Solid State Imaging Device With Adaptive Pixel Correction

1. Field of the Invention

The present invention relates to a solid state imaging device which utilizes a solid state sensor having a substantial number of pixels (picture elements).

2. Description of the Prior Art

FIG. 2 illustrates a block diagram of an example of a prior art solid state imaging device. In FIG. 2, the reference numeral 1 designates a solid state sensor for imaging an external scene; 2 denotes an A/D converter for converting analog image output signals of the solid state sensor 1 into digital signals; 3 represents a frame memory for storing offset signals of the solid state sensor; 4 denotes an arithmetic unit for subtracting the offset signals of the frame memory 3 from the output digital signals of the A/D converter 2 during a normal operation; and 5 stands for a D/A converter/synchronization circuit for converting the outputs of the arithmetic unit 4 into analog video signals and for adding the synchronizing signals.

Next, attention will be focused on the operation of the prior device. The incident light beams OP are converted into analog image signals by means of the solid state sensor 1, and the analog image signals are further converted into digital image signals by use of the A/D converter 2. The outputs of the A/D converter 2 with respect to the uniform light beams incident upon the entire picture are stored as offset signals in the frame memory 3. The arithmetic unit serves to subtract the offset signals stored in the frame memory 3 from the outputs of the A/D converter 2 with respect to an object to be imaged during working time, thereby obtaining corrected outputs. The corrected outputs are transmitted to the D/A converter/synchronization adding circuit 5 which effects conversion into analog signals and multiplexing of synchronizing signals. The above-described control is performed by a timing generator 6 for creating a specific timing, in which case a clock for driving the solid state sensor is defined as the reference. Japanese Patent Published Disclosure No. 36987/87 discloses such a solid state imaging device.

There are, however, some inherent problems with the thus constructed conventional device. In the case of an image which is so defective that a certain pixel of the solid state sensor 1 is completely dead or insensitive even if the incident light beam emerges, the output of this pixel does not vary or varies only a little, and it follows that spots unfavourably appear on the picture as shown in FIG. 3. This phenomenon is an obstacle to obtaining high-quality images.

Moreover, where the above-described conventional device is employed in searching for a small body, it is likely that a mistaken target may be assumed to be the small body of interest. As a result, a solid state sensor having no defect of the kind described is required, and this demand will inevitably lead to an increase in the overall cost of a device produced to meet this demand.

It is a primary object of the present invention to provide a solid state imaging device which overcomes the above-mentioned obstacles or problems and is capable of outputting high-quality image signals even if there are defective pixels in a solid state sensor.

This and other objects are achieved by a solid state imaging device including a level detector connected to an output of the frame memory, means for detecting defective picture elements, and spatial arithmetic means for defining compensation signals by utilizing image signals associated with neighbouring pixels and for substituting the compensation signals as image signals in a position corresponding to the defective picture elements.

Other features and advantages of the invention will become apparent from the following description taken in connection with the accompanying drawings.

FIG. 1 is a block diagram of a solid state imaging device, illustrating one embodiment of the present invention;

FIG. 2 is a block diagram showing one example of a prior art solid state imaging device;

FIG. 3 is a conceptual display of a screen which is produced from a solid state sensor having a defective pixel;

FIG. 4 is a block diagram showing one example of a delay circuit;

FIG. 5 is a typical block diagram of a spatial arithmetic unit suited for the present invention; and

FIG. 6 is a block diagram showing one example of a level detector.

Referring first to FIG. 1, a block diagram of one embodiment of the present invention is shown. In the figure, components which are the same as those depicted in FIG. 2 are marked with the same symbols. A level detector, more precisely a digital level comparator generally indicated at 7, serves to judge each level of offset signals stored in a frame memory 3. The capacity of the frame memory may be determined, in accordance with the pixel number on a picture. The offset signals are obtained through the receipt of uniform light beams from uniform object by the solid state sensor and are stored in the frame memory, which is the same means as that employed in the prior art solid state imaging device. The numeral 8 represents a spatial processor composed of shift registers SR1 through SR3 and a spatial arithmetic circuit SU. The circuit SU, as is well known, controls parallel-simultaneous operation of the output signals from the shift registers SR1 through SR3 in which raster format scanned serial signals are converted into parallel-simultaneous signals. The capacity of each shift registers SR1 through SR3 corresponds to the pixel number on a horizontal scan line. The numeral 9 designates a delay circuit for adjusting a given timing by delaying signals of the level detector 7.

The delay circuit 9 has shift registers 20 coupled in series as illustrated in FIG. 4. The delay time of the circuit 9 is settled as the total delay time of the circuit 9 and said level detector 7 is equal to the total delay time of said arithmetic unit 4 and said shift registers in the spatial processor 8, as the delay circuit 9 can delay the output of said level detector 7 which corresponds to a certain pixel until the signals of that certain pixel and the adjacent pixels are output.

A spatial arithmetic unit suitable for this invention has seven adders 30, a 3-bit shifter 32 for dividing by 8 and a selector 34 that is controlled by the signal from said delay circuit 9, all as shown in FIG. 5. The spatial arithmetic unit receives nine signals from said shift registers SR1 through SR3 and a signal from said level detector 7 through said delay circuit 9.

In the thus constructed solid state imaging device, each level of the images is held by the frame memory 3, and the defective images are detected by the level detector 7 which judges the severity of the defect from the level thereof. The defective images detected are caused to be replaced by compensation signals of neighbouring peripheral pixels by means of the spatial processor 8.

An explanation will next be given relating the following formulae to FIG. 1.

Let the content stored in the frame memory 3 be b, and let the judging threshold value of the level detector 7 be .epsilon.. Based on the relationship of magnitude between the content b and the threshold value .epsilon., the output C of a delay circuit 9 will be generated after a predetermined delay:

The spatial processor 8 executes processing as follows: signals of eight neighbouring pixels in regard to a certain pixel X22 are fetched from part of shift registers SR1, SR2 and SR3, each having a delay time corresponding to a scanning line of the image; and spatial arithmetic is effected under the condition of the formula (1), while the signal of the pixel X22 is output under the condition of the formula (2).

To formulate the output of the spatial arithmetic unit SU shown in FIG. 5 in spatial processor 8, let this output be Y, let the signal corresponding to the pixel X22 be x22, and similarly let the eight neighbouring signals corresponding to X11, X12, X13, X21, X23, X31, X32 and X33 of FIG. 1 be x11, x12, x13, x21, x23, x31, x32 and x33, respectively. The output Y is expressed as:

Hence, at the pixel which has been judged to be defective, an average value for the eight neighbouring pixels is identified as a substitute signal. If there is no defect, the original signal is output as it is.

The above-described embodiment, as indicated by the formula (3), utilizes eight neighbouring pixels. However, the same effects can obviously be attained by performing the following processing which involves the use of four neighbouring pixels or four different neighbouring pixels. Namely, the formulae (5) and (6) which may substitute for the formula (3) are as follows:

In this case, the spatial arithmetic unit SU is composed of three adders, a 2-bit shifter and a selector, and five output terminals of the shift registers SR1 through SR3 are coupled to this spatial arithmetic unit.

In the above embodiment of the present invention, a defective pixel is as one in which the signal level is lower than the predetermined level of the digital comparator; however, a defective pixel may also be defined as one in which the signal level is not between a predetermined lower level and a predetermined upper level. In this case, a level detector having an upper level digital comparator 51, a lower level digital comparator 52 and an OR gate 53, as shown in FIG. 6, is used. The output signal of this detector (in FIG. 6) will be "0" if the input signal level of the detector is between a lower level set in the comparator 52 and an upper level set in the comparator 51, while the output signal will be "1" if the input signal level is outside the above range.

As discussed above, the present invention provides advantages in that the signal of the defective pixels can be compensated by the signals of the neighbouring pixels, and the yield of elements of a solid state camera is thereby remarkably improved.

While the present invention has been described in detail and with reference to a specific embodiment thereof, it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope thereof.