KR0153418B1 - Image data compression method - Google Patents

Abstract

[Technical field to which the invention described in the claims belongs]

The present invention relates to an image data reduction method for selecting and reducing original image data.

[Technical Problem to Solve]

The present invention provides an image data reduction method for preserving a feature portion of an image discarded by a sussampling rule when the image data is reduced, and emphasizing the outline of the image.

[Summary of the solution of the invention]

The present invention recognizes whether a pixel is selected according to a subsampling rule when the image data is reduced, and if a current pixel is selected after the previous pixel is not selected, outputs a result of logical operation between the previous pixel and the current pixel, and outputs the previous pixel. If is selected and the current pixel is selected, the current pixel is output as it is.

[Important Uses of the Invention]

The present invention can be importantly used when reducing image data in an image forming apparatus.

Description

How to reduce image data

1 is a state diagram schematically showing a selection state of a pixel according to a general subsampling.

2 is a state diagram showing reduction of conventional image data by subsampling.

3a and 3b are state diagrams showing an example of a conventional reduction method using a logical sum, and FIG. 3a shows a reduction process and results for image data having a large number of outline protrusions. FIG. Shows the reduction process and results for missing image data.

3 is a block diagram of a digital facsimile for carrying out a preferred embodiment of the present invention.

5 is a flow chart for performing a vertical reduction in accordance with an embodiment of the present invention.

6 is a flow chart of performing a horizontal reduction in accordance with a preferred embodiment of the present invention.

7 is a state diagram of original 10 × 10 image data for illustrating an example of a reduction process of a preferred embodiment of the present invention.

8 is a memory state diagram for recording the image data of FIG.

FIG. 9 is a diagram illustrating a state in which the image data of FIG. 7 is reduced by 70% in a vertical direction according to an exemplary embodiment of the present invention.

10 is a state diagram of a memory for recording the image data of FIG. 9 reduced in the vertical direction in accordance with a preferred embodiment of the present invention.

FIG. 11 is a diagram illustrating a state in which image data of one horizontal line of FIG. 9 is reduced by 70% in a horizontal direction according to an exemplary embodiment of the present invention.

FIG. 12 is a diagram illustrating a state in which the image data of FIG. 7 is reduced by 70% in the vertical and horizontal directions, respectively, according to an exemplary embodiment of the present invention.

13 is a state diagram of a memory for recording the image data of FIG. 12 reduced in the vertical and horizontal directions according to one preferred embodiment of the present invention.

14 is a reduced state diagram showing a state in which the original image data shown in FIG. 2 described above is reduced by 70% in the vertical direction according to the present invention.

15 is a reduced state diagram showing a state in which the original image data shown in FIG. 2 described above is reduced by 70% in the vertical and horizontal directions, respectively, according to the present invention.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of reducing image data, and more particularly, to a method of subsampling and logically summating original image data.

In general, in an image forming apparatus that requires a scanner such as a facsimile or digital copier, the scanned image data is often reduced in accordance with the intention of the user.

The method of subsampling the original image data is reduced conversion when a is less than 1 when the conversion ratio is defined as a, and enlarged when a is greater than 1. For example, when a = 0.7, the pixel selection state of the general subsampling method of reducing the line of pixel data having 10 input pixels by 70% is shown in FIG. This subsampling method selects a specific pixel to reduce and convert the image data according to the conversion ratio, and discards the remaining pixels.

In FIG. 1, I (i) is an i-th input pixel, a is a conversion ratio (also called intrinsic probability), d (i) is a diffusion probability, S (i) is a pixel selection probability, O (i) represents an output pixel. Referring to FIG. 1, a general subsampling method will be described. Therefore, the inherent probability of each pixel is equal to the conversion ratio a, and thus becomes 0.7. Therefore, the sum of the intrinsic probability and the diffusion probability d (i) of each pixel becomes the selection probability S (i) of the corresponding pixel.

Referring to FIG. 1 as an example, the first input pixel I (1) is not selected because the selection probability S (i) is 0.7, and the second input pixel I (2) has a selection probability S (i) of 1.4. Is selected and output as O (1). According to the subsampling method described above, when the tenth input pixel I 10 is calculated, the remaining pixels except for the first, fourth and seventh pixels among the ten input pixels are selected and output.

2 is a state diagram illustrating an example of reducing conventional image data according to subsampling, and shows a state in which 10 × 10 image data is reduced according to the subsampling method described above. Referring to FIG. 2, the original image data is reduced-converted to 7x7 image data except for the input pixels of the 1st, 4th, 7th rows and the 1st, 4th, 7th columns.

As described above, whether or not the pixel is selected in the conventional image data reduction method by subsampling is determined by the conversion ratio according to the subsampling rule regardless of the value of the pixel. That is, the conventional reduction method for subsampling selects image data according to the conversion ratio irrespective of whether the pixel value is '0' or '1', so that the image data whose pixel value is '1' among the pixels that cannot be selected is discarded. There is a problem that is lost.

In the conventional reduction method by subsampling as described above, the selection of pixels during subsampling is performed regardless of the brightness value of the pixels. Thus, as shown in FIG. 2, 10 × 10 original image data is reduced by 7 × 7. There was a problem in that the letter K is converted to an unrecognizable shape when converted into image data.

That is, in general, a document is expressed by an outline component, but when the outline component is discarded when the image data is reduced according to the conventional subsampling, important information of the document is lost.

Therefore, US Patent Nos. 5, 068, 905 (hereinafter referred to as Kodak Patents), filed by Kodak, have been published in order to improve the problems caused by the above-described subsampling conventional image data reduction method. The conventional image data reduction method according to the Kodak patent includes the concept of ORing of pixels when the image data is reduced.

Conventional reduction method according to the Kodak patent is not applied to the concept of 1% unit reduction, only 1/5, 1/4, 1/3,... There is a drawback of only reducing and enlarging. For example, in the case of reduction to 1/3, three consecutive pixels are ORed together to output the value.

3A and 3B are state diagrams showing an anomaly with conventional image data reduction methods according to the above-described Kodak patent. FIG. 3A shows a reduction process and results for image data having a large number of outline protrusions, and FIG. 3B shows a reduction process and results for image data without a protrusion outline.

First, referring to FIG. 3a described above, a conventional image data reduction method according to the Kodak patent is described. In FIG. 3a, X denotes an original image, and X denotes scanned image data of Z. The conventional image data reduction method according to the Kodak patent has a result of logically combining data of a vertical line first when the image data is reduced, and a result of logically combining data of a horizontal line after the logical data of the vertical line.

Meanwhile, referring to FIG. 3 described above, in FIG. 3B, X denotes an original image, and X denotes scanned image data of Z. The conventional image data reduction method according to the Kodak patent has a result of logically combining data of a vertical line first when the image data is reduced, and a result of logically combining data of a horizontal line after the vertical line.

As described above, the conventional image data reduction method according to the Kodak patent compares the result of the reduction in FIG. 3a with the result of the reduction in FIG. 3b, and the reduced result is the same even though the original image data is different. Do. Therefore, the conventional image data reduction method according to the Kodak patent has the same problem as the result of the reduction even though the original image data is different due to unconditional logical operation when the image data is reduced.

Thus, the conventional image data reduction method according to the Kodak patent has the disadvantage that the information on the outline of the image data is lost due to unconditional logical operation resulting in distortion of the reduced image. Therefore, the conventional image data reduction method according to the Kodak patent causes the smoothing of the image when the image data is reduced even in the case of an original having a large number of outlines, which causes a problem of not preserving the characteristics of the original image.

It is therefore an object of the present invention to provide a method for reducing image data having feature portions of an image not selected when the image data is reduced.

Another object of the present invention is to provide an image data reduction method which can preserve the characteristics of an outline when the image data is reduced.

It is still another object of the present invention to provide an image data reduction method capable of preserving outline components by logically combining image data not selected when the image data is reduced when selecting the next image data.

Still another object of the present invention is to provide a method for reducing image data by 1% even when performing an OR operation between image data.

According to an aspect of the present invention, there is provided a method of reducing an image data, comprising: a first step of specifying a vertical processing line of image data desired to be reduced recorded in a predetermined recording means, and receiving image data within the designated processing line; A second step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the vertical processing line, and a third step of determining whether the selectivity of the processing line specified in the second step is less than 1; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the third step, the process line is designated as an erase line, the selectivity of the processing line is designated as a diffusion probability, and the fourth step is returned to the first step. If the selectivity of the processing line is greater than or equal to 1 as a result of the determination of the third step, subtract 1 from the selectivity of the processing line. Assigning a value as the diffusion probability and selecting the processing line; and after performing the fifth step, image data of the previously erased designated processing line when there is a line previously designated as an erase line And ORing the pixel data of the selected processing line and outputting the logical sum value as a vertically reduced line, and outputting the processing line as a vertically reduced line as long as there is no previously designated line as an erase line. And after performing the sixth step, search whether the processing line is the last vertical line, and if the processing line is not the last vertical line, returns to the first step, and if the processing line is the last vertical line, shrink the vertical line. A seventh step of terminating the step; designating the horizontal processing line of the image data desired to be reduced recorded in the predetermined recording means and An eighth step of receiving image data in a line; a ninth step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the horizontal processing line; and a selectivity of the horizontal processing line specified in the ninth step being 1 Determining whether the horizontal processing line is smaller than 1 as a result of the determination in the tenth step and determining that the horizontal processing line is smaller than 1, and selecting the horizontal processing line as an erase line, and selecting the horizontal processing line If the selectivity of the processing line is greater than or equal to 1 as a result of the determination of the eleventh step and the tenth step to return to the eighth step, and the diffusion probability of the processing line A twelfth step of selecting 1 from the selectivity and selecting the processing line; and after performing the twelfth step, a process previously designated as an erase line When phosphor is present, the logical sum of the pixel data of the previously erased processing line and the pixel data of the selected processing line are output, and the logical sum is output as a horizontally reduced line. A thirteenth step of outputting the processing line as a horizontally scaled line as it is, and searching whether the processing line is the last horizontal line after performing the thirteenth step, and if the processing line is not the last line, goes to the eighth step In step 14, if the processing line is the last line, horizontal line reduction is terminated.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

4 is a block diagram of a digital facsimile for carrying out a preferred example of the present invention, wherein the central processing unit 3 processes the overall control operation and image reduction of the digital facsimile according to a predetermined program. The main memory 4 stores program data, protocol data and problem data, and accesses or stores data under the control of the central processing unit 3. OPE (OPERATING PANNEL EQUIPMENT; 13) is provided with a keypad to output the key data corresponding to the key input from the user to the central processing unit (3), receives the display data from the central processing unit (3) to the user It is provided with a display device which can display. The scanner 1 scans image data from an original and outputs it as photoelectrically converted image data. The scanner interface 2 outputs image data input from the scanner 1 as serial data. The pixel density converter 9 stores serial image data output from the scanner interface 2 into a page memory 10 to be described later, inputs received image data accessed from the page memory 10, and then executes the central processing unit 3. The image is reduced and output by the control signal. The page memory 10 stores scanned image data output from the pixel density converter 9 and received image data accessed and decoded from the main memory 4 in units of pages. The printer 8 prints the image data output from the pixel density converter 9 by the control signal of the central processing unit 3. The modem 6 is controlled by the central processing unit 3, and analog-modulated outputs the output data of the central processing unit 3, and demodulates and outputs the analog received image data. The public network interface 7 operates under the control of the central processing unit 3 to form a call loop with the public network, and interfaces the signal of the modem 6 with the signal of the telephone line. The ISDN interface 12 operates under the control of the central processing unit 3 to form an ISDN call loop, and interfaces scanned image data, received image data, and signals of the ISDN line. The codec 11 encodes the scanned image data to the main memory 4, and decodes the image data received through the modem 6 or the ISDN interface 12 to output the page data to the page memory 10. The DMAC 5 transmits the scanned image data hatched by the codec 12 to the modem 6 or the ISDN interface 12, and transmits the image data received through the modem 6 or the ISDN interface 12 to the main memory. (4) is transmitted to the codec 11 for decoding.

FIG. 5 is a flowchart showing a reduction of vertical components according to a preferred embodiment of the present invention, and FIG. 6 is a flowchart showing a reduction of horizontal components according to a preferred embodiment of the present invention.

FIG. 7 is a state diagram of original 10 × 10 image data for illustrating an example of a reduction process of a preferred embodiment of the present invention, and FIG. 8 is a state diagram of the page memory 10 for recording the image data of FIG. .

FIG. 9 is a state diagram showing the arrangement of vertically reduced image data P (m, j) after repeating the above-described process and ending 70% reduction in the vertical direction until s (10, 10) = s100. It is a state diagram showing the memory storage state of the vertically reduced image data P (m, j). FIG. 11 is a diagram illustrating a state in which image data of one horizontal line of FIG. 9 is reduced by 70% in the horizontal direction according to an exemplary embodiment of the present invention.

FIG. 12 is a diagram illustrating a state in which the image data of FIG. 7 is reduced by 70% in the vertical and horizontal directions, respectively, according to an exemplary embodiment of the present invention. FIG. 13 is a state diagram of a memory for recording the image data of FIG. 12 reduced in the vertical and horizontal directions according to one preferred embodiment of the present invention.

FIG. 14 is a reduced state diagram showing a state in which the original image data shown in FIG. 2 described above is reduced by 70% in the vertical direction according to the present invention. FIG. 15 is a reduced state diagram showing a state in which the original image data shown in FIG. 2 described above is reduced by 70% in the vertical and horizontal directions, respectively, according to the present invention.

Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to FIGS. 4 to 15. First, the reduction process of the image data can be divided into the reduction of the vertical component and the reduction of the horizontal component.

In general, the remotely scanned image data is stored in the page memory 10 as shown in FIG. 8 and then reduced, so it does not matter which component is first reduced. In a preferred embodiment of the present invention, for convenience, the vertical component is first reduced and then the horizontal component is reduced. In a specific embodiment of the present invention, when the image data is reduced by 70%, the conversion ratio a is 0.7.

First, in step 502 of FIG. 5, each variable value for determining the processing order of each input image data S (i, j) is initialized. Thereafter, in step 503, each variable d (i), K, a necessary for OR and image reduction is initialized, and a conversion ratio a is received. The diffusion probability d (i) means that if no line is selected, the line selectivity L (i) is diffused to the next line. Of course, when a line is selected (that is, when the line selectivity L (i) is greater than 1), the probability of decreasing 1 is selected. The logical sum determination value K represents a criterion for determining whether to perform a logical sum operation (AND, 간의) between the currently selected line and the previous line. That is, when the OR value K is 1, the OR operation is performed between the currently selected line and the previous line. When the OR value K is 0, the OR operation is not performed on the currently selected line and the previous line. I denotes a line number of 10 x 10 image data shown in FIG. 7, and j denotes a column number.

Whenever initialization or one line is processed in step 504, the line selection value i is increased by one. For example, when i = 1, the first line whose input image data s (1, j) is s1 to s10 is selected. . Thereafter, in step 505, the line selectivity L (i) is determined according to the following equation.

L (i) = d (i-1) + a

The expansion probability d (1-1) at i = 1 is 0, and the conversion ratio a is 0.7, so the line selectivity L (1) is 0.7. In this case, the line selectivity L (i) refers to a criterion for determining whether a certain line is selected.

In step 506, it is determined whether the current processing line is selected by comparing whether the line selectivity L (i) is greater than or equal to one. That is, if the line selectivity L (i) is less than 1, the current processing line is not selected. If the line selectivity L (i) is greater than or equal to 1, the current processing line is selected. Therefore, since the line selectivity L (1) is 0.7, which is smaller than 1, in step 507, without selecting the current processing line, a logical sum determination is performed to inform that the next processing line needs a one-to-one logical sum with the image data of the previous processing line. Set the value K to 1. In step 508, the line selectivity L (1) of the current processing line is diffused to the diffusion probability d (i) of the next line (d (1) = 0.7), and the process returns to step 504.

Thereafter, when i = 2 through the above-described steps 504 to 505, the second line s11 to s20 is processed, and the line selectivity L (2) becomes 1.4.

Therefore, in step 506, since the probability value of the line selectivity L (2) is 1.4, which is larger than the comparison value 1, the second lines s11 to s20 are selected, and in step 509, the diffusion probability d (2) is expressed by the following equation. Is calculated. In other words,

d (2) = L (2) -1 = 1.4-1 = 0.4

Thereafter, steps 510 to 517 are processes for processing the current processing line. First, m representing the line value of the pixel reduced in step 510 is increased by one. In step 511, it is determined whether the logical sum determination value K is 1.

Therefore, since the first line (i = 1) is not selected, since the logical sum value K is 1 according to the above-described process, the value of j is increased by 1 in step 512 so that the first pixel of the second line (i = 2) is increased. s (2, 1) = s11) is selected, and in step 513, the vertically reduced image data P (m, j) is determined and output according to the following equation.

P (m, j) = s (i, j) ∨ s (i-1, j)

For example, P (1, 1) = s (2, 1) ∨ s (1, 1)

As described above, when the logical sum determination value K is 1, the result of the logical sum operation of the pixel s (i-1, j) of the processing line previously designated as the erase line and the pixel s (i, j) of the currently selected processing line are shown. The image data is output as the vertically reduced image data P (m, j). For example, the vertically reduced image data P (1,1) is formed by ORing the corresponding image data s (1, 1) = s1 of the previous line and the image data s (2, 1) = s11 of the current processing line. ) = P1

Thereafter, in step 514, it is determined whether j representing the order of the processing image data of the current processing line is the end of the current processing line (j = 10). If not, the process returns to step 512 to increase the value of j by one. However, if it is the end of the line, it is determined whether the last line (i = 10) is read in step 518. If it is not the last line in step 514, since the current processing line is selected in step 518, the logical sum determination value K is determined as 0 and the process returns to step 504. However, if the last line in step 514, the reduction processing routine of the vertical component is terminated.

If the logical sum determination value K is not 1 in step 511, the j value indicating the order of the processed image data of the current processing line is increased by 1 in step 515. In operation 516, the input image data s (i, j) is output as the vertically reduced image data P (m, j). In step 517, the sequence value j of the current processing pixel is read, and it is determined whether j is the end of the current processing line. If not, the process returns to step 515 to increase the j value by one. However, if it is the end of the line, the value i is read in step 518 to determine whether it is the last vertical line.

In addition, it looks at the image data reduction method of the horizontal component according to an embodiment of the present invention. 6 is a flow chart showing a preferred embodiment of the present invention for the reduction of the horizontal component, the concept is the same as the concept of the reduction of the vertical component.

Once the reduction process of the vertical component is completed, as shown in FIG. 5, the horizontal component is reduced. As shown in FIG. 6, when the horizontal pixel reduction algorithm is started, first, in step 602, each variable value (i, j) for determining the processing order of the input image data P (i, j) is initialized. In operation 603, the variables d (i) and K necessary for the logical sum and image reduction are initialized, and a conversion ratio a is received. In the flowchart of FIG. 6, the diffusion probability d (i) means that the pixel selectivity S (j) is diffused to the next pixel when an arbitrary pixel is not selected in the horizontal line. Of course, when an arbitrary pixel is selected (i.e., when the pixel selectivity S (i) is greater than 1), the probability value 1 is reduced from the pixel selectivity S (i) is determined as the diffusion probability S (i) of the next pixel. . The logical sum determination value K represents a criterion for determining whether to perform a one-to-one logical sum with the next selected pixel when an unselected pixel occurs.

In step 604, each time the initialization or one row is processed, i is increased by 1 to select the next line to be processed. For example, when i = 1, the input image data p (1, j) is p1 to p10. Select the first line that is Thereafter, in step 605, each time one pixel in the line selected in step 604 is finished, j is increased by 1 to select a pixel to be processed, and the pixels in the line to be processed are sequentially processed once in step 605. Is selected.

Thereafter, in step 606, the pixels p (i, j) selected in steps 604 and 605 are received. In operation 607, the pixel selectivity S (j) of the pixel p (i) is determined according to the following equation.

S (i) = d (i-1) + a

For example, the diffusion probability d (1-1) at i = 1 is 0, and the conversion ratio a is 0.7, so the pixel selectivity S (1) is 0.7.

The above-described pixel selectivity S (i) is a criterion for determining whether to select a target pixel of the current processing line.

In operation 608, it is determined whether the target pixel of the current processing line is selected by comparing whether the pixel selectivity S (i) is greater than or equal to one. That is, if the pixel selectivity S (i) is less than 1, the target pixel of the current processing line is not selected. If the pixel selectivity S (i) is greater than or equal to 1, the target pixel of the current processing line is not selected. Select the target pixel. Therefore, since the pixel selectivity S (1) is 0.7, which is smaller than 1, in step 610, the pixel selectivity S (1) is diffused to the pixel selectivity S (1) of the current target pixel (d (1) = 0.7), and the ration decision value K is set to 1. The process then returns to step 605.

Thereafter, the second pixel p2 is selected when i = 2 and the pixel selectivity S (2) is 1.4 through the above-described steps 605 to 607.

Therefore, since the probability value of the pixel selectivity S (2) in step 611 is greater than the comparison value 1 as 1.4, the diffusion probability d (2) is calculated according to the following equation.

In other words,

d (2) = S (2) -1 = 1.4-1 = 0.4

In step 612, it is determined whether the logical sum determination value K is one. Therefore, the first pixel j = 1 is an unselected pixel, and since the logical sum value K is set to 1 by the above-described process, the output pixel r (j) is determined and output in step 613 according to the following equation. .

r (j) = p (i) ∨p (i-1)

When the ration decision value K is 1, an OR operation is performed between the previous pixel p (i-1) designated as the erase pixel and the current target pixel p (i), and the result is output to the output pixel r (j). do.

For example, the previous pixel p (1) and the current target pixel p (2) are ORed and output to the horizontally reduced output pixel r (1).

If the logical sum determination value K is 0 in step 612, the output pixel r (j) horizontally reduced in step 614 becomes the current target pixel p (i).

In step 616, it is determined whether the current target pixel is the last pixel of the processing line. If it is not the last pixel, in step 617, the logical sum correction value K is initialized and the process returns to step 605. However, if the current target pixel is the last pixel of the processing line by repeating the above steps 605 to 616, it is determined in step 618 whether it is the last line of the processing line, and if not, the process returns to step 604 to select the next processing line. If it is the last line, the processing routine ends.

Therefore, if the above-described horizontal component reduction process is summarized, as shown in FIG. 11, when the conversion ratio is 70% in any horizontal line having 10 pixels, the conversion ratio according to the general subsampling rule is first applied. Accordingly, the pixel to be selected and the pixel not to be selected are determined. That is, in the example of FIG. 11, the 1st, 4th, 7th pixels p1, p4, p7 are determined as pixels which are not selected according to the 70% conversion ratio, and the 2nd, 3rd, 5th, 6th, 8th, 9th pixels p2. , p3, p5, p6, p8, p9) are determined as the selected pixel. Then, the image data reduction method according to the present invention performs a logical OR with the next pixel selected instead of unconditionally discarding the unselected pixel. That is, in the present invention, the brightness value of the pixels p1, p4, and p7 that are not selected as the previous pixels of the selected pixels p2, p5, and p8 is logically combined with the brightness values of the selected pixels. Therefore, the image data reduction method according to the present invention selects the first, fourth, and seventh pixels p1, p4, and p7 that are not selected and the next, second, fifth, and eighth pixels * p2, p5, p8 that are selected. By performing each logical OR, the outline characteristic of the image to be reduced can be maximized.

As a result, the first image shown in FIG. 7 is finally reduced to 70% as shown in FIG. 12 according to the preferred embodiment of the present invention shown in FIG. 5 and FIG. It is stored in the page memory 10 in the state of the memory map shown in FIG.

According to the present invention, when the vertical and horizontal components of the original image shown in FIG. 2 are reduced by 70%, outlines are emphasized while maximizing the characteristics of the finally discarded pixel as shown in FIG.

Therefore, as described above, the present invention has an advantage of recognizing the shape even when the image data is reduced by preserving the characteristics of the pixels which are not selected. In addition, the present invention can minimize distortion of an image when the image is reduced by emphasizing the outline component as much as possible by including the characteristic of the pixel discarded when the image data is reduced in the processing of the next pixel. In addition, the present invention can reduce the image data by 1%, unlike the conventional image data reduction method according to the Kodak patent that can be reduced only by 1/2, 1/3 units.

Claims (10)

1. An image data reduction method, comprising: a first step of designating a vertical processing line of image data desired to be reduced recorded in a predetermined recording means and receiving image data in a designated processing line; A second step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A third step of determining whether the selectivity of the processing line specified in the second step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the third step, the process line is designated as an erase line, the selectivity of the processing line as a diffusion probability, and the fourth step returns to the first step; ; A fifth step of designating a probability value by subtracting 1 from the selectivity of the processing line as a diffusion probability and selecting the processing line when it is determined that the selectivity of the processing line is greater than or equal to 1 as a result of the determination in the third step; After performing the fifth step, if the previous line is designated as the erase line, logical operation is performed on the pixel data of the erase line and the pixel data of the selected processing line, and the result is output as a vertically reduced line. A sixth step of outputting the current processing line vertically reduced line if it is not designated as an erase line; After performing the sixth step, it is searched whether the processing line is the last vertical line. If the processing line is not the last vertical line, the process returns to the first step. And a seventh step. The image data reduction method according to claim 1, wherein each of the probabilities is calculated as a percentage. 1. An image data reduction method, comprising: a first step of designating a horizontal processing line of image data desired to be reduced recorded in a predetermined recording means and receiving image data within the designated processing line; A second step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A third step of determining whether the selectivity of the processing line specified in the second step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the third step, the process line is designated as an erase line, the selectivity of the processing line as a diffusion probability, and the fourth step returns to the first step; ; If it is determined that the selectivity of the processing line is greater than or equal to 1 as a result of the determination in the third step, the spreading probability is designated as a probability value obtained by subtracting 1 from the selectivity of the processing line, and selecting the processing line; Wow; After performing the fifth step, when there is a line previously designated as an erasing line, the logical sum of the pixel data of the erase line and the pixel data of the selected processing line is output, and the logical sum value is output as a horizontally reduced line. A sixth step of horizontally outputting the processing line when there is no line designated as a previous erase line; After performing the sixth step, it is searched whether the processing line is the last horizontal line. If the processing line is not the last horizontal line, the process returns to the first step. And a seventh step. The image data reduction method according to claim 3, wherein each of the probabilities is calculated as a percentage. In the image data reduction method; A first step of specifying a vertical processing line of image data desired to be reduced recorded in predetermined recording means and receiving image data in the designated processing line; A second step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A third step of determining whether the selectivity of the processing line specified in the second step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the third step, the process line is designated as an erase line, the selectivity of the processing line as a diffusion probability, and the fourth step returns to the first step; ; If it is determined that the selectivity of the processing line is greater than or equal to 1 as a result of the determination of the third step, a fifth step of designating a probability value by subtracting 1 from the selectivity of the processing line as a probability probability and selecting the processing line Wow; After performing the fifth step, when there is a line previously designated as an erase line, logical operation is performed on the pixel data of the erase line and the pixel data of the selected processing line, and the logic sum value is output as a vertically reduced line. A sixth step of outputting a processed line vertically reduced when there is no line designated as a previous erase line; After performing the sixth step, it is searched whether the processing line is the last vertical line, and if the processing line is not the last line, the process returns to the normal step. If the processing line is the last line, the vertical line is reduced. Terminating a seventh step; An eighth step of specifying a horizontal processing line of image data desired to be reduced recorded in predetermined recording means and receiving image data in the designated processing line; A ninth step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A tenth step of determining whether the selectivity of the processing line designated in the ninth step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in step 10, the processing line is designated as an erase line, the selectivity of the processing line is designated as a diffusion probability, and the process returns to the horizontal processing line designation step. Steps; In the determination result of the tenth step, if it is determined that the selectivity of the processing line is greater than or less than 1, the diffusion probability is set to a value obtained by subtracting 1 from the selectivity of the processing line, and the processing line is selected. And after performing the twelfth step, when there is a line previously designated as an erase line, logically perform operation on the pixel data of the erase line and the pixel data of the selected processing line, and output the logical sum value as a vertically reduced line. And a thirteenth step of outputting a horizontally reduced line of the selected processing line when there is no line previously designated as an erase line; Searching for whether the processing line is the last horizontal line after performing the thirteenth step; returning to the eighth step if the processing line is not the last horizontal line; and ending the horizontal line reduction if the processing line is the last line. The image data reduction method characterized by consisting of 14 steps. The image data reduction method according to claim 5, wherein each of the probabilities is calculated as a percentage. In the image data reduction method; A first step of designating a horizontal processing line of image data desired to be reduced recorded in predetermined recording means and receiving image data in the designated processing line; A second step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A third step of determining whether the selectivity of the processing line specified in the second step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the third step, the process line is designated as an erase line, the selectivity of the processing line as a diffusion probability, and the fourth step returns to the first step; ; If it is determined that the selectivity of the processing line is greater than or equal to 1 as a result of the determination in the third step, a probability value obtained by subtracting 1 from the selectivity of the processing line is designated as a spread probability and the fifth step of selecting the processing line. Wow; After performing the fifth step, when there is a line previously designated as an erase line, the logic sum of the pixel data of the process line previously designated as the erase line and the pixel data of the selected process line is performed to vertically reduce the logical sum value. A sixth step of outputting the processed line and outputting the processed line horizontally reduced when there is no line designated as the previous erase line; After performing the sixth step, it is searched whether the processing line is the last horizontal line. If the processing line is not the last line, the processing returns to the horizontal processing line designation step. Terminating a seventh step; An eighth step of specifying a vertical processing line of image data desired to be reduced recorded in predetermined recording means and receiving image data in the designated processing line; A ninth step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A tenth step of determining whether the selectivity of the processing line specified in the ninth step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the tenth step, the processing line is designated as an erase line, the selectivity of the processing line as a diffusion probability, and the eleventh step of returning to the ninth step; ; If it is determined that the selectivity of the processing line is greater than or equal to 1 as a result of the determination of the tenth step, a probability value of subtracting 1 from the selectivity of the processing line is designated as a diffusion probability and the 12th step of selecting the processing line; Wow; After performing the twelfth step, when there is a line previously designated as an erase line, the logical sum of the pixel data of the previously erased line and the pixel data of the selected processing line is performed, and the logical sum is converted into a vertically reduced line. A thirteenth step of outputting and outputting a processing line vertically reduced when there is no line previously designated as an erase line; After performing the thirteenth step, it is searched whether the processing line is the last vertical line. If the processing line is not the last line, the processing returns to the vertical processing line designation step. The image data reduction method comprising the fourteenth step. 8. The method of claim 7, wherein each of the probabilities is calculated as a percentage. 1. An image data reduction method comprising: a first step of designating a processing line of image data desired to be reduced recorded in a predetermined recording means and receiving image data in a designated processing line; A second step of designating a sum of a predetermined conversion ratio and a predetermined diffusion probability as a selectivity of the processing line; A third step of determining whether the selectivity of the processing line specified in the second step is less than one; If it is determined that the selectivity of the processing line is less than 1 as a result of the determination in the third step, the process line is designated as an erase line, the selectivity of the processing line as a diffusion probability, and the fourth step returns to the first step; ; A fifth step of selecting a processing line and designating a probability value of subtracting 1 from the selectivity as a diffusion probability if the selectivity of the processing line is greater than or equal to 1 as a result of the determination in the third step; After performing the fifth step, when there is a line previously designated as an erasing line, the logical sum of the pixel data of the previously erased line and the pixel data of the selected processing line is output, and the logical sum value is output as a reduced line. If there is no line designated as a previous erase line, outputting the selected processing line as it is; and after performing the sixth step, searching whether the processing line is the last line, and if the processing line is not the last line, And a seventh step of returning to the first step and ending the line reduction if the processing line is the last line. The method of claim 9; And calculating each of the probabilities as a percentage.