JP2004336211A - Compression processing method and compression system for image data - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a compression processing method for image data which efficiently separates an original picture into a color binary picture and a multi-color picture and compresses them with a little calculation quantity without ensuring a large memory areas. <P>SOLUTION: The method comprises a blocking step of dividing an original picture containing a color picture into a plurality of pixel blocks containing a specified number of pixels; a picture area discriminating step of discriminating each divided pixel block into a few-color pixel block with a small dispersion and a multicolored pixel block with a large dispersion, by obtaining and comparing the degree of dispersion obtained from a histogram formed about attribute information containing at least density information with a predetermined threshold; and a pixel data picture area separation compressing step which adds background colors to the little-color pixel block discriminated by the picture area discriminating step to generate color binary image data, compresses the image data in a specified binary picture compression form, adds the background colors to the multicolored pixel block discriminated by the picture area discriminating step to generate natural background image data, and compresses the image data in a specified binary picture compression form. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention separates the color image data read by the image reading means into a multicolor image area and a color binary image area, and compresses the multicolor image area according to a predetermined image compression format. The present invention relates to a method for compressing image data to be compressed by a predetermined binary compression format, and a system for implementing the method.
[0002]
[Prior art]
In recent years, with the spread of color scanners, color documents including multi-color images such as natural images and color binary images such as characters and diagrams have been read and digitized, and distributed to customers on CD-ROMs or via the Internet. Although distribution is increasing, it is necessary to read color images at a high resolution of about 300 to 400 dpi in order to digitize color images with high quality. Its delivery time takes time.
[0003]
For this reason, it is necessary to reduce the data amount of the digitized color image by compressing it. However, the multi-color image and the color binary image are grouped together and the conventionally used JPEG compression method is used. When compression is performed in a multivalued image compression format such as that described above, the quality of a multicolor image can be ensured, but the compression ratio of a color binary image deteriorates and the image quality around characters and diagrams deteriorates.
[0004]
Therefore, as shown in Patent Documents 1 to 5 below, after a color image is digitized and before compression, a multicolor image and a color binary image are separated in advance, and the multicolor image is compressed in a multivalued image compression format. On the other hand, a technique for compressing a color binary image in a binary image compression format has been proposed.
[0005]
However, in these conventional techniques, an algorithm for distinguishing a multicolor image from a color binary image is particularly complicated.
[0006]
That is, in Patent Literatures 1 and 2, when compressing an digitized image (hereinafter referred to as an original image), the outline of a character or a diagram is traced, and a portion surrounded by the outline is recognized as a color binary image. A color binary image is extracted and separated into a multicolor image and a color binary image, and the multicolor image is compressed in a multivalued image compression format, while the color binary image is compressed in a binary image compression format. ing.
[0007]
In Patent Document 3, an original image is divided into a plurality of blocks, and each image is classified into a foreground color (dark part) and a back color (bright part) in the block. The color binary image is determined by updating the scenery and reducing the size of the grid until it converges.
[0008]
[Patent Document 1]
JP-A-11-308463 [Patent Document 2]
JP-A-2002-369011 [Patent Document 3]
US Pat. No. 5,900,953
[Problems to be solved by the invention]
However, in the above-mentioned Patent Documents 1 and 2, the detection of a color binary image requires tracing the outline of a character or a diagram. there were. In Patent Document 3, since it is necessary to reduce the size of a block recursively, it takes a long time to perform the processing, and the calculation time varies greatly depending on the type of the original image. It was unsuitable for incorporation into equipment.
[0010]
The present invention has been made in view of the above circumstances, and efficiently separates an original image into a color binary image and a multicolor image with a small amount of calculation and without securing a large memory area. It is an object of the present invention to provide a method for compressing image data which can be compressed, and a system for performing the method.
[0011]
[Means for Solving the Problems]
In order to achieve the above object, claims 1 to 6 propose a compression processing method. That is, the method divides the image data into a plurality of pixel blocks, and executes an algorithm for identifying one-color blocks, two-color blocks, and multi-color blocks using the degree of dispersion of the pixel distribution information of each pixel block. It suggests a way to do it.
[0012]
That is, in the first aspect, a block processing for dividing an original image including a color image into a plurality of pixel blocks including a predetermined number of pixels, and a process of dividing pixels such as density and hue for each of the divided pixel blocks. For the distribution information, the degree of variance of the pixel block as a whole, the degree of variance of an area obtained by dividing the pixel block based on a threshold specified in advance, and a predetermined calculation process are performed to obtain a small-color pixel block with small variance. Image area discriminating processing for discriminating between a group and a multi-color pixel block group having a large variance; and generating color binary image data based on the small color pixel block group identified by the image area discriminating processing. Image data is compressed in a predetermined binary image compression format, while natural background image data is compressed based on the multicolor pixel block group identified by the image area identification processing. Produced, and an image data image area separation compression process for compressing the image data in a predetermined multilevel image compression format.
[0013]
Here, the degree of dispersion of the pixel block is obtained by using a variance value or a weighted average of the entire pixel block by a simple method, but is not limited to such.
[0014]
When dividing a pixel block based on a threshold value, the threshold value may be a predetermined value, an average value of the variance of the entire block, or may be obtained by a predetermined calculation. As one of the calculation methods, there is a method in which an arbitrary threshold value is changed, and the threshold value is fixed at a point where the variance ratio of each class exceeding and not exceeding the threshold value is maximized.
[0015]
In addition, the area divided based on the threshold value here means an area of pixel density and hue, but is not limited to this.
[0016]
Furthermore, in the image data image area separation compression processing, the natural background image data is obtained by filling a multicolor pixel block extracted from the original image with a portion corresponding to the small color pixel block with the background color of the small color pixel block. It is formed by interpolation. In this case, an image in which only the low-color pixel blocks are made transparent and an image in which only the multi-color pixel blocks are made transparent and the low-color pixel blocks are filled with the background color as an effective image are used as a layer structure. And may be formed in combination. Then, in the multi-valued image compression format, the natural background image data generated in this manner is compressed according to JPEG, wavelet, or the like after thinning out as necessary.
[0017]
On the other hand, the color binary image data is formed by interpolating a transparent block of a portion corresponding to a multi-color pixel block into a small-color pixel block. The color binary image referred to in the present invention means data in which different color characters can be displayed for each pixel block, and is different from a normal binary image composed of white and black. In this case, the multi-color pixel block portion may be made transparent and an image having only the small-color pixel block portion as an effective image may be formed to have a layer structure. The color binary image data generated in this manner is compressed in a binary image compression format by using a method such as MR or MMR.
[0018]
According to another aspect of the present invention, the image area discriminating process includes identifying a one-color block and a two-color block as a small-color pixel block group, and identifying a multi-color block as a multi-color pixel block group. The compression processing is performed. Such a method of claim 2 may be executed in place of the image area separation and compression processing of claim 1.
[0019]
According to a third aspect of the present invention, in the image area discriminating process, a histogram is created as pixel distribution information such as density and hue for a pixel block, and an arithmetic process is performed on the histogram to obtain a small color with a small variance. It is characterized in that it is distinguished into a pixel block and a multi-color pixel block having a large variance.
[0020]
In the image area discriminating process, in the discriminating process, for a pixel block determined as one color block, a change in a representative color in a pixel block adjacent in one direction to the pixel block is determined. In the gradation determination process, if the representative color is constant, the block is determined to be a one-color block. If not, the block is determined to be a multi-color block. From the above, while a pixel block determined to be a one-color block or a two-color block is regarded as a small-color pixel block, a pixel block determined to be a multi-color block is regarded as a multi-color pixel block, and pixel data image area separation compression processing is performed. It is characterized by performing.
[0021]
Such a method according to claim 4 may be executed in place of the image area separation and compression processing according to claim 1.
[0022]
In this case, even if the pixel block is a one-color block, if the representative color of the one-color block continuous in one direction changes, the natural background image is regarded as a multi-color block. Can display gradation. Here, the representative color may be determined based on the average value, frequency, and the like of the pixel blocks.
[0023]
According to a fifth aspect of the present invention, in the determination process, for a pixel block determined to be a two-color block, the pixel block is set as a target pixel block, and the same hue as the foreground color is determined in the outer contour of the pixel block. Further includes a contour discriminating process of discriminating whether or not there is a pixel forming the following pixel.If the result of the discriminating process indicates that a pixel having the same hue as the foreground color exists in the contour of the target pixel block by a predetermined value or more, A pixel block adjacent to the contour of the target pixel is regarded as a two-color block, and the pixel data image area separation compression processing is performed.
By adding such a contour discriminating process, when a pixel block adjacent to a two-color block includes a pixel having the same hue as the foreground color included in the two-color block, the image is compressed in a binary image compression format. Since the image data is compressed, a pixel block in which a part of a character, a line or the like protrudes is also processed as a two-color block. Therefore, a color image including character data and ruled lines can also be compressed without impairing features such as characters and ruled lines.
[0024]
Claims 6 to 10 simultaneously propose image data compression processing systems for implementing the methods of claims 1 to 5.
[0025]
Here, in claim 6, a block processing means for dividing the original image data including the color image into a plurality of pixel blocks including a predetermined number of pixels, and at least a density and a hue are given to each of the divided pixel blocks. A histogram is created for attribute information such as, for example, and an image is identified based on the created histogram into a small-color pixel block including only the foreground and / or the back-color and a multi-color pixel block including other multicolors. Based on the area identification processing means and the small-color pixel blocks identified by the image area identification processing means, color binary image data is generated, and the generated image data is compressed in a predetermined binary image compression format. On the other hand, while compressing, natural background image data is generated based on the multicolor pixel block identified by the image area identification processing means, and the generated original image data is generated. And an image data image area separation compression means for compressing the pixel data at a constant multi-valued image compression format.
[0026]
Such a compression processing system for image data can compress a color image including characters, line segments, and ruled lines by using an image area separation method regardless of the method of the first aspect.
[0027]
According to claim 7, the image area discriminating processing means determines that a pixel block including only one of the foreground color and the back color is a one-color block, and determines a pixel block including only both of them. It can be determined as a two-color block, otherwise it can be determined as a multi-color block, and based on the results of these determinations, image data image area separation compression processing can be performed.
[0028]
In the eighth aspect, the gradation determination processing described in the fourth aspect can be performed, and in the ninth aspect, the contour determination processing can be performed. In the tenth aspect, the image area identification processing described in the first aspect can be performed.
[0029]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described. FIG. 1 is a diagram illustrating the basic processing of the present invention.
[0030]
According to the method of the present invention, a block processing (1) for dividing a color original image generated by the electronic processing into a plurality of pixel blocks including a predetermined number of pixels (for example, 256 dots) is performed. Next, for each of the divided pixel blocks, a histogram of attribute information including at least density and hue information is created, a variance value is obtained from the created histogram, and a variance is calculated by comparing with a predetermined threshold. Is performed to distinguish between a small-color pixel block group having a small pixel number and a multi-color pixel block group having a large variance (2).
[0031]
Then, binary image data is generated from the small-color pixel block group identified by the image area identification process (2) (3), and the image data is compressed in a predetermined binary image compression format (4).
[0032]
On the other hand, a background color is added to the multicolor pixel block group identified by the image area identification process (2) to generate natural image data (4), and the natural image data is converted into pixel data in a predetermined multivalued image compression format. Is compressed (5).
[0033]
Here, a series of processing in steps (3) to (6) is referred to as pixel data image area separation / compression processing (6).
[0034]
Next, each of the block processing (1) and the image area identification processing (2) will be described in detail with reference to FIG.
[0035]
That is, it is assumed that there is an original image shown in FIG. In this original image, a character image such as “HG” or “ABC” or a color binary image such as a straight line is drawn for a multicolor image.
[0036]
In the blocking process (1), such an original image is first divided into a plurality of blocks including a predetermined number of pixels. FIG. 2B shows the result of division into a plurality of blocks.
[0037]
An image area discriminating process (2) is performed on the original image subjected to the blocking process as shown in FIG. That is, for each pixel in each block, a histogram is created for attribute information (here, density or hue) of the pixel.
[0038]
Here, assuming that the block 10 is a multicolor image, the histogram thereof is uniformly distributed from a small density to a large density (see FIG. 5A). If the block 11 is a one-color image, the density of the histogram is somewhat constant (see FIG. 5C). Further, assuming that the block 12 is a two-color image, two histograms whose density is constant to some extent appear (see FIG. 5B).
[0039]
In the image area identification process (2), such a histogram is extracted. That is, a variance value is obtained for each block, and the processing described later is performed to discriminate between a small-color pixel block having a small variance and a multicolor pixel block having a large variance. As a result, the blocks 11 and 12 are identified as small-color pixel blocks, and the block 10 is identified as a multi-color pixel block.
[0040]
In the image area discriminating process (2), the distribution degree of each block as a whole may be obtained without creating a histogram.
[0041]
The flowchart in FIG. 4 shows the basic operation of the image area identification processing.
[0042]
First, an image is divided into a plurality of pixel blocks in the blocking process (1) (S1). Thereafter, the image area identification process (2) starts.
[0043]
In this image area identification processing, pixel distribution information on the density and hue of each block is extracted, and the density and hue variance σ1 of the entire block are obtained (S2, 3).
[0044]
If the degree of the variance σ1 of the entire block is smaller than a preset constant K (S4), the block is identified as a one-color block (see FIG. 5C) (S9).
In the example shown in the figure, in step S9, it is not immediately determined as a one-color block but is treated as a one-color block candidate. However, this is for performing two-color block peripheral processing and gradation processing, which will be described later. If not, it is determined as a one-color block.
[0045]
On the other hand, if either the density or the degree of the hue variance σ1 is sufficiently larger than the constant K, it is identified as a two-color block (see FIG. 5B) or a multi-color block (see FIG. 5A). However, for this determination, a determination method as shown in FIG. 6 is used.
[0046]
In FIG. 6, an appropriate threshold value is determined for the density and the hue, and the threshold value is fixed so that the variance ratio between the foreground color and the back color is maximized. It is determined whether the block is a multicolor block.
[0047]
In other words, when the degree of variance of the divided area is sufficiently smaller than the variance of the entire block in the pixel distribution information obtained by dividing the pixel block as described above, the histogram shown in FIG. Since it is estimated, it is identified as a two-color block. If the degree of these variances is not sufficiently smaller than the variance of the entire block, it is presumed that a histogram as shown in FIG. Is identified.
[0048]
More specifically, for the divided pixel block, the variances σf and σb of the classes of the foreground and background colors are obtained (S6), and the variances σf and σb of the density and the hue are set to be smaller than the threshold σ1 of the entire block. If it is sufficiently small (S7), it is identified as a two-color block (S8), but if not (S7), it is determined as a multi-color block (S5).
[0049]
Here, the determination of such a two-color block or a multi-color block can be performed, for example, in the following three ways.
[0050]
That is, simply, satisfies σf <σ1 × constant and σb <σ1 × constant, then (σf + σb) {2 <σ1 × constant, and furthermore, (σf × Nf + σb × Nb)} Whether (Nf + Nb) <σ1 × constant is satisfied can be determined. Here, Nf means the number of pixels in the class belonging to σf, and Nb means the number of pixels in the class belonging to σb.
[0051]
If the processing of steps S2 to S9 is performed for all blocks (S10 and S11), all the original images can be classified into multicolor, two-color, and one-color blocks.
[0052]
FIG. 7 is a flowchart illustrating the two-color block peripheral processing. Here, it is determined whether or not compression processing should be performed by regarding blocks adjacent to the block identified as one color block in the above-described image area identification processing as two-color blocks.
[0053]
In this process, the outer contour frame in the block determined to be a two-color block includes the same foreground pixel as that of the two-color block (hereinafter referred to as a pixel block of interest), and the number of pixels is equal to the threshold nth If so (S20, S21), the adjacent block is stored as a two-color block (S22). On the other hand, if not, it is stored as the initially set block, that is, a one-color block or a two-color block (S23), and the image area compression / separation processing (3) is performed.
[0054]
FIG. 7A is a diagram for explaining this processing. Here, assuming that the two-color blocks B0 and B0 'are the target pixel blocks, are the outer contour frames S1 to S4 inside the two-color blocks B0 and B0' have the same pixels as the foreground color of the two-color blocks B0 and B0 '? See if you can. Then, it is determined whether the number of pixels is equal to or larger than a threshold value nth. Then, each of the blocks B1 and B1 and the block B1 'on which a part of a character or a straight line is applied is treated as a two-color block.
[0055]
The two-color block peripheral processing is not limited to this method. For the adjacent blocks adjacent to the two-color block, the same pixel as the representative color (foreground color) of the two-color block (hereinafter, referred to as a target pixel block) is used. It is also possible to check whether there is a certain amount or more, and if so, a method of storing the block as a two-color block.
[0056]
FIG. 8 is a flowchart illustrating the gradation determination processing. Here, since a series of blocks recognized as one color block by the above-described image area discriminating process may form a gradation, a block in which the representative color (foreground color) is not constant is a multicolor block. Identify as
[0057]
That is, if one color block is continuous and the representative color has changed (S30), these continuous blocks are stored as multicolor blocks (S31). On the other hand, if not, it is stored as the original one-color block (S32), and the image area is compressed.
[0058]
In this way, when a continuous block determined to be a one-color block is complemented by the above-described background color, it can be prevented from being filled with the same color and lost in gradation.
[0059]
FIG. 9 is a diagram illustrating a configuration example of a main part of a compression processing system capable of realizing the above-described compression processing method. The image processing apparatus includes a block processing unit A for performing a block processing (1), and an image processing unit C for performing an image area identification processing (2) and performing a two-color block peripheral processing and a gradation determination processing. Processing means B includes image area separation / compression processing means D for performing image area separation / compression processing (3).
[0060]
【The invention's effect】
As can be understood from the above description, according to the method of the present invention of claims 1 to 4, a color original image can be separated into a color binary image and a natural background image by a simple algorithm. In addition, data can be compressed at high speed without securing a large memory area.
[0061]
In particular, according to the second aspect, the characteristics of the histogram of the pixel block are analyzed by a simple method and are discriminated as one-color, two-color, and multi-color blocks. Therefore, a small amount of calculation and a large memory area can be ensured. And can compress data at high speed.
[0062]
According to the fourth aspect, even when a color original image has a gradation, it can be identified by a simple method and can be compressed without deteriorating its characteristics.
Furthermore, according to the present invention, even when characters, lines, ruled lines, and the like are included in the color original image, they can be efficiently compressed without missing, blurring, or deterioration.
[0063]
According to the sixth aspect of the present invention, the method according to the first aspect includes the means for performing the blocking processing, the means for performing the image area separation processing, and the means for performing the pixel data compression processing according to the first aspect. Can be realized. Further, in this system, a color binary image and a natural background image can be generated and compressed by identifying the foreground color, the background color, and these distribution patterns in the pixel block. And can be compressed.
[0064]
According to the system of the present invention, the characteristics of the histogram of the pixel block can be analyzed by a simple method, and can be discriminated as one-color, two-color, and multi-color blocks and compressed.
[0065]
Further, according to the system of the present invention, when a color original image has a gradation, it can be efficiently compressed without deteriorating its characteristics. According to the ninth aspect of the present invention, even when characters, lines, ruled lines, and the like are included in a color original image, they can be efficiently compressed without missing, blurred, or degraded. Further, according to the system of the tenth aspect of the present invention, the image area identifying process of the first aspect can be easily performed.
[Brief description of the drawings]
FIGS. 1A and 1B are diagrams for explaining a basic process of the present invention. FIGS. 2A and 2B are explanatory diagrams of a blocking process and an image area discriminating process in the present invention. FIGS. FIG. 4B is an explanatory diagram of natural image data and extracted image data generated in the present invention. FIG. 4 is a flowchart showing a basic operation of image area discriminating processing. FIGS. 5A, 5B, and 5C. Is a diagram showing a histogram of each of a multi-color block, a two-color block, and a one-color block. FIGS. 6A and 6B are explanatory diagrams of a multi-color pixel block and a two-color pixel block. FIG. 7A is a flowchart for explaining a basic operation of a color block peripheral process. FIG. 7A is an explanatory diagram of a peripheral process for a two-color block. FIG. 8 is a flowchart for explaining a basic operation of a gradation process. Diagrams [Description of symbols]
DESCRIPTION OF SYMBOLS 1 ... Blocking process 2 ... Image area discriminating processing 3-6 ... Image area separation / compression processing A ... Blocking means B ... Image area discriminating processing means C ... Arithmetic processing Means D: image area separation / compression processing means

Claims (10)

Blocking processing of dividing an original image including a color image into a plurality of pixel blocks including a predetermined number of pixels,
For each of the divided pixel blocks, regarding the distribution information of the pixels such as the density and the hue, the degree of dispersion of the pixel block as a whole and the degree of dispersion of an area obtained by dividing the pixel block based on a predetermined threshold value are described. Image area discriminating processing for obtaining and performing predetermined arithmetic processing to identify a small-color pixel block group having a small variance and a multi-color pixel block group having a large variance;
Color binary image data is generated based on the small-color pixel block group identified by the image area identification processing, and the image data is compressed in a predetermined binary image compression format. Image data compression processing comprising: generating natural background image data based on the identified multi-color pixel block group; and compressing the image data in a predetermined multi-valued image compression format. Method. In claim 1,
The image area discriminating process includes:
The one-color block and the two-color block are identified as the small-color pixel block group, and the multi-color block is identified as the multi-color pixel block group, and the pixel data image area separation and compression process is performed. Image data compression processing method. In claim 1,
In the image area discrimination process, a histogram is created for pixel blocks as pixel distribution information such as density and hue, and arithmetic processing is performed on the histogram. A compression processing method of image data, characterized in that it is identified as a large multi-color pixel block. In claim 2 or 3,
The image area discriminating process includes:
The above-described determination processing further includes, for the pixel block determined as one color block, a gradation determination processing of determining a change in a representative color in a pixel block adjacent in one direction to the pixel block,
In the gradation determination processing, if the representative color is constant, it is determined to be a one-color block. If not, it is determined to be a multi-color block. A pixel block determined as a color block is regarded as a small-color pixel block, while a pixel block determined as a multi-color block is regarded as a multi-color pixel block, and the pixel data image area separation and compression process is performed. Image data compression processing method. In any one of claims 2 to 4,
The discrimination process is as follows:
For a pixel block determined to be a two-color block, whether or not there is a pixel having the same hue as the foreground color in the target pixel block in the outer contour of the pixel block as the target pixel block Further includes a contour determination process of determining
As a result of the discrimination processing, if a pixel having the same hue as the foreground color is present in the outline of the target pixel block at a predetermined value or more, a pixel block adjacent to the outline of the target pixel is regarded as a two-color block. A method for compressing image data, comprising performing pixel data image area separation compression processing. Blocking processing means for dividing original image data including a color image into a plurality of pixel blocks including a predetermined number of pixels,
Image area discriminating processing means for discriminating between a small-color pixel block group and a multi-color pixel block group including other multi-colors based on pixel distribution information such as density and hue for each of the divided pixel blocks. When,
Color binary image data is generated based on the small-color pixel block group identified by the image area identification processing means, and the generated image data is compressed in a predetermined binary image compression format. An image data image area for generating natural background image data based on the multicolor pixel block group identified by the area identification processing means and compressing the generated original image data in a predetermined multi-valued image compression format A compression processing system for image data, comprising: a separation compression processing unit. In claim 6,
The image area identification processing means,
When it is determined that only one of the foreground color and the back color is included in the pixel block, the pixel block is determined to be a one-color block. Otherwise, the pixel block is further determined to be a predetermined color block. Using the threshold of, the image is divided into regions on the foreground color side and the background color side, and the degree of dispersion of each is obtained.
The next determination processing, that is, further includes an arithmetic processing means for determining a two-color block if the degree of dispersion of the divided area is sufficiently smaller than a predetermined threshold, and otherwise determining a multi-color block. ,
The image data image area separating / compressing means regards the pixel block determined as a one-color block or a two-color block as a small-color pixel block group based on the result of the determination processing, and determines a pixel block determined as a multi-color block as a small-color pixel block group. An image data compression processing system, wherein the image data image area separation compression processing is performed by regarding the image data as a group of multicolor pixel blocks. In claim 6 or 7,
The arithmetic processing means includes:
An image data compression processing system that further executes the gradation determination processing according to claim 3. In any one of claims 7 to 8,
5. A compression processing system for image data, wherein the arithmetic processing means further executes the contour determination processing according to claim 4. In any one of claims 7 to 9,
The image area identification processing means creates a histogram based on pixel distribution information such as density and hue for the plurality of divided pixel blocks, and calculates the degree of dispersion of the entire pixel block and the dispersion of the divided areas. A predetermined arithmetic processing is executed after obtaining the degree of the image data, and a compression processing system for the image data identified as the small-color pixel block group and the multi-color pixel block group.

Images