Resolution Conversion Method of Pictorial Image Processing System

The present invention relates to a resolution conversion method of a pictorial image processing system, and more particularly to a resolution conversion method for converting pictorial image data of resolution of 200 dpi read by a scanner to pictorial image data of resolution of 300 dpi so that the converted pictorial image data is outputted to a printer and/or any displaying apparatus of resolution of 300 dpi.

Generally a pictorial image processing apparatus such as a facsimile adopts a heat sensitive recording method utilizing a thermal printer head element for recording a processed document. In the heat sensitive recording method, the record paper is expensive and has poor preservation of record even if black dots are expressed relatively well by using a special heat sensitive paper recorded by the thermal printer head. Recently a facsimile using a normal paper as a record paper is getting popular, and the facsimile using the normal paper adopts an electrophotographic method utilizing a LBP(laser beam printer) or LED(light emitting diode). However, a resolution is deteriorated owing to a quality of the normal paper and a minute detail is not expressed well in the electro-photographic method as compared with the heat sensitive recording method. Consequently a dark recorded paper is produced. Accordingly the facsimile using a normal paper should increase a resolution of its recording device more than a resolution of a manuscript scanner, and a number of data processed and transmitted to the recording device in proportion to the increased resolution. In 1980, Consultant Committee of International Telegraph & Telephone (hereinafter called as "CCITT") recommends that a scanner of G3 facsimile apparatus should have about 203 dots per inch ( 8 mm per character) for a horizontal resolution and about 196 dots per inch (7.7 mm per line of the character) for a vertical resolution. As compared with the recommendation of CCITT, a current recording device of the normal paper facsimile has normally a resolution of 300 dots per inch for a vertical and a horizontal directions. Accordingly the number of data to be recorded should be enlarged vertically and horizontally by 1.534 times. An pictorial image quality of an output is influenced by a method of magnification when the data to be recorded is magnified according to a change of resolution. A zero-order interpolation is used mostly for magnification because of simplicity of embodiment. As the zero-order interpolation repeats simply a pixel of every even number in a horizontal direction and a line of every even number in a vertical direction, the black dots are converted asymmetrically partially, and furthermore it is occurred often that the black dots are piled on one another. Consequently the zero-order interpolation may deteriorate greatly an pictorial image quality of half tone expressing an intensity by a distribution density of random discrete black dots in the same way of an error diffusion method.

There is disclosed a new magnification method utilizing a combination of logical product and sum so as to eliminate such problems as deterioration of pictorial image quality in the zero-order interpolation and to easily perform such a thing, when it is necessary to convert a resolution of the data to be recorded in a normal paper facsimile.

With reference to FIG. 1, a configuration of conventional facsimile system is described as follows. A central processing unit 2 (hereinafter called as "CPU") controls a system according to a given program. A memory 3 stores program data, protocol data and character data, and accesses and/or stores the data by a control of the CPU 2. An operating panel equipment 1 comprises a plurality of keys for generating key data and a device for displaying data of the CPU, and transmits the key data to the CPU 2 if the key data is generated. A charge coupled device 8 (hereinafter called as "CCD") reads pictorial image data from a manuscript and performs a photoelectric conversion of the pictorial image data. A scanner 6 transmits the converted pictorial image data from the CCD 8 to the CPU 2. A modem 4 modulates output data of the CPU 2 into an analog and demodulates received analog data by a control of the CPU. A signal processor 9 decodes and/or encodes output pictorial image data from the scanner 6 and the modem 4. A PSTN line interface circuit 5 establishes a speech loop of a telephone line and interfaces the signal of the modem 4 with a signal of the telephone line by a control of the CPU 2. A printer 7 prints input data by a control signal of the CPU 2.

FIG. 2 illustrates an example of conventional conversion from a resolution of 200 dpi to a resolution of 300 dpi and FIGS. 3A, 3B illustrates an example of a conventional resolution conversion of pictorial image data.

With reference to FIGS. 1 to 3, a conventional resolution conversion method by a magnification utilizing the disclosed zero-order interpolation is described as follows. Pictorial image data of 200 dpi read by the CCD 8 or received from the modem 4 through the PSTN line interface circuit 5 is stored in the memory 3. With reference to FIG. 2, the CPU 2 converts a resolution of first line by writing a pixel of odd number once and writing a pixel of even number twice repeatedly in a horizontal direction (as shown in FIG. 2) after reading by two pixels of the pictorial image data of a first line at a time from the memory 3. After completing a resolution conversion of a first line, the CPU 2 performs a resolution conversion of a second line in the same way of processing the first line, and writes the second line of resolution conversion one more time repeatedly under the second line in a vertical direction. After finishing the resolution conversion, the converted pictorial image data is stored in the memory 3 and transmitted to the printer 7 for printing. By repeating such an operation, the CPU 2 converts an input pictorial image signal of resolution of 200 dpi as shown in FIG. 3A to pictorial image data of resolution of 300 dpi as shown in FIG. 3B by writing a pixel of even number twice repeatedly in a horizontal direction and a line of even number twice repeatedly in a vertical direction.

Since pictorial image data of resolution of 200 dpi is converted to pictorial image data of resolution of 300 dpi by writing twice repeatedly a pixel of even number in horizontal direction and a line of even number in vertical direction respectively, a horizontal and a vertical stripes are appeared, pictorial image quality is deteriorated, and pictorial input image signal is distorted.

It is an object of the present invention to provide a resolution conversion method for securing a high quality of pictorial image.

It is another object of the present invention to provide a resolution conversion method for converting pictorial image data of resolution of 200 dpi to pictorial image data of resolution of 300 dpi so as to print without deteriorating a quality of a picture in a facsimile.

It is still another object of the present invention to provide a resolution conversion method for converting pictorial image data of resolution of 200 dpi to pictorial image data of resolution of 300 dpi so as to prevent a distortion of an input pictorial image signal in a facsimile.

According to the present invention, a resolution conversion method of a pictorial image processing system comprises the steps of storing pictorial image data of two lines received as an unit of line in an odd and an even line of a two-line buffer, covering the pictorial image data stored in the two-line buffer with a mask by 2.times.2 pixels so as to prevent an overlapping one another, performing a resolution conversion by a logical operation after reading the pictorial image data covered by the mask in the second step and storing the converted pictorial image data in a three-line buffer, and recording the pictorial image data stored in the three-line buffer on a memory when the resolution conversion of two lines is finished, and repeating the above steps until all lines of the pictorial image data are recorded completely.

The present invention will now be described more specifically with reference to the drawings attached only by way of example.

FIG. 1 shows a block diagram of a general facsimile;

FIG. 2 shows an example of conventional conversion from resolution of 200 dpi to resolution of 300 dpi;

FIGS. 3A, 3B show an example of conventional pictorial image data conversion;

FIG. 4 shows a flowchart according to the present invention;

FIG. 5A, 5B show an example of inventive conversion from resolution of 200 dpi to resolution of 300 dpi;

FIG. 6A, 6B show a view of pictorial image data covered with a 2.times.2 mask according to the present invention;

FIGS. 7A, 7B show an example of pictorial image data conversion of the present invention; and

FIGS. 8A to 8P show examples of conversion of the present invention according to a state of pictorial image data.

Pictorial image data of two lines received as an unit of line is stored each in an odd and an even line buffers of a two-line buffer. And the pictorial image data of the two-line buffer is covered with a mask by 2.times.2 pixels so as not to be overlapped one another. Thereafter a resolution is converted by a logical operation after reading the pictorial image data covered with a mask by 2.times.2 pixels, and the converted data is stored in a three-line buffer. The pictorial image data stored in the three-line buffer is transmitted to a memory when the resolution coversion of two lines is finished. The resolution coversion is performed by repeating the above operation until all lines of the pictorial image data is transmitted and recorded completely in the memory.

With reference to FIG. 4, pictorial image data of two lines is received as an unit of line and stored in an odd and an even line buffers of a two-line buffer in a first step. A given number of the pictorial image data stored in the two-line buffer are covered with a mask in a second step. A resolution of the pictorial image data covered by the mask is converted by a logical operation and the converted resolution is stored in a three-line buffer in a third step. In a fourth step, the above steps are repeated until all lines of the pictorial image are recorded completely in a memory, after recording the pictorial image data stored in the three-line buffer on a memory when the resolution conversion of the two lines is finished in the third step.

With reference to FIG. 1 and 4 to 8, an example of embodiment is described more specifically as follows.

A configuration of the inventive facsimile system is the same one as shown in FIG. 1 and described more specifically as follows. A central processing unit 2 (hereinafter called as "CPU") receives pictorial image data of 200 dpi read as an unit of line by a CCD 8 or a modem 4 through a PSTN line interface unit 5, and reproduces and stores the pictorial image data in a two-line buffer after encoding or decoding the pictorial image data by a signal processor 9. At this time, the pictorial image data is binarization data as "black" or "white", and for example, the "black" is indicated as "1" and the "white" as "0, or vice versa "black" as "0" and "white" as "1". Hereby it is assumed that the "black" is "1" and the "white" is "0". The CPU 2 covers the two-line buffer with a mask by 2.times.2 pixels as shown in FIG. 6A. The CPU 2 reads the pictorial image data covered with the mask by 2.times.2 pixels as shown in FIG. 7A, and performs a resolution conversion of the pictorial image data into pictorial image data of 9 pieces by a logical operation as shown in FIG. 7B and stores the converted data in the memory 3. And the CPU 2 transmits the pictorial image data stored in the memory 3 to a printer 7 for printing.

With reference to FIG. 4, the CPU 2 stores pictorial image data of 200 dpi received as an unit of line from a scanner 6 or a modem 4 and processed by a signal processor 9 in an odd and an even line buffers of a two-line buffer in a step of 4A. In a step of 4B, the pictorial image data of the odd and the even line buffers of the two-line buffer is covered with a 2.times.2 mask by 2.times.2 pixels so as not to be overlapped as shown in FIG. 6A. In a step of 4C, the pictorial image data of 2.times.2 pixels covered with the 2.times.2 mask (as shown in FIG. 5A) is read and converted to pictorial image data of 9 pieces as shown in FIG. 5B by a logical operation and thereby designating each pixel for the pictorial image data of 9 pieces. The pictorial image data of the 9 pieces is obtained as follows. ##EQU1## A resolution converted data according to a state of the pictorial image data of the two-line buffer covered with the 2.times.2 mask has one of 16 cases as shown in FIGS. 8A to 8P, and a truth table is as follows.

In the table 1, "1" stands for pictorial image data of "black" pixel and "0" stands for pictorial image data of "white" pixel. In a step of 4D, the resolution converted pictorial image data of the 9 pieces are stored in a three-line buffer so as not to be overlapped one another. In a step of 4E, it is checked whether the resolution conversion of the pictorial image data for the two lines is completed. If the resolution conversion of the pictorial image data for the two lines is not completed in the step of 4E, the mask is moved to a right side by 2.times.2 pixels so as to cover next 2.times.2 pixels in the two-line buffer without overlapping in a step of 4F, and the CPU 2 returns to the step of 4C. If the resolution conversion of pictorial image data for the two lines is completed in the step of 4E, the resolution converted pictorial image data of the two lines stored in the three-line buffer is recorded in a memory 3 in a step of 4G, and then a step of 4H is performed. In the step of 4H, it is checked whether all lines of the pictorial image data received are recorded completely or not, and if not, the next two lines of the pictorial image data are read and stored in the two-line buffer in a step of 4I, and the CPU 2 repeats the above steps after returning to the step of 4A. But if all lines of pictorial image data is recorded completely in the step of 4H, the resolution conversion process is finished.

Consequently a high quality of pictorial image is secured because a distortion of data from a telephone line is prevented and stripes of a vertical and a horizontal directions are not appeared by converting from resolution of the pictorial image data of 200 dpi of two lines received as an unit of line to resolution of pictorial image data of 300 dpi of three lines in pictorial image processing system such as a facsimile.

While the invention has been particularly shown and described with reference to the preferred embodiment of the present invention thereof, it will be understood by those skilled in the art that the foregoing and other changes in form and details may be made without departing from the spirit and scope of the present invention.