading apparatus of FIG. 4;

FIG. 29 is a similar view but illustrating operation of the transport system at a different stage;

FIG. 30 is a similar view but illustrating operation of the transport system at another different stage;

FIG. 31 is a front elevational view showing an operation panel of the image reading apparatus shown in FIG. 4;

FIG. 32 is a block diagram schematically showing the construction of a control system of the image reading apparatus shown in FIG. 4;

FIG. 33 is a diagrammatic view of part of another image reading apparatus showing another preferred embodiment of the present invention;

FIG. 34 is a diagrammatic view illustrating a location of a start mark provided on a paper sheet;

FIG. 35 is a block diagram showing an image data processing system of the image reading apparatus shown in FIG. 33; and

FIG. 36 is a block diagram of a paper end detection apparatus incorporated in the image reading apparatus shown in FIG. 33.

Referring first to FIG. 1, there is shown an image reading apparatus in which a paper end detection apparatus according to an aspect of the present invention is incorporated. The image reading apparatus includes a paper transport path 310 for transporting therealong a paper sheet 40 from which an image is to be read, and an optical image reading unit 410 for optically reading paper image information from a paper sheet 40 being transported at a fixed location along the paper transport path 310.

The paper end detection apparatus is denoted at 170 and detects an end of a paper sheet 40 from an image signal obtained by the optical image reading unit 410. The paper end detection apparatus 170 includes one-line peak value detection means 171, storage means 172 and comparison means 173.

The one-line peak value detection means 171 detects, from an image signal from the optical image reading unit 410, a peak value in one line along a direction perpendicular to the direction in which the paper sheet 40 is transported.

The storage means 172 stores a past peak value in one line detected prior to the current peak value detected by the one-line peak value detection means 171.

The comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and the past peak value from the storage means 172 and outputs a result of the comparison as a paper end detection signal.

The image reading apparatus further includes image signal processing means 180 for processing the image signal obtained by the optical image reading unit 410 in response to a result of detection of the paper end detection apparatus 170.

The paper end detection apparatus may further comprise a shift register for storing both of the current peak value detected by the one-line peak value detection means 171 and the past peak value in one line detected prior to the current peak value.

Alternately, the paper end detection apparatus may be constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means 172 and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and an addition correction past peak value from the addition means and outputs a result of the comparison as a paper end detection signal.

The paper end detection apparatus may be constructed such that it further comprise subtraction means for subtracting a predetermined value from the past peak value from the storage means 172 and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal.

The paper end detection apparatus may be constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means 172, and subtraction means for subtracting a predetermined value from the past peak value from the storage means 172, and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 with an addition correction past peak value from the addition means and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal.

In this instance, the paper end detection apparatus may be constructed further such that the comparison means 173 includes first comparison means for comparing the current peak value detected by the one-line peak value detection means 171 and the addition correction past peak value from the addition means with each other, and second comparison means for comparing the current peak value detected by the one-line peak value detection means 171 and the subtraction correction past peak value from the subtraction means with each other, and that it further comprises outputting means for outputting, when a paper end detection signal is outputted from at least one of the first comparison means and the second comparison means, the paper end detection signal as a paper end detection signal of the paper end detection apparatus.

The paper end detection apparatus may further comprise latching means for latching an output of the comparison means 173.

In the paper end detection apparatus for an image reading apparatus and the image reading apparatus with a paper end detection apparatus according to the present invention described above, paper image information is optically read from a paper sheet 40 being transported along the paper transport path 310 at the fixed location along the paper transport path 310 using the optical image reading unit 410. Further, the image signal obtained by the optical image reading unit 410 is processed in response to a result of detection of the paper end detection apparatus 170 by the image signal processing means 180.

In this instance, in the paper end detection apparatus 170, an end of the paper sheet 40 is detected from the image signal obtained by the optical image reading unit 410. More particularly, the one-line peak value detection means 171 detects, from the image signal from the optical image reading unit 410, a peak value in one line along a direction perpendicular to the direction in which the paper sheet 40 is transported. The storage means 172 stores a past peak value in one line detected prior to the current peak value detected by the one-line peak value detection means 171. Further, the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and the past peak value from the storage means 172 and outputs a result of the comparison as a paper end detection signal.

The paper end detection apparatus may further comprise a shift register for storing both of the current peak value detected by the one-line peak value detection means 171 and the past peak value in one line detected prior to the current peak value.

Or, the paper end detection apparatus may be constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means 172 and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and an addition correction past peak value from the addition means and outputs a result of the comparison as a paper end detection signal.

Alternatively, the paper end detection apparatus may be constructed such that it further comprises subtraction means for subtracting a predetermined value from the past peak value from the storage means 172 and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal.

The paper end detection apparatus may be constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means 172, and subtraction means for subtracting a predetermined value from the past peak value from the storage means 172, and wherein the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 with an addition correction past peak value from the addition means and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal.

In this instance, the paper end detection apparatus may be constructed further such that first comparison means of the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and the addition correction past peak value from the addition means with each other and second comparison means of the comparison means 173 compares the current peat value detected by the one-line peak value detection means 171 and the subtraction correction past peak value from the subtraction means with each other, and outputting means outputs, when a paper end detection signal is outputted from at least one of the first comparison means and the second comparison means, the paper end detection signal as a paper end detection signal of the paper end detection apparatus.

The paper end detection apparatus may further comprise latching means for latching an output of the comparison means 173.

With the paper end detection apparatus for an image reading apparatus and the image reading apparatus with a paper end detection apparatus according to the present invention described above, the following effects or advantages are achieved.

1. Since the paper end detection apparatus includes the one-line peak value detection means 171 for detecting, from an image signal from the optical image reading unit 410, a peak value in one line along a direction perpendicular to the direction in which the paper sheet is transported, the storage means 172 for storing a past peak value in one line detected prior to the current peak value detected by the one-line peak value detection means 171, and the comparison means 173 for comparing the current peak value detected by the one-line peak value detection means 171 and the past peak value from the storage means 172 and outputting a result of the comparison as a paper end detection signal, the leading end and the trailing end of a paper sheet can be detected with certainty with a simple structure, and consequently, reading timings of or a changing over timing of reading between front face data and rear face data of a paper sheet can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

2. Where the paper end detection apparatus further comprises a shift register for storing both of the current peak value detected by the one-line peak value detection means 171 and the past peak value in one line detected prior to the current peak value, comparison calculation by the comparison means 173 can be performed readily with certainty.

3. Where the paper end detection apparatus is constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means 172 and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and an addition correction past peak value from the addition means and outputs a result of the comparison as a paper end detection signal, the leading end or the trailing end of a paper sheet can be detected with certainty with a simple structure, and consequently, reading timings or a reading changing over timing can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is similarly comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

4. Where the paper end detection apparatus is constructed such that it further comprises subtraction means for subtracting a predetermined value from the past peak value from the storage means 172 and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal, the leading end or the trailing end of a paper sheet can be detected with certainty with a simple structure, and consequently, reading timings or a reading changing over timing can be controlled precisely irrespective of the type of the paper sheet being transported. As a result, it is similarly comparatively easy to transport paper sheets of a variety of sizes to the image reading apparatus so that images of them may be read by the image reading apparatus.

5. Where the paper end detection apparatus is constructed such that it further comprises addition means for adding a predetermined value to the past peak value from the storage means 172, and subtraction means for subtracting a predetermined value from the past peak value from the storage means 172, and that the comparison means 173 compares the current peak value detected by the one-line peak value detection means 171 with an addition correction past peak value from the addition means and a subtraction correction past peak value from the subtraction means and outputs a result of the comparison as a paper end detection signal, a paper leading end detection circuit and a paper trailing end detection circuit can be constructed with a common circuit construction.

6. Where the paper end detection apparatus is constructed further such that the comparison means 173 includes first comparison means for comparing the current peak value detected by the one-line peak value detection means 171 and the addition correction past peak value from the addition means with each other, and second comparison means for comparing the current peat value detected by the one-line peak value detection means 171 and the subtraction correction past peak value from the subtraction means with each other, and that it further comprises outputting means for outputting, when a paper end detection signal is outputted from at least one of the first comparison means and the second comparison means, the paper end detection signal as a paper end detection signal of the paper end detection apparatus, a paper leading end detection circuit and a paper trailing end detection circuit can be constructed with a common circuit construction readily.

7. Where the paper end detection apparatus further comprises latching means for latching an output of the comparison means 173, noise or chattering can be prevented with certainty.

Referring now to FIG. 2, there is shown, in block diagram, an image reading apparatus in which a paper end detection apparatus according to another aspect of the present invention is incorporated. The image reading apparatus comprises an optical image reading unit 411 which scans a paper sheet in two perpendicular scanning directions to optically read paper image information from the paper sheet.

The image reading apparatus further comprises a paper end detection apparatus 170 which detects an end of a paper sheet from an image signal obtained by the optical image reading unit 411. To this end, the paper end detection apparatus 170 includes a first start mark SM1 and a second start mark SM2, first start mark detection means 174, second start mark detection means 175, and paper trailing end detection signal outputting means 176.

The first start mark SM1 and the second start mark SM2 are provided on the paper sheet and extend in two perpendicular directions.

The first start mark detection means 174 detects the first start mark SM1 from an image signal from the optical image reading unit 411.

The second start mark detection means 175 detects the second start mark SM2 from the image signal from the optical image reading unit 411.

The paper leading end detection signal outputting means 176 outputs a paper leading end detection signal when both of the first start mark detection means 174 and the second start mark detection means 175 detect the start marks.

The image forming apparatus further comprises image signal processing means 180 for processing an image signal obtained by the optical image reading unit 411 in response to a result of detection of the paper end detection apparatus 170.

In the paper end detection apparatus for an image reading apparatus and the image reading apparatus with a paper end detection apparatus according to the present invention described above, the optical image reading unit 411 scans a paper sheet in two perpendicular scanning directions as seen in FIG. 2 to optically read paper image information from the paper sheet. Further, the image signal processing means 180 processes the image signal obtained by the optical image reading unit 411 in response to a result of detection of the paper end detection apparatus 170.

In this instance, the paper end detection apparatus 170 detects an end of the paper sheet from the image signal obtained by the optical image reading unit 411. More particularly, the first start mark detection means 174 detects the first start mark SM1 from the image signal from the optical image reading unit 411, and the second start mark detection means 175 detects the second start mark SM2 from the image signal from the optical image reading unit 411. Further, the paper leading end detection signal outputting means 176 outputs a paper leading end detection signal when both of the first start mark detection means 174 and the second start mark detection means 175 detect the start marks.

In this manner, since the paper end detection apparatus for an image reading apparatus wherein paper image information is optically read from a paper sheet using an optical image reading unit 411 of the type which scans the paper sheet in two perpendicular directions comprises the first start mark SM1 and the second start mark SM2 provided on the paper sheet and extending in two perpendicular directions, the first start mark detection means 174 for detecting the first start mark SM1 from an image signal from the optical image reading unit 411, the second start mark detection means 175 for detecting the second start mark SM2 from the image signal from the optical image reading unit 411, and the paper leading end detection signal outputting means 176 for outputting a paper leading end detection signal when both of the first start mark detection means 174 and the second start mark detection means 175 detect the start marks, an end of a paper sheet can be detected accurately from image signals from the two different start marks obtained by the optical image reading unit. Consequently, an accurate data transfer timing can be obtained readily.

b. Embodiment of the Invention

An image reading apparatus according to a preferred embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

1. General Construction of the Image Reading Apparatus

Referring first to FIGS. 4 to 7, there is shown an image reading apparatus according to a preferred embodiment of the present invention. The general structure of the image reading apparatus shown can be divided into an apparatus body 10 and an apparatus lid unit 20. The apparatus lid unit 20 is mounted for pivotal motion around a fulcrum 32 to open or close the apparatus body 10. When the image reading apparatus is used, the apparatus lid unit 20 is fixed to such a closing condition as indicated by solid lines in FIGS. 4 to 6 by a body-lid unit locking mechanism 30. Various other components of the image reading apparatus are mounted on the apparatus body 10 and the apparatus lid unit 20.

Referring to FIGS. 4 and 7, the image reading apparatus includes, as components thereof, a paper supply mechanism 200 which can successively supply paper sheets 40 accommodated therein, a paper transport mechanism 300 for transporting a paper sheet 40 supplied from the paper supply mechanism 200, an optical image reading mechanism 400 for optically reading information on a paper sheet 40 being transported by the paper transport mechanism 300, and a paper stacking mechanism 500 for receiving a paper sheet 40 discharged from the paper transport mechanism 300 to stack such paper sheets 40.

The paper supply mechanism 200 includes a paper supply hopper 210 which can accommodate therein paper sheets 40 to be read, a paper supply roller 220 located above the paper supply hopper 210 for supplying one of paper sheets 40 accommodated in the paper supply hopper 210 toward the paper transport mechanism 300, a paper supply roller driving mechanism 230 for driving the paper supply roller 220 to rotate, a paper supply hopper driving mechanism 240 for driving the paper supply hopper 210 to an inclined position in response to the amount of paper sheets 40 accommodated in the paper supply hopper 210, and a paper separation mechanism 800 for preventing two or more paper sheets supplied by the paper supply roller 220 from being fed to the paper transport mechanism 300.

The paper supply hopper 210 includes a hopper table 212 supported for pivotal motion on a rotatable shaft 212A located at a rear end portion (right end portion in FIGS. 4 and 7) of the image reading apparatus. The hopper table 212 is driven at an end portion (left end portion in FIGS. 4 and 7) thereof by a rack-and-pinion mechanism 248 (including a pinion 248A and a rack 248B) of the paper supply hopper driving mechanism 240 so that it is pivoted upwardly and downwardly and adjusted to a predetermined inclined position. The hopper table 212 includes, as shown in FIG. 5, a pair of tiltable paper edge guide members 214 for guiding the opposite side edges of the paper sheets 40 accommodated in the paper supply hopper 210.

The paper supply hopper driving mechanism 240 includes, as a driving source, a hopper motor 242 constituted from a stepper motor. The driving force of the hopper motor 242 is transmitted to the rack-and-pinion mechanism 248 by way of a belt-and-pulley mechanism 244.

Meanwhile, the paper supply roller 220 is supported for rocking motion around an axis of a separation roller 820 by way of a rockable arm 292 such that it can be retracted upwardly from a space above the paper supply hopper 210 by a paper supply roller retraction mechanism 270. Such upward retraction of the paper supply roller 220 can be performed artificially. However, in a normal condition wherein no artificial force is applied to the paper supply roller 220, the paper supply roller 220 remains at a position suitably moved down by the weight of the paper supply roller 220 itself or by means of a spring not shown, at which the paper supply roller 220 is received by the hopper table 212 below. Then, when the paper supply hopper 210 is pivoted, the paper supply roller 220 is moved upwardly by a required amount in response to the position of the upper face of the paper sheets 40 accommodated in the hopper table 212 which is moved upwardly or downwardly by pivotal motion of the paper supply hopper 210.

Referring now to FIGS. 4, 7, 8 and 9, the paper supply roller driving mechanism 230 for driving the paper supply roller 220 to rotate includes, as a driving source, a transport motor 342 constituted from a stepper motor. The paper supply roller driving mechanism 230 further includes a first belt-and-pulley mechanism 344 and first to third gear mechanisms 852, 856 and 232 interposed between the transport motor 342 and the paper supply roller 220. A pick clutch 238 constituted from an electromagnetic clutch is provided at an inputting portion of the driving force to the paper supply roller 220 from the third gear mechanism 232.

The paper supply roller driving mechanism 230 is controlled by paper supply roller driving mechanism control means 250 in response to the paper supplying position (hopper paper supplying position) of the paper supply hopper 210. More particularly, the paper supply roller driving mechanism control means 250 controls the pick clutch 238 between on and off states to control operation of the paper supply roller driving mechanism 230, that is, the rotation condition of the paper supply roller 220.

The paper separation mechanism 800 includes a separation roller 820, a rotation member 830 disposed in an opposing relationship to the separation roller 820 with a small gap left therebetween, and a separation roller driving mechanism 850 for driving the separation roller 820 to rotate.

The rotation member 830 is located below the separation roller 820, that is, nearer to the apparatus body 10 than the separation roller 820, with a small gap left therebetween. The rotation member 830 includes a pair of pulleys 834 and 836 disposed in a spaced relationship from each other in the paper transporting direction and an endless belt 838 wound between and around the pulleys 834 and 836.

The separation roller driving mechanism 850 is constituted from components substantially common to those of the paper supply roller driving mechanism 230 described hereinabove. In particular, as shown in FIGS. 4, 7, 8 and 9, the separation roller driving mechanism 850 includes the transport motor 342 described hereinabove as a driving source and further includes the first belt-and-pulley mechanism 344 and the first and second gear mechanisms 852 and 856 interposed between the transport motor 342 and the paper supply roller 220. A separation clutch 854 constituted from an electromagnetic clutch is interposed in the first gear mechanism 852. In short, the paper supply roller driving mechanism 230 has a construction wherein the third gear mechanism 232 is provided in addition to the separation roller driving mechanism 850. It is to be noted that operation of the separation clutch 854 is controlled by separation clutch control means 858.

Meanwhile, the paper transport mechanism 300 includes a paper transport path 310 for transporting a paper sheet 40 supplied thereto from the paper supply mechanism 200, a plurality of paper transporting rollers 320 to 328 disposed along the paper transport path 310, a roller driving mechanism 340 for driving the paper transporting rollers 320 to 328, and roller driving mechanism control means 350 for controlling the roller driving mechanism 340. Idler rollers 330 to 338 are provided corresponding to the paper transporting rollers 320 to 328, respectively.

The paper transport path 310 includes an inclined transport path 312 for transporting a paper sheet supplied thereto from the paper supply mechanism 200 in an inclined condition, and a paper reversing transport path 314 provided contiguously to the inclined transport path 312 for reversing the paper sheet 40 transported by the inclined transport path 312.

Due to the construction of the paper transport path 310, the posture of one of the paper sheets 40 supplied from the paper supply hopper 210 is changed first from a substantially horizontal posture in the paper supply hopper 210 to a rearwardly inclined posture in the inclined transport path 312 and is then reversed by the paper reversing transport path 314, and then, in this posture, the paper sheet 40 is discharged to the paper stacking mechanism 500.

Consequently, a paper sheet which is directed upwardly in the paper supply hopper 210 is directed downwardly in the paper stacking mechanism 500, and the paper sheets 40 accommodated one on another in the paper supply hopper 210 are successively stacked into the paper stacking mechanism 500 without changing the order of them.

Meanwhile, the paper transporting rollers 320 to 328 and the idler rollers 330 to 338 are disposed in a condition distributed discretely at a distance smaller than the length of the paper sheets 40 in the transporting direction as seen from FIGS. 4, 7, 8 and 9.

The roller driving mechanism 340 includes the transport motor 342 described above as a driving source and further includes a second belt-and-pulley mechanism 348 in addition to the first belt-and-pulley mechanism 344. The first and second belt-and-pulley mechanisms 344 and 348 will be described here. The first belt-and-pulley mechanism 344 includes a pulley 344A mounted on a rotary shaft of the transport motor 342, another pulley 344B mounted on a rotary shaft 320A of the paper transporting roller 320, and a belt 346A wound between and around the pulleys 344A and 344B. The second belt-and-pulley mechanism 348 includes pulleys 320B to 328B mounted on the rotary shafts 320A to 328A of the paper transporting rollers 320 to 328, respectively, and a belt 346B wound between and around the pulleys 320B to 328B.

Accordingly, when the transport motor 342 operates, the driving force is transmitted from the rotary shaft of the transport motor 342 to the pulley 344B by way of the pulley 344A and the belt 346A so that the rotary shaft 320A of the paper transporting roller 320 is driven to rotate. Further, from the pulley 320B, the rotary shafts 322A to 328A of the other paper transporting rollers 322 to 328 are driven to rotate by way of the belt 346A and the pulleys 322B to 328B so that the paper transporting rollers 320 to 328 are driven to rotate simultaneously.

It is to be noted that reference numeral 360 denotes a tension pulley.

The paper transport mechanism 300 described above is schematically shown in FIGS. 10 and 11. Referring to FIGS. 10 and 11, the components are shown such that the paper supply hopper 210 is positioned on the left side while a paper stacker 510 is positioned on the right side and a paper sheet 40 is transported from the left to the right side reversely to those in FIGS. 4 to 9 so as to conform to time charts which will be hereinafter described.

Referring now to FIG. 12, the optical image reading mechanism 400 includes an optical image reading unit 410 having a reading point 422 located intermediately of the inclined transport path 312 for optically reading information on a paper sheet 40, and image information extraction control means 440 for controlling extraction of image information read by the optical image reading unit 410.

Referring to FIGS. 4 and 7, the optical image reading unit 410 includes, in the arrangement shown, two units of a first optical image reading unit 412 and a second optical image reading unit 414. The optical image reading units 412 and 414 are located intermediately of the inclined transport path 312, and the first optical image reading unit 412 optically reads information on the front face 42 of a paper sheet 40 while the second optical image reading unit 414 optically reads information on the rear face 44 of the paper sheet 40.

Here, each of the optical image reading units 412 and 414 is constituted as an image reading unit of common specifications. For example, FIG. 12 is a schematic side elevational view showing the construction of the image reading unit of common specifications. Referring to FIG. 12, the optical image reading unit 410 shown includes a fluorescent lamp unit 420 serving as a lighting element for irradiating light upon the reading point 422 on the inclined transport path 312, a CCD (charge coupled device) circuit board 436 including a CCD array 436A (CCD array 436A denotes a plurality of CCDs arranged in an array) for optically reading information on a paper sheet 40, and a video circuit board 438 for processing information from the CCD array 436A. It is to be noted that reference numeral 434 in FIG. 12 denotes a black box.

A light path 418 from the reading point 422 to the CCD array 436A is constituted from a plurality of (in the arrangement shown, three, first to third) mirrors 418A, 418B and 418C for reflecting light. A shading plate 430 and a lens 432 are located intermediately of the light path 418 between the mirror 418C and the CCD array 436A so that an image from the mirror 418C may be introduced into the CCD array 436A by way of the lens 432 after it is corrected, particularly at peripheral portions thereof, by the shading plate 430.

Since the light path 418 is formed by the plurality of mirrors 418A, 418B and 418C for reflection of light, the light path 418 can have a sufficient length while the reading point 422 and the CCD circuit board 436 are located at comparatively near locations to each other. Consequently, even where the lens 432 has a great focal length, the reading point 422 can be disposed at a focus position of the lens 432.

A paper sheet 40 from which information has been read by the optical image reading mechanism 400 in this manner is discharged from the paper transport mechanism 300 to the paper stacking mechanism 500. Here, at the terminal end of the paper transport mechanism 300, a paper discharge roller mechanism 540 is located so that the paper sheet 40 may be discharged to the paper stacking mechanism 500 while being driven by the paper discharge roller mechanism 540.

The paper stacking mechanism 500 includes a stacker table 520 having, at the bottom thereof, the paper stacker 510 on which paper sheets 40 can be stacked. A paper trailing end guide mechanism 550 for guiding the rear end 48 of a paper sheet 40 to be stacked into the paper stacker 510.

Referring back to FIGS. 4, 7 and 11, several sensors 610 to 618, 620A to 620D and 622 are provided. Thus, operations of the driving systems described above, that is, operations of the hopper motor 242 of the paper supply hopper driving mechanism 240, the pick clutch 238 of the paper supply roller driving mechanism 230, the separation clutch 854 and the roller driving mechanism 340 of the separation roller driving mechanism 850, and the transport motor 342 for the separation roller driving mechanism 850 and the paper supply roller driving mechanism 230 and extraction operations of the image information extraction control means 440 of the first optical image reading unit 412 and the second optical image reading unit 414 are controlled in response to detection signals from the sensors 610 to 618 and 620A to 620D.

The sensor (SHE) 610 is a hopper empty sensor for detecting whether or not the paper supply hopper 210 is empty. The sensor (SPK) 612 is a paper supply sensor for detecting whether or not the posture of the paper supply hopper 210 is in an optimum condition (that is, a hopper paper supplying position) for supplying a paper sheet. Here, since the paper supply roller 220 is put into a paper supplying position (optimum condition) in response to the paper supplying position of the paper supply hopper 210, the sensor 612 actually detects whether or not the paper supply hopper 210 and the paper supply roller 220 are in their individual paper supplying positions. The hopper empty sensor 610 and the paper supply sensor 612 may each be constituted from, for example, a photo-interrupter.

The sensor (SF1) 614 and the sensor (SF2) 616 are transport sensors for detecting a paper sheet 40 is transported by the paper transport mechanism 300. The sensor (SF3) 618 is a discharge sensor for detecting whether or not a paper sheet 40 is discharged from the paper transport mechanism 300 to the paper stacking mechanism 500. The transport sensors 614 and 616 and the discharge 618 may each be constituted from, for example, a photo-sensor. Here, the transport sensor 614 is a transmission type photo-sensor which includes a light emitting element and a light receiving element located on the opposite sides of the paper transport mechanism 300, and each of the transport sensor 616 and the discharge sensor 618 is a reflection type sensor wherein a light emitting element and a light receiving element are provided as a unitary member.

The sensor (SBS) 620A, the sensor (SA4) 620B, the sensor (SB4) 620C and the sensor (SA3) 620D are sheet width detection sensors. The sensor 620A is a B5 width detection sensor provided for detection of a paper width of a paper sheet of the "B5 size"; the sensor 620B is an A4/LT width detection sensor provided for detection of a paper width of a paper sheet of the "A4 size" or "LT size"; the sensor 620C is a B4 width detection sensor provided for detection of a paper width of a paper sheet of the "B4 size"; and the sensor 620D is an A4/DL width sensor provided for detection of a paper width of a paper sheet of the "A3 size" or "DL size". The sensors 620A to 620D may each be constituted from, for example, a photo-sensor (in the arrangement shown, a reflection type photo-sensor is employed).

Meanwhile, the sensor 622 is a bottom sensor for discriminating whether or not the hopper table 212 of the paper supply hopper 210 is at its lowermost position (bottom position). The sensor 622 may be, for example, a photo-interrupter.

For starting and stopping operations, setting of an operation condition and so forth of the image reading apparatus described above, an operation panel 920 is provided at the front of the image reading apparatus as shown, for example, in FIG. 5.

2. Image Reading Mechanism

The optical image reading unit 410 includes, in the arrangement shown, two units of the first optical image reading unit 412 and the second optical image reading unit 414 as described hereinabove. The optical image reading units 412 and 414 are located intermediately of the inclined transport path 312, and the first optical image reading unit 412 optically reads information on the front face 42 of a paper sheet 40 while the second optical image reading unit 414 optically reads information on the rear face 44 of the paper sheet 40.

As described above, since the optical image reading units 412 and 414 are constructed as image reading units of common specifications, when there is no necessity of distinguishing them from each other in the description of each optical image reading unit, the optical image reading unit is represented by the optical image reading unit 10. In particular, the light path 418 from the reading point 422 to the CCD circuit board 336 in the optical image reading unit 410 is schematically shown in FIG. 13 wherein the light path 418 is generally represented as a straight line omitting the reflections by the mirrors 418A, 418B and 418C. Referring to FIG. 13, pieces of image information arranged in the widthwise direction of a paper sheet 40 are collected by the lens 432 and come to the CCD circuit board 436. The CCD circuit board 436 is constituted from a plurality of CCDs arranged in a juxtaposed relationship to each other so as to catch the pieces of information arranged in the widthwise direction.

The shading plate 430 located forwardly of the lens 432 corrects the image information since the image information is distorted by an increasing amount toward the opposite left and right ends and of the paper sheet 40.

The CCD array 436A operates under the control of respective CCD drivers to catch image information, and the image information is sent to and processed by a video circuit provided on the video circuit board 438.

By the way, in each of the optical image reading units 410, the fluorescent lamp unit 420 is provided in order to make the reading point 422 light.

It is to be noted that a heater (not shown) is provided along the rear face of the fluorescent lamp of the fluorescent lamp unit 420. When the temperature is low, the heater is rendered operative, and after it is started, the fluorescent lamp is warmed up rapidly so that the reading point 422 can be illuminated with a sufficient amount of light.

3. Read Image Data Processing

3-1. Outline of the Read Image Data Processing System

Referring to FIG. 3, the image reading apparatus includes a first image data processing system D1 for processing image data read by the first optical image reading unit 412 for reading information on the front face of a paper sheet, and a second image data processing system D2 for processing image data read by the second optical image reading unit 414 for reading information on the rear face of the paper sheet.

The first image data processing system D1 includes a CCD array 436AA of the first optical image reading unit 412, an amplification circuit (AMP) 64A, a sample hold circuit 66A, an analog to digital (A/D conversion circuit 60A, and an image processing section 68A. Meanwhile, the second image data processing system D2 includes a CCD array 436AB of the second optical image reading unit 414, an amplification circuit 64B, a sample hold circuit 66B, an analog to digital conversion circuit 60B and an image processing section 68B.

The CCD arrays 436AA and 436AB read image data of a paper sheet by way of the image reading units 412 and 414, respectively, as described hereinabove. The amplification circuits 64A and 64B amplify the image data of the paper sheet obtained from the CCD arrays 436AA and 436AB, respectively, and the sample hold circuits 66A and 66B sample and hold the image data of the paper sheet after amplified by the amplification circuits 64A and 64B, respectively.

The analog to digital conversion circuits 60A and 60B convert analog data obtained from the image reading units 412 and 414 into digital data using white level information and black level information of the image information of the paper sheet as indices for a conversion criterion. The image processing sections 68A and 68B process the digital data from the analog to digital conversion circuits 60A and 60B, respectively, by various processes such as binary digitization, emphasis and smoothing.

The image data processing system further includes an outputting section 90 for selectively sending out paper front face image data and paper rear face image data to a host computer (not shown) in response to an instruction from an output control circuit 100 which is part of the image information extraction control means 440. In particular, in the present embodiment, since the image reading units 412 and 414 are provided in the proximity of each other, it sometimes occurs that the image reading units 412 and 414 read images simultaneously. Therefore, information from the second optical image reading unit 414 for reading information on the rear face of a paper sheet is stored once into a buffer storage apparatus (DRAM) of the rear face reading board 944 (refer to FIG. 32) and, after information from the first optical image reading unit 412 is sent to the host computer, the information from the second optical image reading unit 414 is sent from the buffer storage apparatus to the host computer. Such control means is provided in the outputting section 90, and details of the same will be hereinafter described.

In this manner, image information read by the image reading units 412 and 414 is read out under the control of the output control circuit 100 of the image information extraction control means 440 as seen from FIG. 7. In this instance, in the image information extraction control means 440, transfer control of image information to the host computer and like control are performed in response to results of detection of a paper leading end detection circuit (paper leading end detection means) 450 and a paper trailing end detection circuit (paper trailing end detection means) 451. It is to be noted that the paper leading end detection circuit 450 and the paper trailing end detection circuit 451 will be hereinafter described.

In particular, the paper leading end detection circuit 450 detects the paper leading end 46 from a variation of the output of each of the optical image reading units 410, and the paper trailing end detection circuit 451 detects the paper trailing end 48 from a variation of the output of each of the optical image reading units 410. The paper lead end detection circuit 450 and the paper trailing end detection circuit 451 are both provided in the image information extraction control means. It is to be noted that also the paper leading end detection circuit 450 and the paper trailing end detection circuit 451 will be hereinafter described.

Further, the image information extraction control means 440 controls extraction of image information obtained from the first optical image reading unit 412 and the second optical image reading unit 414 in response to a result of selection by the original selection switch 924L serving as the paper reading selection means and a discrimination mark (not shown) applied to a paper sheet 40.

In particular, it can be selected by the original selection switch 924L whether both face reading should be performed or one face reading should be performed, and the image information extraction control means 440 performs reading control in response to a result of the selection by the original selection switch 924L. However, paper sheets which require both face reading and paper sheets which allow one face reading may possibly be present in a mixed condition. In this instance, when paper sheets should be read in a different manner from other paper sheets in which the paper sheets are mixed, a discrimination mark is applied to each of the paper sheets so that they may be read in a different manner. The discrimination mark is provided for discrimination whether the paper sheet should be read by one face reading or by both face reading, and is applied to a location outside an original reading area such as, for example, a corner of the leading end of the paper sheet 40 so that it may be distinguished from image information in the original reading area which should originally be read.

Therefore, for example, when one face reading originals are mixed in both face reading originals, if a discrimination mark which designates one face reading is applied to each of the one face reading originals the quantity of which is smaller than that of the both face reading originals and it is selectively set by way of the original selection switch 924L that both faces of each paper sheet 40 should usually be read, then image information on both faces of a paper sheet is normally read by both of the first optical image reading unit 412 and the second optical image reading unit 414. However, when a discrimination mark 50 is detected, image information only on the front face or the rear face of the paper sheet 40 is read by the first optical image reading unit 412 or the second optical image reading unit 414.

On the contrary, when both face reading originals are mixed in one face reading originals, if a discrimination mark which designates double face reading is applied to each of the double face reading originals the quantity of which is smaller than that of the one face reading originals and it is selectively set by way of the original selection switch 924L that one face of each paper sheet 40 should usually be read, then image information only on the front face or the rear face of a paper sheet is normally read by the first optical image reading unit 412 or the second optical image reading unit 414. However, when a discrimination mark is detected, image information on the both faces of the paper sheet is read by both of the first optical image reading unit 412 and the second optical image reading unit 414.

The image information extraction control means 440 further includes discrimination mark image erasure means 460 so that the image of such discrimination mark applied to a paper sheet 40 may be erased and only image information to be read originally may be outputted.

By the way, the apparatus body 10 or the apparatus lid 20 assures an upper mounting space (space for the front face reading unit) 26 and a lower mounting space (space for the rear face reading unit) 16 having substantially similar sizes and shapes to each other to allow the optical image reading units 412 and 414 to be mounted in them, respectively (refer to FIG. 4). In the meantime, the optical image reading unit 410 is prepared by a plural number having different specifications having different performances but having substantially common sizes and profiles.

While, in the image reading apparatus of the present embodiment, the first optical image reading unit 412 and the second optical image reading unit 414 are constructed with common specifications, it is easy to construct the first optical image reading unit 412 and the second optical image reading unit 414 so as to have different specifications such that, for example, the optical image reading unit for front face reading of the construction described above has higher performances than the optical image reading unit for rear face reading of the construction described above.

Further, each of the optical image reading units 410 includes detection means (front/rear face detection means) 630 which can detect that it is installed as a unit for front face reading when it is installed in the upper mounting space 26 but detect that it is installed as a unit for rear face reading when it is installed in the lower mounting space 16. The detection means 630 may be constructed such that, for example, a front surface detection projection (not shown) is provided only in the upper mounting space 26 while a rear face detection projection (not shown) is provided only in the lower mounting space 16, and a front face detection switch (not shown) which is automatically contacted, when it is installed in the upper mounting space 26, by the front face detection projection to switch to an on-state and a rear face detection switch (not shown) which is automatically contacted, when it is installed in the lower mounting space 16, by the rear face detection projection to switch to an on-state are provided on each of the optical image reading units 410.

Information detected by the detection means 630 in this manner is sent to the image information extraction control means 440 and used for extraction control of image information.

The image data processing system shown in FIG. 3 further includes a pair of timing circuits 55A and 55B which define, for example, sample holding timings of the sample hold circuits 66A and 66B, respectively.

3-2. White Level Information Used upon Analog to Digital Conversion of Image Data and Associated Factors

As shown in FIG. 3, each of the analog to digital conversion circuits 60A and 60B includes a white level information correction circuit (white level information correction apparatus) 70A or 70B and a black level setting circuit 61A or 61B.

The white level information correction circuits 70A and 70B individually set white level information to be used as indices for a conversion criterion of the analog to digital conversion circuits 60A and 60B, respectively, and suitably correct the thus set while level information. The black level setting circuits 61A and 61B individually set black level information to be used as indices for a conversion criterion of the analog to digital conversion circuits 60A and 60B, respectively. It is to be noted that the black level setting circuits 61A and 61B are each constructed as a sample hold circuit.

The white level information correction circuits 70A and 70B will be described in more detail below. Here, since the white level information correction circuits 70A and 70B have a same construction, reference characters to components of the white level information correction circuits 70A and 70B are not distinguished between A and B.

In particular, referring to FIG. 14, the white level information correction circuit 70 includes a plurality of (for example, two) memory circuits 72-1 and 72-2 and registers 73-1 and 73-2, selection circuits 75a, 75b, 75c and 75d, a data magnification variation circuit 74, a white level algorithm circuit 77, and so forth.

The memory circuits 72-1 and 72-2 store a plurality of pieces of white level information (individual pieces of white level information correspond to originals having different ground colors) to be used as indices for a conversion criterion. Write/read control of each of the memory circuits 72-1 and 72-2 is performed in response to an instruction which is received from an MPU (microprocessor unit) circuit 150 by an address controller 72b by way of an input/output port (I/O port) 72a. It is to be noted that a RAM (random access memory) may be employed for the memory circuits 72-1 and 72-2.

The registers 73-1 and 73-2 serve as buffer circuits for temporarily storing the outputs of the memory circuits 72-1 and 72-2, respectively. Each of the selection circuits 75a, 75b, 75c and 75d selectively outputs desired data to a required output line. For example, a multiplexer is used for the selection circuits 75a, 75b, 75c and 75d.

In particular, the selection circuit 75a selectively outputs data from the register 73-1 or 73-2 to the data magnification variation circuit 74 side. The selection circuit 75b selectively outputs white level information varied by magnification variation by the data magnification variation circuit 74 or white level information from the memory circuit 72-1 or 72-2. The selection circuit 75c stores the output of the white level algorithm circuit 77 into and updates a required one of the memory circuits 72-1 and 72-2. The selection circuit 75d supplies the output of one of the memory circuits 72-1 and 72-2 to the analog to digital conversion circuit 60 side by way of a digital to analog conversion circuit 62 (actually, analog to digital conversion circuits 62A and 62B are provided in the data processing systems D1 and D2, respectively).

The data magnification variation circuit 74 varies data by magnification variation by multiplying white level information from the memory circuit 72-1 or 72-2 by a desired coefficient (m: in order to lower the white level, a value between 1 and 0 is selected for m, but in order to raise the white level, a value higher than 1 is selected for m). For example, a digital multiplier is used for the data magnification variation circuit 74. Further, the magnification variation coefficient m of the data magnification variation circuit 74 can be varied by an instruction from the MPU circuit 150.

The white level algorithm circuit 77 compares white level information and data obtained from the optical image reading unit 410 with each other and corrects the white level information in accordance with a result of the comparison. Referring to FIG. 15, the white level algorithm circuit 77 includes a video signal comparator 77a serving as a digital comparison circuit for comparing digital white level information selected by the selection circuit 75b and digital data obtained from the optical image reading unit 410 with each other, and an addition circuit 77b serving as a white level information correction circuit for correcting the white level information in accordance with a result of the comparison by the video signal comparator 77a.

The white level information correction circuit 70 will be described in more detail.

Referring first to FIG. 14, analog video signals from the CCD arrays 436AA and 436AB are amplified by the amplification circuits 64A and 64B, respectively, and, for example, those analog video signals of the outputs of the amplification circuits 64A and 64B in portions (bits) in which photosensitive portions of the CCD arrays 436AA and 436AB are masked are sampled and held by black level setting circuits (sample hold circuits) 71A and 71B, respectively. The thus held analog video signals are connected as reference signals for a black level to the lower limit sides (VRB) of the analog to digital conversion circuits 60A and 60B. Meanwhile, reference signals for a white level are obtained by converting white level values in lines obtained in the last scanning cycle and stored in the memory circuit 72-1 or 72-2 into analog signals by means of the digital to analog conversion circuits 62A and 62B, and are connected to the upper limit sides (VRT) of the analog to digital conversion circuits 60A and 60B, respectively.

Consequently, the analog to digital conversion circuits 60A and 60B output digital signals on the scale of 256 gradations between the white reference level (VRT) and the black reference level (VRB). In this instance, for the white reference level, an analog value of a white level produced corresponding to a white level obtained in the last scanning cycle for an image is used, and for the black reference level, an analog value of a dot at which the photosensitive portion of the CCD array 436AA or 436AB is masked is used.

By the way, white reference level data extracted from the memory circuit 72-1 or 72-2 is fetched into the corresponding register 73-1 or 73-2, and one of the outputs of the registers 73-1 and 73-2 is selected by the selection circuit 75a and then multiplied by m by the data magnification variation circuit 74.

Further, the output of the data magnification variation circuit 74 or data extracted from the memory circuit 72-1 or 72-2 is selected by the selection circuit 75b and inputted to the terminal b of the white level algorithm circuit 77.

Meanwhile, a digital value of a video signal which is the output of the analog to digital conversion circuit 60 is inputted to the other terminal a of the white level algorithm circuit 77. Consequently, the thus inputted digital value is inputted to comparators (COMP) 77a-0 to 77a-2 of the video signal comparator 77a shown in FIG. 15 and then outputted from the video signal comparator 77a as one of, for example, three different outputs including X"FF" (white represented by X"FF" by the 256 gradation representation), X"F7" to X"FE" (a little dark white represented by X"F7" to X"FE" by the 256 gradation representation) and X"F6" or less (white represented by X"F6" or less by the 256 gradation representation).

In the video signal comparator 77a, when the comparators (COMP) 77a-0 to 77a-2 detect that the digital output of the analog to digital conversion circuit 60 is X"FF" mentioned above (that is, when a coincidence output is obtained), it is recognized that the analog video signal obtained by scanning of the current scanning line of the image is equal to or much higher than a white level obtained by scanning in the last scanning cycle, and the white level value of the preceding cycle is incremented, for example, by one.

However, when it is detected that the digital output of the analog to digital conversion circuit 60 falls within the range from X"FE" to X"F7", it is recognized that the analog video signal is a little lower than the white level obtained in the last scanning cycle, and the white level value of the last cycle is incremented by, for example, "-1", that is, decremented by one. Particularly, since no carry need be taken into consideration, X"FF", which is a complementary number on 2, should be added.

When it is detected that the digital output of the analog to digital conversion circuit 60 is equal to or lower than X"F6", it is recognized that not the white level varies but the image now is on the gray level and is considered that it is not related to correction of the white level, and such control as to perform nothing (particular, to add X"00") is performed to calculate a new white level and determine the new white level as a correction value for scanning of the present scanning line.

When one of the comparators (COMP) 77a-0, 77a-1 and 77a-2 outputs "1", only a corresponding one of gate circuits (DV) 77a-3 to 77a-5 in the video signal comparator 77a outputs the value to be added (X"01", X"FF" or X"00" in FIG. 15) while the other gate circuits (DV) exhibit a high impedance state. For example, when the output is extracted from the gate circuit 77a-3, the other gate circuits 77a-4 and 77a-5 exhibit a high impedance state. Thus, the gate circuits 77a-3 to 77a-5 operate as tri-state elements.

Then, the output from one of the gate circuits (DV) 77a-3 to 77a-5 outputted from the video signal comparator 77a is added to a digital value (W0 to W7 in the last cycle) of the white level of the last cycle, which is the output of the selection circuit 75b, by the addition circuit 77b. Then, a result of the addition is outputted as a correction value (W0 to W7 in the present cycle) of the white level for the present scanning cycle from the terminal c of the white level algorithm circuit 77 and is stored into the memory circuit 72-1 or 72-2 selected by the selection circuit 75c. It is to be noted that, as shown in FIG. 15, the addition circuit 77b is constituted from an adder 77b-0 and a flip-flop 77b-1 of one stage. The flip-flop (FF) 77b-1 operates as a hazard prevention mechanism when the correction value in the present cycle is added to the white level corrected in the last scanning cycle and a result of the addition is stored into the memory circuit 72-1 or 72-2.

Such new white level values based on analog video signals of picture elements of a line obtained by scanning the image by means of the CCD array 436A in such a manner as described above are stored into an area of one of the memory circuits for corresponding picture elements of the line selected by the selection circuit 75d. Then, each time a next line is read and analog to digital conversion is performed, the white level values are read out as correction values and are each used as the upper limit value (VRT) for the analog to digital conversion circuit 60 and further referred to for correction of the white level of each picture element of the line.

Naturally, processing of scanning a certain line of the image by means of the CCD array 436A and reading out analog video signals of the individual picture elements and processing of reading out values of white levels of a line scanned in the last scanning cycle from the memory circuit 72-1 or 72-1 are synchronized with a shift pulse signal used to scan the image by means of the CCD array 436A, and the address in the scanning line and the address of the memory circuit 72-1 or 72-2 described hereinabove (the address of the memory circuit of the 8 Kword capacity) are synchronized with each other with an offset of one address.

Further, in the white level information correction circuit 70, for example, in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level value to be provided to the analog to digital conversion circuit 60A or 60B is varied in response to an instruction signal from the MPU circuit 150 when a manual instruction or an automatic instruction is provided to the MPU circuit 150. To this end, white level information stored in the memory circuit 72-1 (or 72-2) is taken out and stored into the register 73-1 (or 73-2), and then the output of the register 73-1 is selected by the selection circuit 75a so that it is subsequently multiplied by m by the data magnification variation circuit 74. The magnification variation factor m can be modified freely in response to an instruction from the MPU circuit 150. This increases the degree of freedom in correction of a white level.

Then, the output of the data magnification variation circuit 74 is selected by the selection circuit 75b and then processed by required processing by the white level algorithm circuit 77, and then the white level of the thus varied magnification is stored into the other memory circuit 72-2 (or 72-1) by way of the selection circuit 75c. Then, the white level of the varied magnification is used as a conversion reference of the analog to digital conversion circuit 60.

It is to be noted that signals at several locations of the circuitry shown in FIG. 14 in this instance (locations denoted at (1) to (8) in FIG. 14, and the output enable signal OE and the write enable signal WE for a memory circuit) are illustrated in the time chart of FIG. 18.

Thereafter, so far as such blue print paper sheets are used, white level information from the memory circuit 72-2 (or 72-1) in which data of the white level information multiplied by m is stored is extracted, and now, the output of the memory circuit 72-2 (or 72-1) is inputted by way of the selection circuit 75b to the white level algorithm circuit 77, by which the processing described above is performed subsequently so that the white level value may have an appropriate value to update the white level.

Consequently, even in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level information correction apparatus copes with this sufficiently and can perform analog to digital conversion with a high degree of accuracy. Besides, by constructing the correction circuit for a white level from a digital circuit, analog parts in the entire circuit and patterns of a printed circuit board of an image inputting device can be reduced to the minimum. Further, since correction of a white level is performed by digital processing, no oscillation occurs in a high frequency band whereas it occurs often with an analog comparator. Accordingly, also the advantage that an increase in stability of operation and in efficiency and certainty in designing can be achieved is obtained.

Further, the white level algorithm circuit 77 may include, as shown in FIG. 16, a video signal comparator 77c serving as a control signal generation circuit for comparing digital white level information selected by the selection circuit 75b and digital data obtained from the optical image reading unit 410 with each other and outputting, in accordance with a result of the comparison, a control signal representing that the digital data is a predetermined value, a counting section (counting circuit) 77d for counting the number of times by which a control signal is successively outputted in the direction of a line from the video signal comparator 77c, an addition value selecting multiplexer 77e serving as a white level information correction circuit for correcting white level information in accordance with a counted value of the counting section 77d, and an addition circuit 77f.

In particular, also in this instance, a digital signal from the analog to digital conversion circuit 60 is inputted to comparators (COMP) 77c-0 to 77c-2 in the video signal comparator 77c, by which it is divided, for example, into three different outputs (X"FF", X"F7" to X"FE", and X"F6" or less) by similar operation to that described hereinabove with reference to FIG. 15.

The outputs of the comparators (COMP) 77c-0 to 77c-2 in the video signal comparator 77c in this instance are outputted by way of gate circuits (G) 77c-3 to 77c-5 and inputted to the counting section (Count) 77d as seen from FIG. 16.

In this instance, each of the memory circuits 72-1 and 72-2 stores information of 8 bits (W0 to W7) necessary to store digital values of white levels of ordinary picture elements obtained by scanning in the last scanning cycle. The 8 bits mentioned above are a new white level signal to be used for a next line which has been produced from a digital value outputted from the analog to digital conversion circuit 60 described hereinabove and an upper limit signal used thereupon by the analog to digital conversion circuit 60. Each of the memory circuits 72-1 and 72-2 further stores, for example, the output of four bits (Qa to Qd) of the counting section (Count) 77d corresponding to each bit of the line.

The output signal of the comparator (COMP) (=FF?) 77c-0 mentioned above is connected to a count enable (EN) terminal of the counting section (Count) 77d, and the four bits (Qa to Qd) of the count value for each picture element obtained by scanning in the last scanning cycle and stored in the memory circuit 72-1 or 72-2 are inputted to terminals Da to Dd of the counting section (Count) 77d. While the count value (Da to Dd) is inputted, when the output of the analog to digital conversion circuit 60A or 60B in the present scanning cycle is X"FF" and consequently a coincidence signal is outputted from the comparator (COMP) (=FF?) 77c-0 so that the count enable (EN) terminal of the counting section (Count) 77d is energized, the counting section (Count) 77d counts up the input value (Da to Dd). However, when the count enable (EN) terminal is not energized and the output signal of one of the other comparators (COMP) (=F7 to FE?, -F6?) 77c-1 and 77c-2, the reset terminal (RST0 or RST1) of the counting section (Count) 77d is energized so that the count value (Da to Dd) for each picture element is cleared.

In particular, in the counting section (Count) 77d, the count value (Qa to Qd) in the last scanning line for each dot of the CCD array 436A is read out from a corresponding address of the memory circuit 72-1 or 72-2 in synchronism with a shift pulse for shifting the CCD array 436A and is loaded to the Da to Dd terminals of the counting section (Count) 77c so that X"FF" which represents a white level is counted for each dot to determine by what number of success lines X"FF" appears. When the digitally converted value of the dot falls within the range of X"F7" to X"FE" or X"F6" or less, the count value for the dot loaded to the Da to Dd terminals of the counting section (Count) 77d is cleared.

In the addition value selecting multiplexer (correction value conversion circuit) 77e shown in FIG. 16, one of gate circuits (DV) 77e-0 to 77e-4 is selected in response to a signal obtained from a decoder (DEC) 77g by decoding an output value of the counting section (Count) 77d and signals .alpha., .beta. and .gamma.(7) outputted from the comparators (COMP) 77c-0, 77c-1 and 77c-2 of the video signal comparator 77c by way of the gate circuits (G) 77c-3, 77c-4 and 77c-5, respectively, to select an addition value (X"01", "X02" or X"04") to be added to a white level of each picture element of a line obtained by the line scanning in the last scanning cycle and stored in the memory circuit 72-1 or 72-2.

The decoded signal from the decoder 77g is "01" when the count value of the counting section 77d is "1", that is, when the count value signifies that the white level corrected in the last scanning cycle is not X"FF" but the white level in the present scanning cycle is X"FF"; the decoded signal is "02" when the count value is "02", that is, when the count value signifies that the white level corrected in the last scanning cycle is X"FF" and also the white level in the present scanning cycle is X"FF"; and the decoded signal is "03" when the count value is "03", that is, when the count value signifies that the white level corrected in the second last scanning cycle is X"FF" and also the white level corrected in the last scanning cycle is X"FF" and besides also the white level in the present scanning cycle is X"FF".

Accordingly, when the count value (Qa to Qd) of the counting section (Count) 77d is, for example, "01", it is recognized that the corrected white level value in the last scanning cycle was not X"FF", and the gate circuit (DV) 77e-2 of the addition value selecting multiplexer 77e is selected. Consequently, the input .alpha. to the gate circuit (G) 77e-2 is energized to determine "+1" as the correction value for the white level value of the last scanning cycle so that the correction value "+1" may be added by the addition circuit 77f.

Similarly, when the count value (Qa to Qd) of the counting section (Count) 77d is, for example, "02", it is recognized that the corrected white level value in the last scanning cycle was X"FF" and then the white level of the dot in the present scanning cycle is X"FF", and the gate circuit (DV) 77e-3 is selected. Consequently, the input .alpha. to the gate circuit (G) 77e-3 is energized to determine "+2" as the correction value for the white level value of the last scanning cycle.

Further, similarly, when the count value (Qa to Qd) of the counting section (Count) 77d is, for example, equal to or higher than "03", it is recognized that the successive corrected white level values in the last scanning cycle and the second last scanning cycle were X"FF" and then also the white level of the dot in the present scanning cycle is X"FF", and the gate circuit (DV) 77e-4 is selected. Consequently, the input .alpha. to the gate circuit (G) 77e-4 is energized to determine "+4" as the correction value for the white level value of the last scanning cycle.

In any other instance, depending upon whether an output is outputted from the comparator (=F7 to FE?) 77c-1 or the comparator (-F6?) 77c-2 of the video signal comparator 77c, the corresponding gate circuit (DV) 77e-0 or 77e-1 is selected so that the input .beta. or .gamma.(7) of the gate circuit (G) 77e-1 or 77e-0 is energized. Consequently, same correction as that described hereinabove with reference to FIG. 15 is performed.

In particular, the control method in the present example described above is characterized in that, when X"FF" as a white level successively appears in successive lines at a certain dot, it is recognized that a sudden variation in white has occurred and thus such a correction is performed that the white level value is raised progressively in accordance with such sudden variation in white.

FIG. 17 shows a modification to the white level algorithm circuit 77 shown in FIG. 16. In particular, referring to FIG. 17, the modified white level algorithm circuit 77 includes, for example, a read-only memory (ROW) 77h in place of the addition value selecting multiplexer 77e and the addition circuit 77f of the white level algorithm circuit 77 shown in FIG. 16. In particular, the output value (4 bits) of the counting section (Count) 77d, the output (3 bits) of the video signal comparator 77c and a white level value (8 bits) in the memory circuit 72-1 or 72-2 obtained by scanning in the last scanning cycle are applied as an address signal to the ROW 77h so that a white level value calculated and stored in advance in the ROW 77h is outputted from the ROW 77h.

The white level value outputted in this manner is stored into the position corresponding to the dot together with the value (Qa to Qd) of the counting section (Count) 77d and then used for calculation for white level correction upon analog to digital conversion in a next line.

Then, also in this instance (in the case of FIG. 16 or 17), in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level value to be provided to the analog to digital conversion circuit 60 is varied in response to an instruction signal from the MPU circuit 150 by the white level information correction circuit 70 in a similar manner as described above. In particular, white level information stored in the memory circuit 72-1 (or 72-2) is taken out and stored into the register 73-1 (or 73-2), and then the output of the register 73-1 (or 73-2) is selected by the selection circuit 75a so that it is subsequently multiplied by m by the data magnification variation circuit 74. Then, the output of the data magnification variation circuit 74 is selected by the selection circuit 75b and then processed by required processing by the white level algorithm circuit 77, and then the white level of the thus varied magnification is stored into the other memory circuit 72-2 (or 72-1) by way of the selection circuit 75c. Then, the white level of the varied magnification is used as a conversion reference of the analog to digital conversion circuit 60.

Thereafter, so far as such blue print paper sheets are used, the white level from the memory circuit 72-2 (or 72-1) is extracted, and now, the output of the memory circuit 72-2 (or 72-1) is inputted by way of the selection circuit 75b to the white level algorithm circuit 77. Then, the processing described above is performed subsequently by the white level algorithm circuit 77 shown in FIG. 16 or 17 so that the white level value may have an appropriate value to update the white level.

Consequently, even in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level information correction apparatus copes with this sufficiently and can perform analog to digital conversion with a high degree of accuracy. Besides, by constructing the correction circuit for a white level from a digital circuit, analog parts in the entire circuit and patterns of a printed circuit board of an image inputting device can be reduced to the minimum. Further, since correction of a white level is performed by digital processing, no oscillation occurs in a high frequency band whereas it often occurs with an analog comparator. Accordingly, also the advantage that an increase in stability of operation and in efficiency and certainty in designing can be achieved is obtained.

Further, in the case described above, since the digital circuit portion of the white level algorithm circuit 77 can be constructed only from ordinary logical OR and AND gate circuits, it can be included readily into a large scale integrated circuit (LSI).

Further, in the arrangement shown in FIG. 17, since the addition value selecting multiplexer 77e and the addition circuit 77f described hereinabove with reference to FIG. 15 are replaced, for example, with the ROW 77h, the number of parts can be reduced, and higher density mounting can be anticipated.

By the way, in the arrangement shown in FIG. 14, the plurality of (two) memory circuits 72-1 and 72-2 are used in order to store white level information, and the memory circuits 72-1 and 72-2 are selectively used using a chip selection function of the address controller 72b. However, such an alternative arrangement as shown in FIG. 19 may be employed wherein a single memory circuit 72 is employed and an address of the memory circuit 72 is designated by the address controller 72b so as to selectively use a pair of different storage areas 72-11 and 72-12 of the memory circuit 72.

Also in this instance, for example, in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level value to be provided to the analog to digital conversion circuit 60 is varied by the white level information correction circuit 70 shown in FIG. 19 in a similar manner as described above. In particular, white level information stored in a certain storage area 72-11 (or 72-12) of the memory circuit 72 is taken out and stored into the register 73, and then the output of the register 73 is subsequently multiplied by m by the data magnification variation circuit 74. Then, the output of the data magnification variation circuit 74 is selected by the selection circuit 75b and then processed by required processing by the white level algorithm circuit 77 (refer to FIGS. 15 to 17), and then the white level of the thus varied magnification is stored into the other storage area 72-12 (or 72-11). Then, the white level of the thus varied magnification is used as a conversion reference of the analog to digital conversion circuit 60.

Thereafter, so far as such blue print paper sheets are used, white level information from the storage area 72-12 (or 72-11) is extracted, and now, the output of the storage area 72-12 (or 72-11) is inputted by way of the selection circuit 75b to the white level algorithm circuit 77. Then, the processing described above is performed subsequently by the white level algorithm circuit 77 shown in FIGS. 15 to 17 so that the white level value may have an appropriate value to update the white level.

In this manner, also in this instance, the effects or advantages achieved by the embodiment described above can be achieved. Further, since the single memory circuit 72 is employed and the storage area 72-11 and 72-12 of the memory circuit 72 are selectively used by designating the address by means of the address controller 72b, there is no necessity any more of preparing a plurality of independent memory circuits. Consequently, handling of the white level information correction circuit 70 is facilitated.

Such another alternative arrangement as shown in FIG. 20 may be employed wherein a switching control circuit 78 is additionally provided for automatically controlling selective switching of the selection circuit 75b in accordance with a result of determination which is conducted by the data magnification variation circuit 74 based on data obtained from the optical image reading unit 410 to determine whether white level information should be varied by magnification variation.

In particular, the switching control circuit 78 includes a pair of comparators 78a and 78c and a counter 78b. In the switching control circuit 78, the output of the analog to digital conversion circuit 60 is first compared with a reference value from reference value setting means 78d by the comparator 78a, and if the output of the analog to digital conversion circuit 60 is higher than the reference value, the counter 78b counts up by one. Further, the output of the counter 78b is compared with a dot reference value from dot reference value setting means 78e by the comparator 78c. If the output of the analog to digital conversion circuit 60 is higher by more than a predetermined line number than the reference value, then since the output of the counter 78b is higher than the dot reference value, a signal to instruct the selection circuit 75b to select the data magnification variation circuit 74 is developed from the comparator 78c. In response to the signal, the selection circuit 75b selects the output of the data magnification variation circuit 74, and consequently, the white level is varied more suddenly than that by variation by the white level algorithm circuit 77. Then, the variation is performed automatically based on data obtained from the optical image reading unit 410 as described above. It is to be noted that, when the output of the analog to digital conversion circuit 60 is still higher by more than the predetermined line number than the reference value even after the selection circuit 75b is switched to the data magnification variation circuit 74 side, the white level varied once is further varied by the data magnification variation circuit 74. In particular, if the white level is varied, for example, twice, then the white level is varied to m.sup.2 times.

It is to be noted that, in FIG. 20, a selection circuit 75e is further provided. The selection circuit 75e selects the output of the white level algorithm circuit 77 or predetermined value data (X"FF" which corresponds to the maximum value). In particular, a control signal VABS is supplied from the MPU circuit 150 to the selection circuit 75e so that, in an initial state, the predetermined value data is outputted from the selection circuit 75e, and thereafter, the output of the white level algorithm circuit 77 is outputted from the selection circuit 75e.

In this manner, also with the white level information correction circuit 70 shown in FIG. 20, the effects or advantages which can be achieved by the embodiment described above can be achieved. Further, since it is determined by the data magnification variation circuit 74 based on data obtained from the optical image reading unit 410 whether or not white level information should be varied by magnification variation and then selective switching of the selection circuit 75b is automatically controlled in accordance with a result of the determination, even when paper sheets used are changed, the white level can be automatically varied rapidly.

It is to be noted that, while, in the several arrangements described above, the white level is varied for each picture element, it need not necessarily be corrected for each picture element, but may naturally be corrected, for example, for each line.

Referring back to FIG. 3, an image signal after digital conversion by the analog to digital conversion circuit 60 is transferred to the image processing section 68A or 68B for next image processing such as, for example, emphasis processing to emphasize the contrast between white and black or "dither processing: binary digitization processing" for a net point image (which is constituted from a large number of fine dots; dot image) such as a photograph image.

3-3. Outputting Section and Output Control Circuit

In each of the image data processing systems D1 and D2, information digitized by the analog to digital conversion circuit 60A or 60B is sent, after it is processed by emphasis processing and/or binary digitization processing by the image processing section 68A or 68B, to the outputting section 90 as seen from FIG. 3, and paper front face data and paper rear face data are transferred from the outputting section 90 to the host computer (not shown).

Referring now to FIG. 21, the outputting section 90 includes a latch circuit 91, a DRAM (buffer storage apparatus) 92, a rear face memory control section 93, a read data buffer 94, a rear face timing generation section 95 and a selection circuit 96.

The latch circuit 91 latches paper front face data VDA and front face timing signals VGA, HGA and VCLKA from the image data processing system D1 for processing paper front face data and notifies to the output control circuit 100 that the paper front face data and the surface timing signals have been latched by the latch circuit 91. It is to be noted that a flip-flop may be used for the latch circuit 91.

The timing signal VGA is a gate signal in a horizontal direction (direction of a line; main scanning direction), and the timing signal HGA is a gate signal in a vertical direction (paper transporting direction; sub-scanning direction). One bit of a picture element in one line on the front face of a paper sheet can be extracted using the timing signal VGA and the timing signal HGA. Further, the timing signal VCLKA is a clock signal which defines the transfer rate of front face data.

The DRAM 92 is a memory circuit for storing paper rear face data VDB. Storage and read-out control of the DRAM 92 is performed by the rear face memory control section 93. In particular, the rear face memory control section 93 is constructed as a DMAC (dynamic memory access controller) and stores paper rear face data VDB sent thereto into the DRAM 92. After paper front face data VDA for one sheet are sent, the rear face memory control section 93 controls the DRAM 92 so that, between successive storing operations of paper rear face data VDB into the DRAM 92, the paper rear face data VDB are read out from the DRAM 92 (in this instance, the paper rear face data VDB are read out in units of one bit).

It is to be noted that all data for one full paper sheet from the later operating image reading unit 414 need not necessarily be stored into the DRAM 92, and actually, those data obtained after outputting of data for one full paper sheet from the first operating image reading unit 412 is completed until all information stored in the DRAM 92 is sent out are stored. In particular, data from the later operating image reading unit are stored for a period of time after outputting of data for one full paper sheet from the other first operating image reading unit is completed until it becomes possible to pass and output data from the later operating image reading unit through and from the rear face memory control section 93. This can decrease the memory capacity.

It is a matter of course that data from the later operating image reading unit 414 may all be stored for one full paper sheet into the DRAM 92. This facilitates memory control.

The read data buffer 94 temporarily stores, between successive storing operations of paper rear face data VDB into the DRAM 92, paper rear face data VDB partially read out from the DRAM 92 and produces arranged data to be sent. Where the read data buffer 94 is provided in this manner, data can be read out little from the DRAM 92 while giving priority to writing into the DRAM, and consequently, the data transfer time can be reduced.

It is to be noted that the rate at which data are read out from the DRAM 92 and the read data buffer 94 is set equal to twice the writing rate, that is, the transfer rate of front face data to the host computer.

The rear face timing generation section 95 generates rear face timing signals VGB, HGB and VCLKB. Also in this instance, the timing signal VGB is a gate signal in a horizontal direction (direction of a line; main scanning direction), and the timing signal HGB is a gate signal in a vertical direction (paper transporting direction; sub-scanning direction). One bit of a picture element in one line on the rear face of a paper sheet can be extracted using the timing signal VGB and the timing signal HGB. Further, the timing signal VCLKB is a clock signal which defines the transfer rate of front face data. In this instance, the rate of the rear face timing signal VCLKB is set to twice that of the front face timing signal VCLKA. Due to the rate and also due to the data read-out data rate from the DRAM 92 and the read data buffer 94, rear face data are transferred at the rate equal to twice that of front face data.

The selection circuit 96 receives a control signal from the output control circuit 100 and selectively outputs paper front face data or paper rear face data in accordance with the control signal. In this instance, after paper front face data are transferred for one full paper sheet, the data to be outputted are switched so that paper rear face data are thereafter transferred for one complete paper sheet.

It is to be noted that the output control circuit 100 is switched to the paper front face data side in response to paper leading end detection information but is switched to the paper rear face data side in response to paper trailing end detection information. Such detection of a leading end or a trailing end of a paper sheet will be hereinafter after described.

Therefore, it is considered that the outputting section 90 includes the storage means (DRAM) 92 for storing data (paper rear face data) from that one of the first optical image reading unit 412 and the second optical image reading unit 414 which serves as a later operating image reading unit (in the present example, the second optical image reading unit 414) from which paper image information is read out later than from the other image reading unit.

Also it is considered that the outputting section 90 further includes first data transfer means (the latch circuit 91 and the selection circuit 96) for successively transferring, by way of a data transfer line, paper front face data from that one of the first optical image reading unit 412 and the second optical image reading unit 414 which serves as a first operating image reading unit (in the present example, the first operating image reading unit 412) from which paper image information is read out first, and second data transfer means (the rear face memory control section 93, the rear face timing generation section 95 and the selection circuit 96) for successively transferring, by way of another data transfer line, after paper front face data from the first operating image reading unit 412 are transferred by the first data transfer means, paper rear face data from the later operating image reading unit 414 stored in the storage means 92 at a rate (in the present example, at the twice rate) higher than the data transfer rate by the first data transfer means.

Further, it is also considered that the outputting section 90 further includes auxiliary storage means (the read data buffer 94) for storing partial paper image information read out from the storage means 92 between successive writing operations of data into the storage means (DRAM) 92.

It is to be noted that the reason why paper rear face data are sent at a rate twice that of paper front face data is that it is desired that paper rear face data be transferred to the host computer side before the paper sheet 40 whose front face and rear face have been read is thereafter discharged to the stacking mechanism 500. This is because, if a paper sheet is discharged to the stacking mechanism 500, then transportation of a new paper sheet is started and reading of data of the paper sheet is started. In other words, paper rear face data are transferred at the rate twice that of paper front face data because it is desired to complete transfer of rear face data before transportation of a next paper sheet is started. Accordingly, naturally the value twice may possibly be varied depending upon the length of the paper transport path, the paper transport velocity or the like of the image reading apparatus.

Accordingly, the outputting section 90 operates in such a manner as illustrated in FIG. 22.

Referring to FIG. 22, since the selection circuit 96 is initially switched to the paper front face data side, paper front face data are transferred by through-transfer (step A1). Thereafter, when the signal VGA becomes negated, that is, when transfer of the front face data is completed, the route of "YES" is taken at step A2, and then the selection circuit 96 is switched to the paper rear face data side (step A3). Then, at step A4, paper rear face data are read out from the DRAM 92 and transferred at a high rate to the read data buffer 94 (step A5). Then, this operation is repeated until the DRAM (image sensor) 92 becomes emptied (step A6).

Due to the construction described above, data (paper front face data) from the first operating image reading unit 412 from between the first optical image reading unit 412 and the second optical image reading unit 414 from which paper image information is to be read out first are first transferred successively by way of a data transfer line, and data (paper rear face data) from the later operating image reading unit 414 from between the first optical image reading unit 412 and the second optical image reading unit 414 from which paper image information is to be read out later are temporarily stored into the DRAM 92 and, after the data from the first operating image reading unit 412 are transferred completely, the stored data from the later operating image reading unit 414 are successively transferred at a rate higher than the transfer rate of the data from the first operating image reading unit 412 by way of another data transfer line. Consequently, the following advantages can be obtained.

In particular, even if image reading by the later operating image reading unit 414 is started before image reading by the first operating image reading unit 412 is completed, data within the overlapping period can be held with certainty. Besides, data from the later operating image reading unit 414 from which data are transferred later can be transferred rapidly to the host computer side.

Accordingly, even where the image reading units 412 and 414 are disposed in the proximity of the paper transport path in order to achieve minimization, reduction in weight and compaction of the apparatus, data of the front and rear faces of a paper sheet can be transferred to the host computer side before the paper sheet is discharged to the stacking mechanism 500. Consequently, even if paper sheets are successively transferred at a high speed while achieving minimization, reduction in weight and compaction of the apparatus, data of the front and rear faces of each paper can be read and transferred to the host computer side satisfactorily.

By the way, the output control circuit 100 includes, as described hereinabove, the paper leading end detection circuit 450 for detecting paper leading end information and the paper trailing end detection circuit 451 for detecting paper trailing end information.

The paper leading end detection circuit 450 and the paper trailing end detection circuit 451 have a same circuit construction and each includes, as shown in FIG. 23, magnitude comparators 131, 138 and 139, a multiplexer 132, registers 133, 134, 135 and 141, adders 136 and 137, and an OR gate 140.

Due to the construction, a video signal (paper front face data) from the image reading unit 412 is inputted to an input terminal A of the magnitude comparator 131. The magnitude comparator 131 thus compares the video signal from the image reading unit 412 with the output of the register 133 which is inputted to the magnitude comparator 131 by way of another input terminal B. In this instance, if the video signal is higher, then the video signal is outputted from the multiplexer 132 so that it is latched by the register 133. On the contrary if the output of the register 133 is higher, then the output of the register 133 is outputted from the multiplexer 132 so that it is latched by the register 133. Such comparison in magnitude is repeated. Consequently, at the end of the line, a maximum value (peak value) of the line is latched in the register 133.

Therefore, it can be understood that the magnitude comparator 131, the multiplexer 132 and the register 133 construct one-line peak value detection means for detecting a peak value in one line along the direction perpendicular to the paper transporting direction based on an image signal from an optical image reading unit.

Further, a clock BB (this clock BB is developed once for one line at the end of the line) causes the register 134 to latch a maximum value (peak value) in one line latched in the register 133 and simultaneously causes the register 135 to latch a maximum value (peak value) of the last line. Consequently, the registers 133 and 134 construct a shift register for storing both of a current peak value detected by the one-line peak value detection value and a past peak value in one line detected prior to the current peak value.

Meanwhile, the register 135 is constructed as storage means for storing a past peak value in one line detected prior to a current peak value detected by the one-line peak value detection means.

Thereafter, +n (n is a natural number) is added to a maximum value (peak value) of a preceding line by the adder 136, and -n is added to (that is, n is subtracted from) the maximum value by the adder 137. Then, the output of the adder 136 and the maximum value (peak value) of the current line from the register 134 are compared with each other by the magnitude comparator 138, and the output of the adder 137 and the maximum value (peak value) of the current value from the register 134 are compared with each other by the magnitude comparator 139.

Then, if the maximum value of the current line is higher than the sum of the maximum value of the preceding line and n, the magnitude comparator 138 outputs "1", but if the maximum value of the current line is lower than the difference of n from the maximum value of the preceding line, the magnitude comparator 139 outputs "1". Consequently, if the maximum value of the current line is higher than the sum of the maximum value of the preceding line and n or lower than the difference of n from the maximum value of the preceding line, then the OR gate 140 outputs "1".

Then, the output of the OR gate 140 is latched by the register 141 which operates in response to a clock AA (the clock AA is outputted once for one dot) in order to prevent fluctuation, and the output of the register 141 is used as a paper leading end detection signal or a paper trailing end detection signal. Then, such paper leading or trailing end detection signal is used as an interruption requesting signal (IRQ) to the MPU circuit 150.

Consequently, it can be seen that the magnitude comparators 138 and 139 cooperatively construct comparison means for comparing a current peak value detected by the one-line peak value detection means and a past peak value from the storage means with each other and outputting a result of the comparison as a paper leading or trailing end detection signal.

Meanwhile, the adder 136 constructs addition means for adding a predetermined value (n) to a past peak value from the register (storage means) 135. Thus, the magnitude comparator (comparison means) 138 is constructed so as to compare a current peak value detected by the one-line peak value detection means and an addition correction past peak value from the adder 136 with each other and output a result of the comparison as a paper end detection signal.

Further, the adder 137 constructs subtraction means for subtracting a predetermined value (n) from a past peak value from the register (storage means) 135. Thus, the magnitude comparator (comparison means) 139 is constructed so as to compare a current peak value detected by the one-line peak value detection means and a subtraction correction past peak value from the adder 137 with each other and output a result of the subtraction as a paper leading or trailing end detection signal.

Meanwhile, the magnitude comparator 138 constructs first comparison means for comparing a current peak value detected by the one-line peak value detection means and an addition correction past peak value from the adder (addition means) 136 with each other. The magnitude comparator 139 constructs second comparison means for comparing a current peak value detected by the one-line peak value detection means and a subtraction correction past peak value from the adder (subtraction means) 137 with each other. The OR gate 140 constructs outputting means for outputting, when a paper end detection signal is outputted from at least one of the first comparison means and the second comparison means, the paper end detection signal as a paper end detection signal.

Further, the register 141 constructs latching means for latching the outputs of the comparison means.

It is to be noted that FIG. 24 illustrates signal waveforms (time chart) at several locations of the circuit shown in FIG. 23.

The leading end or the trailing end of a paper sheet is detected in such a manner as described above. The paper leading end detection circuit 450 detects the leading end of a paper sheet based on a paper end detected for the first time, and when the MPU circuit 150 receives such paper leading end detection signal as an interruption requesting (IRQ) signal from the paper leading end detection circuit 450, the MPU circuit 150 develops, in response to the interruption requesting (IRQ) signal, a command signal to read the front face of the paper sheet. As a result, also the selection circuit 96 is switched to the paper front face data side.

Meanwhile, the paper trailing end detection circuit 451 detects the trailing end of the paper sheet based on a paper end detected for the second time, and when the MPU circuit 150 receives such paper trailing end detection signal as an interruption requesting (IRQ) signal from the paper trailing end detection circuit 451, the MPU circuit 150 develops, in response to the interruption requesting (IRQ) signal, a command signal to read the rear face of the paper sheet. As a result, also the selection circuit 96 is switched to the paper rear face data side.

Therefore, the outputting section 90 is constructed as image signal processing means for processing image signals obtained by the optical image reading units in response to a result of detection by the paper end detection apparatus.

Accordingly, if the paper leading end detection circuit 450 and the paper trailing end detection circuit 451 of such circuit construction as described above are employed, the the leading end and the trailing end of a paper sheet can be detected with certainty with a common circuit construction and with a simple construction. Consequently, reading timings of paper front face data and paper rear face data or reading switching timings between them can be controlled precisely irrespective of the type of a paper sheet being transported. As a result, it is comparatively simple for the image reading apparatus to allow paper sheets of various sizes to be transported to read information on them.

It is to be noted that, while any of the leading end and the trailing end of a paper sheet can be detected by the construction shown in FIG. 23, where the ground color of a paper sheet 40 is brighter than the color of the backing member provided on the paper transport path 310, if it is intended to only detect the leading end of the paper sheet 40, then the adder 137, the magnitude comparator 139 and the OR gate 140 can be omitted. Similarly, if it is intended to only detect the trailing end of the paper sheet 40, then the adder 136, the magnitude comparator 138 and the OR gate 140 can be omitted. On the contrary, where the ground color of a paper sheet 40 is darker than the color of the backing member provided on the paper transport path 310, if it is intended to only detect the leading end of the paper sheet 40, then the adder 136, the magnitude comparator 138 and the OR gate 140 can be omitted. Similarly, if it is intended to detect only the trailing end of the paper sheet 40, then the adder 137, the magnitude comparator 139 and the OR gate 140 can be omitted.

4. Control System

4-1. Operation Panel

Referring now to FIG. 31, the operation panel 920 has provided thereon various indication lamps including a power source input indication lamp 922A, a reading enable indication lamp 922B and a check lamp 922C, and a liquid crystal display unit 922D for displaying various information by characters. The liquid crystal display unit 922D suitably displays, for example, information of an operation input, an error message, and so forth.

The operation panel 920 further has provided thereon a plurality of automatic reading mode setting switches 924A and 924B each serving as insertion mode selection means for selectively setting one of a plurality of (two including a mode 1 and a mode 2 here) automatic reading modes, a manual insertion setting switch 924C serving as insertion mode selection means for setting a manual insertion mode, a start switch 924D for starting the image reading apparatus, and a stop switch 924E for stopping the image reading apparatus. In order to start the apparatus, one of the mode 1, the mode and the manual insertion mode is set, and then the start switch 924D will be operated.

The operation panel 920 further has provided thereon an original size inputting switch 924F, a reading concentration setting switch 924G, a reading density setting switch 924H, a landscape switch 924J, a half tone (half tone) setting switch 924K, and the original selection switch (paper reading selection means) 924L. The original selection switch 924L is a switch by which it can be set whether both face reading of an original should be performed or one face reading only of the front face or the rear face should be performed.

4-2. Construction of the Control System

FIG. 32 schematically shows the mechanical components described above and control sections for controlling the mechanical components. Referring to FIG. 32, a control section 930 includes a mechanical section control means 932 including a control circuit for controlling mechanical operations of the mechanical components, and image reading system control means 934 including a control circuit for controlling operation of the image reading system. A pair of power source adjustment sections 940A and 940B for transforming an external power source to required voltages are connected to the image reading system control means 934.

The mechanical section control means 932 controls operation of the transport systems (that is, the paper supply mechanism 200, the paper transport mechanism 300, the paper stacking mechanism 500 and so forth) and the heater of the lamp unit and an invertor of the fluorescent lamp in accordance with an instruction signal received by way of the image reading system control means 934 and detection information from the various sensors of the mechanical components. The mechanical section control means 932 also controls operation of a cooling fan 936 for the control section 930 itself. The paper supply hopper position control means (motor control means) 280, the pick clutch control means 250 serving as paper supply roller driving mechanism control means, the separation clutch control means 858 and the roller driving mechanism control means 350 described hereinabove are included in the mechanical section control means 932.

The image reading system control means 934 controls operation of CCD driver units of the first optical image reading unit 412 and the second optical image reading unit 414, a video circuit and a rear face reading board 944 and outputting to an outputting interface board 938 in response to setting information of the operation panel 920 and information from the mechanical section control means 932. The image information extraction control means 440, the paper leading end detection circuit 450, the paper trailing end detection circuit 451 and the discrimination mark image erasure means 460 described hereinabove are provided in the image reading system control means 934.

Where an endorser (endorsing printer) 942 is provided in the proximity of the terminal end of the paper transport path 310 as shown in FIG. 7, also a driver for the endorser 942 is controlled by the image reading system control means 934 as seen from FIG. 32. Also where an extension memory board and/or an auxiliary printed circuit board (IPC-2) are provided, they are controlled by the image reading system control means 934.

4-3. Operation

Operations of the hopper motor 242, the pick clutch 238, the separation clutch 854 and the transport motor 342 and control by the image information extraction control means 440 proceed, for example, in such a manner as illustrated in time charts of FIGS. 25 to 30.

First, control of the hopper motor 242 will be described. Upon starting of the control, control of an initialization mode is performed as seen from FIG. 25. In particular, in response to an operation starting instruction (that is, a control starting instruction) for the image reading apparatus such as, for example, throwing in of a power source to the apparatus, the hopper motor 242 is rotated in a direction to lower the hopper table 212. Then, when the hopper table 212 comes to its lowermost position, the bottom sensor 622 switches from an off-state (open) to an on-state (closed), and the hopper motor 242 stops in response to such detection signal of the bottom sensor 622. Naturally, the control is not performed if, upon reception of the control starting instruction, the hopper table 212 is already at the lowermost position and the bottom sensor 622 is in an on-state (closed).

The control of the hopper motor 242 after this is different between an automatic reading mode and the manual insertion mode in response to setting information of the switches.

In particular, if paper sheets 42 are accommodated into the hopper table 212 and a switch operation (depression of the start button) for an automatic reading mode is performed, then the hopper motor 242 is rotated in a direction to raise the hopper table 212 as seen from FIG. 26. Then, when the top of the paper sheets 40 in the hopper table 212 rises from a position (bottom position) corresponding to the lowermost position of the hopper table 212, whereupon the hopper empty sensor 610 is turned on ("presence of a paper sheet"), to a prescribed height at which the paper supply sensor 612 is turned on ("presence of a paper sheet").

When the hopper table 212 is raised by a little distance after the paper supply sensor 612 is turned on, the hopper motor 242 is stopped. Thereafter, image reading is performed while paper supplying and transporting operations are performed. During the process, as the paper sheets 40 are supplied, the height of the top of the stack of paper sheets 40 decreases. Consequently, the paper supply sensor 612 is turned off finally, and in response to this, the hopper motor 242 is rotated in the direction to raise the hopper table 212.

Then, as the height of the top of the stack of paper sheets 40 in the hopper table 212 rises again, it finally reaches the prescribed height, whereupon the paper supply sensor 612 is turned on ("presence of a paper sheet"). While such a sequence of operations as described above is repeated to control the height of the top of the paper sheets within a fixed range, image reading operation is performed together with paper supplying and transporting operations.

On the other hand, if a switch operation for the manual insertion mode (depression of the manual insertion button) is performed, then the hopper motor 242 is rotated in the direction to raise the hopper table 212 as seen from FIG. 27. Then, when the hopper table 212 is raised until the height of the top end thereof comes to a prescribed height, then the paper supply sensor 612 is turned on ("presence of a paper sheet"). When the hopper table 212 is further raised a little after the paper supply sensor 612 is turned on, the hopper motor 242 is stopped. Thereafter, the hopper motor 242 is kept stopped and the hopper table 212 keeps the position. Then, manual insertion of a paper sheet is performed as can be seen also from an on/off condition of the hopper empty sensor 610.

Subsequently, operations of the pick clutch 238, the separation clutch 854 and the transport motor 342 and control by the image information extraction control means 440 will be described together with operation of the hopper motor 242. Referring to FIG. 28, paper sheets 40 are first accommodated into the paper supply hopper 210 and a start command to instruct starting of image reading is developed (point T1). At this initial stage, since the paper supply hopper 210 is not at the paper supply position, the paper supply sensor 612 is in an off-state. The hopper empty sensor 610 also provides a signal indicating absence of a paper sheet.

Since the paper supply sensor 612 is in an off-state, the hopper motor 242 is rendered operative to raise the paper supply hopper 210 to the paper supply position (point T2). Consequently, the paper supply sensor 612 is turned on. As a result, the hopper motor 242 is stopped, and the pick clutch 238 and the separation clutch 854 are engaged. Thereafter, the transport motor 342 is started (point T3) after a small time lag (30 ms in the example shown) until the pick clutch 238 and the separation clutch 854 are engaged firmly. By the operation of the transport motor 342, the pick rollers 220 and the separation roller 820 are rotated by way of the pick clutch 238 and the separation clutch 854 to supply and transport a first paper sheet.

The transport motor 342 can be selectively set to one of a low speed mode of a velocity V.sup.1 (for example, 12 to 13 cm/s), a high speed mode of another velocity V.sup.2 (for example, about 50 cm/s) and an intermediate speed mode (mid speed mode) of an intermediate velocity between them. For the first paper sheet upon starting of paper supply, the transport motor 342 operates in the low speed mode. Accordingly, also the transportation speeds of the pick rollers 220 and the separation roller 820 are low.

When the leading end of the paper sheet being transported in this manner passes the transport sensor 614, the transport sensor 614 detects this and is turned on (point T4), and the pick clutch 238 is disengaged. At this point of time, the paper sheet is already at a position at which it can be driven by the separation roller 820, and consequently, the paper sheet is thereafter driven by the separation roller 820.

Then, when the leading end of the paper sheet being transported passes the transport sensor 616, the transport sensor 616 detects this and is turned on (point T5), and the separation clutch 854 is disengaged. At this point of time, the paper sheet is already at a position at which it can be driven by the transport roller 320, and consequently, the paper sheet is thereafter driven by the transport roller 320. Thereafter, the paper sheet is successively driven by the succeeding transport rollers 322 to 328. At the point of time T5, since the transport motor 342 is in the low speed mode, the transportation velocity of the transport roller 320 itself is low.

The transport sensor 616 serves also as a sensor for detecting a reading timing, and when the passage of the leading end of the paper sheet is detected by the transport sensor 616, a read command is developed in response to the detection (point T6). Upon reception of the read command, the transport motor 342 is accelerated from the low speed mode (velocity V.sup.1) to the high speed mode (velocity V.sup.2). Accordingly, also the speed of rotation of the transport rollers 320 to 328, that is, the transportation speed, increases until high speed transportation is reached.

Then, at a point of time T7 after lapse of a predetermined time t.sup.3 after the leading end of the paper sheet passes the transport sensor 616, the first optical image reading unit 412 for reading information of the front face of the paper sheet is put into a reading condition. Thereafter, at another point of time T8 after lapse of another predetermined time t4 after the leading end of the paper sheet passes the transport sensor 616, the second optical image reading unit 414 for reading information on the rear face of the paper sheet is put into a reading condition. In particular, each video gate (not shown) of the video circuit board 438 is put into an on-state.

It is to be noted that the predetermined times t.sup.3 and t.sup.4 are times required for a paper sheet to pass from the transport sensor 616 to the reading points 412A and 414A of the optical image reading units 412 and 414, respectively, and are given, from the distances L.sup.1 and L2 from the transport sensor 616 to the reading points 412A and 414A and the transportation speed V.sup.2 by the transport roller 320, by the following equations, respectively;

During such image reading (at points of time T9 and T10), the transport sensors 614 and 616 are switched from on to off when the trailing end of the paper sheet passes the transport sensors 614 and 616, respectively.

Then, in each of the optical image reading units 412 and 414, when a time t.sup.5 required for image reading passes (point T11 or T12), the video gate is switched from on to off, thereby completing reading (Read Complete). It is to be noted that the time t.sup.5 is given as a product between the reading line number and the integration time (t.sup.5 =reading line number.times.integration time).

In this manner, while the first paper sheet is transported in the high speed mode by the transport rollers 320 to 328, image reading of the front face and the rear face of the paper sheet is performed by the optical image reading units 412 and 414, respectively, and thereafter, the paper sheet is driven by the paper transport roller 328 and stacked into the paper stacker 510.

After reading of the first paper sheet is completed, a start command is developed immediately, and in response to the start command, transportation and reading of a second paper sheet are started. In the operation for the second or following paper sheet, the image reading apparatus operates in such a manner as illustrated in FIG. 29.

In particular, in the present example, since the paper supply hopper 210 is at the paper supply position (that is, the paper supply sensor 612 is in an on-state) when the start command is instructed (point T13), the pick clutch 238 and the separation clutch 854 are engaged simultaneously with the instruction of the start command. Since the transport motor 342 continues to operate in the high speed mode, the pick roller 220 and the separation roller 820 are rotated at a comparatively high speed due to the engagement of the clutches 238 and 854 to transport the second paper sheet. Naturally, in this instance, also the transport rollers 320 to 328 are being rotated by the transport motor 342.

Thereafter, transportation and reading of the second paper sheet are performed substantially in a similar manner to the first paper sheet. However, in transportation and reading of the second or following paper sheet, since the transport motor 342 is operating in the high speed mode from the beginning, the transport motor 342 is controlled to temporarily lower the speed thereof at a point of time when the main element for driving the paper sheet changes over from the separation roller 820 to the transport roller 320, different from the transportation and reading of the first paper sheet.

In particular, when the leading end of the second paper sheet which is supplied and transported at a comparatively high speed by the pick roller 220 and the separation roller 820 passes the transport sensor 614, the transport sensor 614 detects this and is turned on (point T15). Consequently, the pick clutch 238 is disengaged and the paper sheet is thereafter driven by the separation roller 820.

Then, when the leading end of the second paper sheet passes the transport sensor 616, the transport sensor 616 detects this and is turned on (point T19), and the separation clutch 854 is disengaged. Around the point of time T19 (between the points of time T17 to T20), the speed of the transport motor 342 is reduced temporarily from the high speed mode to the intermediate speed mode.

Such speed reduction control is started at a point of time T16 when a required time elapses after the transport sensor 614 is turn on (at a point of time before the leading end of the paper sheet passes the transport sensor 616) and is performed by holding, after the point of time T17 at which the speed drops to an intermediate speed, the intermediate speed till a point of time T20 at which a predetermined time (for example, 50 ms) elapses after the point of time T17.

Due to the speed reduction control, when the main element for driving the paper sheet changes over from the separation roller 820 to the transport roller 320, the transportation speed of the separation roller 820 and the transport roller 320 is suppressed, and consequently, changing over from the separation roller 820 to the transport roller 320 proceeds smoothly. This reduces a cause of a trouble such as paper jamming.

Within the period, a read command is developed (point T18), and similarly as in transportation of the first paper sheet, the first optical image reading unit 412 for reading information of the front face of a paper sheet is put into a reading condition at a point of time T21 at which the predetermined time t.sup.3 elapses after the leading end of the paper sheet passes the transport sensor 616. Then, at another point of time T22 when the predetermined time t.sup.4 elapses after the leading end of the paper sheet passes the transport sensor 616, the second optical image reading unit 414 for reading information of the rear face of a paper sheet is put into a reading condition. In particular, each video gate (not shown) of the video circuit board 438 is put into an on-state. It is to be noted that the predetermined times t.sup.3 and t.sup.4 mentioned above are given similarly as described hereinabove.

During such image reading, the transport sensors 614 and 616 are changed over from an on-state to an off-state (points T23 and T24) as the trailing end of the paper sheet passes the transport sensors 614 and 616, respectively.

Then, in each of the optical image reading units 412 and 414, the video gate is changed over from an on-state to an off-stage to complete the reading (Read Complete) when the time t.sup.5 required for image reading elapses. Also the time t.sup.5 is given similarly as described hereinabove.

In this manner, while the second or following paper sheet is transported in the high speed mode by the transport rollers 320 to 328, image reading of the front face and the rear face of the paper sheet is performed by the optical image reading units 412 and 414, respectively, and thereafter, the paper sheet is driven by the paper transport roller 328 and the paper discharge roller and stacked into the paper stacker 510.

If the paper supply sensor 612 is turned off as a result of reduction in quantity of the paper sheets 40 in the paper supply hopper 210 (point T14 in FIG. 29), then the hopper motor 242 is rendered operative at a point of time (T27) at which the operations of the pick roller 220 and the separation roller 820 and the speed reduction control of the transport motor 342 are completed to raise the paper supply hopper 210 to the paper supply position (point T2). Such height control of the paper supply hopper 210 is performed each time the paper supply sensor 612 is turned off as a result of reduction in quantity of the paper sheets 40 while the paper supplying and transporting operations are performed.

Then, when the paper sheets 40 in the paper supply hopper 210 are reduced in quantity until the paper supply hopper 210 becomes empty, the hopper empty sensor 610 changes over from an off-state ("paper present") to an on-state ("paper absent") (point T28) as seen from FIG. 30, and then the transport sensor 616 changes over from an on-state ("during paper passage") to an off-state ("completion of paper passage") (point T29). Thereafter, the video gate of the second optical image reading unit 414 on the downstream side of the transport path is changed over from an on-state to an off-state and simultaneously the read command is changed over from an on-state to an off-state (point T30), and then the discharge sensor 618 changes over from an on-state ("during paper passage") to an off-state ("completion of paper passage") (point T31). The power supply to the transport motor 342 is cut to stop the transport motor 342 after lapse of a predetermined time t.sup.8 after the discharge sensor 618 changes over to an off-state. The predetermined time t8 corresponds to a time within which a paper sheet 40 is transported from the discharge sensor 618 to the stacker 500.

It is to be noted that, if a paper sheet to be read requires image reading of only one face thereof and it is intended to read, for example, only the front face of the paper sheet, when reading of the video gate of the first optical image reading unit 412 in FIGS. 28 and 29 comes to an end, it is determined that reading for the paper sheet is completed (Read Complete), and next control is started immediately.

Since transportation and image reading of paper sheets is performed in response to the hopper empty sensor 610, the paper supply sensor 612, the transport sensors 614 and 616 and the discharge sensor 618 in this manner, the image reading operation can be performed appropriately in accordance with a transportation condition of a paper sheet, which is suitable to high speed image reading. Further, if paper jamming (paper jamming) should occur intermediately of the paper transport path, this can be detected promptly and the operation of the image reading apparatus can be stopped immediately.

Further, since the control timings by the roller driving mechanism control means 350 and the image information extraction control means 440 are synchronized with each other, even if the processing speed for image reading is increased, the paper transportation operation and the image reading operation can be performed with certainty.

Furthermore, since reading of information of the front face of a paper sheet 40 is performed optically by the first optical image reading unit 412 and reading of information of the rear face of the paper sheet 40 is performed optically by the second optical image reading unit 414, reading of image information on the opposite faces of the paper sheet 40 can be performed rapidly, and the processing speed of a double-side original is improved significantly.

Further, with the image reading apparatus of the present embodiment, the following advantages can be achieved due to its structural characteristics.

In particular, since the paper transport path 310 connected to the paper supply mechanism 200 is constituted from the inclined transport path 312 and the paper reversing transport path 314 without involving a horizontal transport path, the paper transport path 310 requires a comparatively small depthwise space, and accordingly, there is an advantage in that the image reading apparatus can be reduced in size as much. Further, there is another advantage in that a paper sheet can be transported rapidly from the paper supply mechanism 200 to the stacker mechanism 300 and image reading can be performed at a high speed. Naturally, the reduction in space allows an increase in size of the paper sheet hopper or the paper stacker, which allows reading of a paper sheet of a greater size.

By the way, referring back to FIG. 3, analog video signals from the CCD arrays 436AA and 436AB of the image reading units 412 and 414 are amplified by the amplification circuits 64A and 64B, respectively, and the analog video signals of the outputs of the amplification circuits 64A and 64B in portions (bits) in which, for example, the photosensitive portions of the CCD arrays 436AA and 436AB are masked are sampled and held by the black level setting circuits (sample hold circuits) 71A and 71B and are connected as reference signals for a black level to the lower limit sides (VRB) of the analog to digital conversion circuits 60A and 60B, respectively. For a reference signal for a white level, a signal obtained by digital to analog conversion of a white level value of each bit of each line obtained by scanning in the last scanning cycle and stored in the memory circuit 72-1 or 72-2 by the digital to analog conversion circuit 62A or 62B is used and connected to the upper limit side (VRT) of the analog to digital conversion circuit 60A or 60B.

Consequently, the analog to digital conversion circuit 60A or 60B outputs a multiple value digital signal on the scale of 256 gradations between the reference level (VRT) for white and the reference level (VRB) for black. In this instance, for the reference level for white, an analog value of a white level produced with reference to a white level obtained by scanning of the image in the last scanning cycle is used, and for the reference value for black, an analog value at a dot at which the photosensitive element of the CCD array 436AA or 436AB is masked is used.

It is to be noted that, in this instance, for example, in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level value to be provided to the analog to digital conversion circuit 60 is varied in response to an instruction signal from the MPU circuit 150 by the white level information correction circuit 70. In particular, as seen from FIGS. 14 to 20, white level information stored in the memory circuit 72-1 (or 72-2) or the storage area 72-11 (or 72-12) is taken out and stored into the register 73-1 (or 73-2) or 73, and then the output of the register 73-1 (or 73-2) or 73 is multiplied by m by the data magnification variation circuit 74. Then, the output of the data magnification variation circuit 74 is selected by the selection circuit 75b and then processed by required processing by the white level algorithm circuit 77, and then the white level of the thus varied magnification is stored into the other memory circuit 72-2 (or 72-1) or the other storage area 72-12 (or 72-11) by way of the selection circuit 75c. Then, the white level of the varied magnification is used as a conversion reference of the analog to digital conversion circuit 60.

Thereafter, so far as such blue print paper sheets are used, the white level from the memory circuit 72-2 (or 72-1) or the storage area 72-12 (or 72-11) is extracted, and now, the output of the memory circuit 72-2 (or 72-1) or the storage area 72-12 (or 72-11) is inputted by way of the selection circuit 75b to the white level algorithm circuit 77. Then, the processing described above is performed subsequently by the white level algorithm circuit 77 so that the white level value may have an appropriate value to update the white level.

Consequently, even in such a case that paper sheets whose ground color is white have been read till now and blue print paper sheets of a different ground color are to be read subsequently, the white level information correction apparatus copes with this sufficiently and can perform analog to digital conversion with a high degree of accuracy.

Then, the image signal after digital conversion by the analog to digital conversion circuit 60A or 60B is transferred to the image processing section 68A or 68B for next image processing such as, for example, emphasis processing to emphasize the contrast between white and black or "dither processing; binary digitization processing" for a net point image such as a photograph image as described hereinabove.

Further, in each of the image data processing systems D1 and D2, digital information obtained from the analog to digital conversion circuit 60A or 60B is sent, after it is processed by emphasis processing and/or binary digitization processing by the image processing section 68A or 68B, to the outputting section 90 as seen from FIG. 3, and paper front face data and paper rear face data are transferred from the outputting section 90 to the host computer (not shown).

Upon such transfer, data (paper front face data) from the first operating image reading unit 412 from between the first optical image reading unit 412 and the second optical image reading unit 414 from which paper image information is to be read out first are first transferred successively by way of a data transfer line. Meanwhile, data (paper rear face data) from the later operating image reading unit 414 from between the first optical image reading unit 412 and the second optical image reading unit 414 from which paper image information is to be read out later are temporarily stored into the DRAM 92, and, after the data from the first operating image reading unit 412 are transferred completely, the stored data from the later operating image reading unit 414 are successively transferred at a rate higher than the transfer rate of the data from the first operating image reading unit 412 by way of another data transfer line.

Consequently, even if image reading by the later operating image reading unit 414 is started before image reading by the first operating image reading unit 412 is completed, data within the overlapping period can be held with certainty. Besides, data from the later operating image reading unit 414 from which data are to be transferred later can be transferred rapidly to the host computer side.

Accordingly, even where the image reading units 412 and 414 are disposed in the proximity of the paper transport path in order to achieve minimization, reduction in weight and compaction of the apparatus, data of the front and rear faces of a paper sheet can be transferred to the host computer side before the paper sheet is discharged to the stacking mechanism 500. Consequently, even if paper sheets are successively transferred at a high speed while achieving minimization, reduction in weight and compaction of the apparatus, data of the front and rear faces of each paper can be read and transferred to the host computer side satisfactorily.

In this instance, the paper leading end detection circuit 450 detects the leading end of a paper sheet based on a paper end detected for the first time, and then when the MPU circuit 150 receives such paper leading end detection signal as an interruption requesting (IRQ) signal from the paper leading end detection circuit 450, the MPU circuit 150 develops, in response to the interruption requesting (IRQ) signal, a command signal to read the front face of the paper sheet. As a result, also the selection circuit 96 is switched to the paper front face data side.

Meanwhile, the paper trailing end detection circuit 451 detects the trailing end of the paper sheet based on a paper end detected for the second time, and then when the MPU circuit 150 receives such paper trailing end detection signal as an interruption requesting (IRQ) signal from the paper trailing end detection circuit 451, the MPU circuit 150 develops, in response to the interruption requesting (IRQ) signal, a command signal to read the rear face of the paper sheet. As a result, also the selection circuit 96 is switched to the paper rear face data side.

Where the paper leading end detection circuit 450 and the paper trailing end detection circuit 451 of such circuit construction as described above are employed, the leading end and the trailing end of a paper sheet can be detected with certainty with a common circuit construction and with a simple construction. Consequently, reading timings of paper front face data and paper rear face data or reading switching timings between them can be controlled precisely irrespective of the type of a paper sheet being transported. As a result, it is comparatively possible for the image reading apparatus to allow paper sheets of various sizes to be transported to read information on them.

C. Other Embodiment

Referring now to FIG. 33, there is shown an image reading apparatus of another embodiment of the present invention. The image reading apparatus is constructed as an apparatus of a complex type which allows both of automatic paper supplying reading wherein an original is moved while an optical reading unit 1021 is fixed to read an image of the original and flat bed reading wherein an original is fixed while the optical reading unit 1021 is moved to read an image of the original.

Upon automatic paper supplying reading, the optical reading unit 1021 is first moved, before reading of an image, to a position below a white reference sheet and reads a white reference of the write reference sheet, and is then returned to its reading position (first reading position) for an automatically supplied paper sheet to thereafter read an image on the surface of each of originals which successively pass above the optical reading unit 1021.

On the other hand, upon flat bed reading, an original is placed on a flat bed 1700, and the optical reading unit 1021 first reads, below the white reference sheet, the white reference and then moves to another position (second reading position) below the flat bed 1700, and then reads an image of the original on the flat bed 1700 while it is moving.

Further, reference characters a.sup.1 and a.sup.2 in FIG. 33 denote reading positions for a front face and a rear face of an original, respectively. Meanwhile, reference characters 1101a and 1101b denote platen glass plates for use upon reading of a front face and a rear face of an original, respectively.

Reference characters 1416a and 1416b denote white reference sheets for use upon reading of a front face and a rear face of an original applied to the platen glass plates 1101a and 1101b, respectively.

An original is fed between the platen glass plates 1101a and 1101b by a pair of feed rollers 1009 and is discharged by a pair of discharge rollers 1013. An original feed path 1023 along which the original is fed is indicated by an arrow mark.

When no original is fed, the optical reading unit 1021 can read, at the reading position a.sup.1, the white reference sheet 1416b applied to the platen glass plate 1101b through the opposed platen glass plate 1101a. Similarly, an optical reading unit for reading a rear face of an original not shown can read, at the reading position a.sup.2, the white reference sheet 1416a applied to the platen glass plate 1101a.

Then, when an original is fed to the reading position a.sup.1 or a.sup.2, the corresponding white reference sheet is hidden behind the original, and, for example, the optical reading unit 1021 operates to read the front face of the original passing there.

It is to be noted that, upon automatic paper supplying reading, while the optical reading unit 1021 remains fixed at the position shown in FIG. 33, it can successively read the white reference sheet 1416b and the front face of the original, but upon flat bed reading, the optical reading unit 1021 moves, after it reads the white reference sheet 1416b at the position shown in FIG. 33, rightwardly to the position below the flat bed 1700 to read an image of the original on the flat bed 1700. It is to be noted that the flat bed 1700 is used when the original is of the type which cannot be stacked on a hopper such as, for example, a book.

Further, in FIG. 33, reference characters 1419a and 1419b denote each a thin glass plate serving as a cover, 1421a and 1421b denote each a double-sided adhesive tape, 1025a and 1025b denote each a lamp for illuminating an original, reference numeral 1027 denotes a mirror, 1031 a lens, 1033 a CCD, 1428 a leaf spring, and 1429 a projection.

The white reference sheets 1416a and 1416b are sandwiched between the platen glass plates 1101a and 1101b and the thin glass plates 1419a and 1419b, respectively, and the platen glass plates 1101a and 1101b and the thin glass plates 1419a and 1419b are adhered to each other by the double-sided adhesive tapes 1421a and 1421b, respectively. When no original is present at a reading position, the corresponding white reference sheet 1416a or 1416b is illuminated by the lamp 1025a or 1025b so that reflected light may be read by the CCD 1033 in the optical reading unit 1021.

The platen glass plate 1101a is mounted for movement by a small distance (for example, approximately 2 to 3 mm) leftwardly in FIG. 33, but is normally biased rightwardly to the position shown in FIG. 33 by the leaf spring 1428. Upon reading of the front face of an original, the optical reading unit 1021 is moved by a small distance (2 to 3 mm) rightwardly from the position shown in FIG. 33. During such movement, the optical reading unit 1021 reads the white reference sheet 1416b for a plurality of raster lines. A signal obtained by the reading is sent to a white level follower circuit (usually provided in an image reading apparatus), by which a white level is set removing, from the received signal, a disturbance to a white reference level caused by dust sticking to or soiling to the platen glass plate.

On the other hand, upon reading of the rear face of an original, the optical reading unit 1021 is moved leftwardly to move the platen glass plate 1101a leftwardly (2 to 3 mm) by way of the projection 1429 against the force of the leaf spring 1428. During such movement, the white reference sheet 1416a is read for a plurality of raster lines by a CCD (not shown) for reading of the rear face of an original, and a white level is set removing an influence of dust or soiling from a white reference level in a similar manner as described above.

By the way, an image signal read by the CCD 1033 of the optical reading unit 1021 is transferred to a host computer by way of an amplification circuit 181, an analog to digital (A/D) conversion circuit 182, an image processing section 183 and a data transfer processing section 184 as seen from FIG. 35.

Referring to FIG. 35, the image reading apparatus further includes a paper end detection apparatus 190, and data transfer control by the data transfer processing section 184 is performed in accordance with paper end information detected by the paper end detection apparatus 190.

Referring to FIG. 34, in the image reading apparatus of the present embodiment, a white reference providing white tape 1701 which is provided adjacent an end portion of an original frame of the flat bed 1700 has two different start marks SM1 and SM2 disposed in mutually perpendicular directions in an opposing relationship to the optical reading unit 1021 which moves to the white reference providing white tape 1701, that is, on the rear face opposite to the original setting face.

The start mark SM1 is a rectangular mark extending in the longitudinal direction of a paper sheet (sub scanning direction), and the start mark SM2 is a rectangular mark extending in a widthwise direction of a paper sheet (main scanning direction).

The paper end detection apparatus 190 detects an end of a paper sheet from image signals from the two start marks SM1 and SM2 obtained from the optical reading unit 1021. To this end, the paper end detection apparatus 190 includes, as shown in FIG. 36, a comparator 191, a pair of counters 192A and 192B, a pair of comparators 194A and 194B, a pair of latches 196A and 196B, another pair of counters 193A and 193B, another pair of comparators 195A and 195B, an AND circuit 197, a latch 198 and a gate circuit 199.

The comparator 191 binary digitizes a video signal with reference to a predetermined slice level. The counter 192A receives an output of the comparator 191 and counts the width of the start mark SM1 using a clock signal CC (the clock CC is outputted once for one dot). The comparator 194A receives an output of the counter 192A and a width control signal (width control signal 1) for the start mark SM1 and discriminates whether the start mark SM1 has a width greater than a prescribed width. The latch 196A temporarily latches an output of the comparator 194A to remove noise or chattering.

The counter 192B receives the output of the comparator 191 and counts the width of the start mark SM2 using the clock signal CC. The comparator 194B receives an output of the counter 192B and a width control signal (width control signal 2) for the start mark SM2 and discriminates whether or not the start mark SM2 has a width greater than another prescribed width. It is to be noted that the counter 192B receives, at an enable terminal EN thereof, an output of the latch 196A (detection signal of the start mark SM1) so that it may operate after the start mark SM1 is detected. The latch 196B temporarily holds an output of the comparator 194B to remove noise or chattering.

The counter 193A receives the output of the latch 196A and counts the number of lines of the start mark SM1 using a clock signal DD (the clock DD is outputted once for one line (at the end of a line)). The comparator 195A receives an output of the counter 193A and a length control signal (length control signal 1) for the start mark SM1 and discriminates whether or not the start mark SM1 has a length greater than a prescribed length.

The counter 193B receives an output of the latch 196B and counts the number of lines of the start mark SM2 using the clock signal DD. The comparator 195B receives an output of the counter 193B and a length control signal (length control signal 2) for the start mark SM2 and discriminates whether or not the start mark SM2 has a length greater than the prescribed length.

The AND circuit 197 receives outputs of the comparators 195A and 195B and outputs "1" when both of the start marks SM1 and SM2 satisfy the requirements for the width and the length. The latch 198 temporarily holds an output of the AND circuit 197 to remove noise and chattering.

Accordingly, it is recognized that the paper end detection apparatus 190 includes first start mark detection means (the comparator 191, counter 192A, comparator 194A, latch 196A and counter 193A) for detecting the start mark SM1, second start mark detection means (the comparator 191, counter 192B, comparator 194B, latch 196B and counter 193B) for detecting the start mark SM2, and paper leading end detection signal outputting means (the comparators 195A and 195B, AND circuit 197, latch 198 and gate circuit 199) for outputting a paper leading end detection signal when a start mark is detected by both of the first start mark detection means and the second start mark detection means.

In the paper end detection apparatus 190 having the construction described above, a video signal is first binary digitized with reference to the predetermined slice level by the comparator 191 as seen from FIG. 36.

Then, the counter 192A counts the width of the start mark SM1, and the comparator 194A discriminates whether or not the start mark SM1 has a width greater than the prescribed width. The output of the comparator 194A is temporarily held by the latch 196A to remove noise or chattering.

Meanwhile, the counter 192B counts the width of the start mark SM2, and the comparator 194B discriminates whether or not the start mark SM2 has a width greater than the prescribed width. The latch 196B temporarily latches an output of the comparator 194B to remove noise or chattering. It is to be noted that the counter 192B operates after the start mark SM1 is detected.

Further, the counter 193A counts the number of lines of the start mark SM1, and the comparator 195A discriminates whether or not the start mark SM1 has a length greater than the prescribed length.

Meanwhile, the counter 193B counts the number of lines of the start mark SM2, and the comparator 195B discriminates whether or not the start mark SM2 has a length greater than the prescribed length.

Then, when both of the start marks SM1 and SM2 satisfy the requirements for both of the width and the length, the AND circuit 197 outputs "1", and the output is temporarily latched by the latch 198. Consequently, noise or chattering is removed from the output signal of the AND circuit 197. Then, where the system is constructed such that outputting of a gate signal or data is based on an instruction of a CPU, an output of the latch 198 is supplied as an interrupt requesting signal (IRQ) to the CPU. On the contrary, where the system is constructed otherwise such that outputting of a gate signal or data is performed automatically, the output of the latch 198 is used as an enable signal for an outputting IC. The enable signal is outputted by way of the gate circuit 199 of the AND circuit construction.

In this manner, with the paper end detection apparatus 190, since an end of a paper sheet is detected from image signals from the two start marks SM1 and SM2 obtained by the optical reading unit 1021, an end of a paper sheet can be detected accurately. Consequently, an accurate data transfer timing can be obtained readily.

The present invention is not limited to the specifically described embodiment, and variations and modifications may be made without departing from the scope of the present invention.