JP2005223455A - Image processing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of enhancing a compression rate by separating a character part and an image pattern part from image data and applying prescribed processing to them. <P>SOLUTION: The image processing apparatus demarcates a region including a continuous region of valid pixels in image data on the basis of the image data of a processing object, generates at least one of region information denoting regions, integrates a plurality of regions or divides one region from a plurality of the regions according to a prescribed integration or division rule on the basis of the property of each region and a positional relation among the regions demarcated in relation to each of the generated region information items. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus that separates a character part and a picture part from image data and performs predetermined processing.

Raster image data (hereinafter simply referred to as “image data” unless otherwise distinguished) includes many images with different properties, such as character (text) portions and natural image portions (pattern portions). Elements can be included. Due to the difference in properties of these image elements, it is often preferable to perform different image processing for each image element, for example, compression by a different method is suitable for compression processing. Therefore, conventionally, image processing called so-called T / I separation has been researched and developed, and a processing method for separating a character part and a picture part and performing compression processing suitable for each is known (Patent Document 1, Patent Document 1). 2).
JP 2002-369011 A JP 2003-18413 A JP 2002-175532 A

  However, in the conventional method, a continuous area (pixel block) of significant pixels (pixels set to a color different from the background color such as paper color and constituting an image such as a character or a picture) is extracted, and its circumscribed rectangle is extracted. Since the pixel block included in the circumscribed rectangle is classified as a character or a picture, most of the included pixels are pixels of the background color, or the area is too fragmented. On the other hand, the compression rate may be lowered.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide an image processing apparatus capable of improving the compression rate.

  The present invention for solving the problems of the conventional example described above is an image processing device, and defines an area including a continuous area of significant pixels in the image data based on the image data to be processed. A plurality of means for generating at least one area information representing the area, and a predetermined integration rule based on the property of each area defined by each of the generated area information or the positional relationship between the areas. And a means for generating an integrated area obtained by integrating these areas.

  Further, the present invention for solving the problems of the conventional example is an image processing apparatus, and based on image data to be processed, an area including a continuous area of significant pixels is included in the image data. The region is divided according to a predetermined division rule based on the property of each region defined for each of the generated region information and means for generating at least one region information that defines and represents the region. And a means for generating a plurality of divided regions.

  Furthermore, the present invention for solving the above-described problems of the conventional example is an image processing device, and defines an area including a continuous area of significant pixels in the image data based on the image data to be processed. A plurality of regions by means of generating at least one region information representing the region and a predetermined integration rule based on the property of each region defined by each region information or the positional relationship between the regions. For generating new area information, and for each of the area information, a means for dividing the area according to a predetermined division rule based on the property of each area defined by the area information and generating new area information It is characterized by including.

  In this case, re-integration may be performed for a plurality of regions obtained by the above division, which further satisfy the integration rule, and those that further satisfy the division rule for the region obtained by integration. May be subdivided.

  In these, the integration rule may include a condition regarding the size of the region indicated by the region information or the area of the overlapping portion of the region. The division rule may include a condition relating to a ratio of the area occupied by the significant pixels in the region indicated by the region information. Note that the result of the integration or division processing is provided to a predetermined image compression processing or the like.

  An image processing method according to an aspect of the present invention uses a computer to define a region including a continuous region of significant pixels based on image data to be processed, and represents the region At least one piece of information is generated, and for each of the region information, a plurality of regions are integrated according to a predetermined integration rule based on the property of each region defined or the positional relationship between the regions, and new region information is obtained. At least one of the step of generating or the step of dividing the region and generating new region information according to a predetermined division rule based on the property of each region defined by each of the region information is executed. It is characterized by that.

  Furthermore, a program according to another aspect of the present invention provides a computer with region information that defines a region including a continuous region of significant pixels in the image data based on the image data to be processed, and represents the region. For each of the region information, a plurality of regions are integrated according to a predetermined integration rule based on the property of each region defined by the region or the positional relationship between the regions, and new region information is created. For each of the area information, or a process of dividing the area and generating new area information according to a predetermined division rule based on the property of each area defined by the area information. And a procedure for performing the operation.

  As shown in FIG. 1, the image processing apparatus according to the embodiment of the present invention includes a control unit 11, a storage unit 12, an image input unit 13, and an image output unit 14. The control unit 11 operates in accordance with a program stored in the storage unit 12 and performs each image processing described later. The contents of this image processing will be described in detail later.

  The storage unit 12 holds a program executed by the control unit 11. The storage unit 12 also operates as a work memory that stores various data generated during the process of the control unit 11. Specifically, the storage unit 12 can be implemented as a computer-readable recording medium and a device that writes data to or reads data from the recording medium (for example, a hard disk device or a memory device).

  The image input unit 13 is, for example, a scanner, and outputs image data obtained by optically reading a document to the control unit 11. Here, it is assumed that the value of each pixel is expressed in an RGB (red, green, blue) color space in the image data output from the image input unit 13. The image output unit 14 outputs image data in accordance with an instruction input from the control unit 11. For example, the image output unit 14 outputs the image data to an image forming unit (printer or the like) or transmits it to an external device via a network. Is what you do.

  Next, the content of the process of the control part 11 is demonstrated. As functionally shown in FIG. 2, the control unit 11 according to the present embodiment sets the image data input from the image input unit 13 as a processing target, and performs predetermined preprocessing on the image data that is the processing target. A pre-processing unit 21 for performing a character line drawing extraction processing unit 23 for extracting a character line drawing, a pattern extraction processing unit 24, a character line drawing processing unit 25, a design part processing unit 26, a compression processing unit 27, and a format process. Part 28.

  Hereinafter, each of these parts will be described in detail.

[Pre-processing section]
In the preprocessing unit 21, the value of each pixel of the image data (processing target image data) input from the image input unit 13 is converted from RGB to YCbCr (a value composed of luminance and color difference). Specifically, the conversion can be performed using the following equation (1). Here, the value of each component of RGB is assumed to be a value from 0x00 (“0x” indicates a hexadecimal number) to 0xFF. Further, the pre-processing unit 21 may correct the gradation of each pixel value based on the luminance and saturation of the background area. However, the gradation correction process is not always necessary.

[Character line drawing extraction processing section]
The character / line drawing extraction processing unit 23 performs processing for extracting a character / line drawing part from the image data expressed in the YCbCr color space output from the preprocessing unit 21. For this process, for example, the process disclosed in Patent Document 3 can be used. That is, the character / line drawing extraction processing unit 23 demarcates a plurality of processing regions on the image output by the previous output unit 21 while allowing overlap. Then, a moving average is calculated for each pixel value of the pixels included in each processing region, and each pixel in the processing region is determined to be white or black by comparing the moving average with the value of each pixel. As a result, the character / line drawing extraction processing unit 23 binarizes the image data based on the threshold value for each region determined by the moving average. Then, a continuous area of black pixels included in the binarized image data is found, and the size of the rectangle circumscribing the continuous area is within a predetermined font size range (for example, from a size corresponding to 9 points to a size corresponding to 24 points). In the case where a condition such as the above is satisfied, the continuous area of the black pixels is determined as a character. Similarly, a line drawing is extracted based on information on the width of a continuous area of black pixels.

  The character / line drawing extraction processing unit 23 stores in the storage unit 12 coordinate information of individual characters and rectangles (character / line drawing circumscribed rectangles) surrounding them.

[Pattern extraction processing unit]
The pattern extraction processing unit 24 removes significant pixels in the character / line drawing portion extracted by the character / line drawing extraction processing unit 23 from the image data expressed in the YCbCr color space output from the preprocessing unit 21. A significant pixel is a pixel that represents a color different from the background color, such as the paper color. Specifically, the background color is determined using the mode value of the pixel value group near the four corners of the processing target image data, and the distance in the color space between the background color and the color represented by the pixel value A pixel having a value greater than or equal to a predetermined threshold is defined as a significant pixel.

  The pattern extraction processing unit 24 binarizes the image data after removing the significant pixels of the character / line image portion, and finds a continuous region of black pixels included in the binarized image data. Further, a rectangle circumscribing the continuous area is generated, and coordinate information of the rectangle is stored in the storage unit 12 as pattern rectangle information.

[Character line drawing processing]
The character / line drawing processing unit 25 extracts information on the circumscribing rectangle of the character / line drawing from the storage unit 12, extracts significant pixels in each circumscribing rectangle of the character / line drawing from the image data expressed in the YCbCr color space output from the preprocessing unit 21. Compress. Specifically, for each character / line drawing circumscribing rectangle, a representative pixel value is determined from the value of a significant pixel inside the character / line drawing circumscribing rectangle, and character line drawing circumscribing rectangles related to substantially the same representative pixel value are associated with each other. Then, grouped data is generated by grouping significant pixels inside the character line drawing circumscribed rectangles associated with each other. Further, the grouped data is binarized to generate a set of representative pixel values and binarized grouped data. For example, the pixels constituting the red character are associated with each other to form grouped data, the red portion is set to black, the other is set to white, and the representative pixel value representing “red” Will be generated. Further, the binarized data may be run-length compressed (compressed by MMR or the like).

[Pattern processing unit]
The pattern part processing unit 26 reads the pattern rectangle information from the storage unit 12. Then, it is determined whether or not each pattern rectangle information can be integrated with other pattern rectangle information. This determination is made according to a predetermined integration rule.

  This integration rule can be defined as a condition regarding the property (for example, size) of a rectangular area defined by the design rectangle information and the positional relationship between a plurality of the rectangular areas. Specifically, the pattern part processing unit 26 executes an integration process as shown in FIG.

  In this integration process, first, for each pattern rectangle information, the center of gravity of significant pixels in the rectangular area defined by the information is calculated (S1). Here, a well-known process can be used as the center of gravity calculation method. Next, the area ratio of the corresponding rectangular area is calculated for each picture rectangle information (S2). Here, the area ratio is a value obtained by dividing the area of the rectangular area by the area of the entire image data to be processed.

  Then, the pattern portion processing unit 26 selects, as the integration candidate, the pattern rectangle information whose area ratio is less than a predetermined value (for example, 0.5) (S3). Thereby, for example, a large-sized picture is not integrated with another picture. Here, although it is based on the size of the rectangular area, a value obtained by dividing the number of significant pixels included in the rectangle by the number of pixels of the entire image data may be calculated as the area ratio.

  The pattern part processing unit 26 selects different pairs of pattern rectangle information from the pattern rectangle information selected as integration candidates in the process S3 (S4). Here, one unselected combination is selected from among the combinations for extracting two pieces of picture rectangle information from the integration candidates. Then, based on the size of each piece of rectangular information relating to this pair and the distance between the centers of gravity, it is determined whether or not the rectangular regions relating to this pair are to be integrated (S5).

であるとき、統合するものと判断して、このペアに係る矩形領域を統合する（Ｓ６）。なお、ここでＨ１，Ｖ１は上記ペアに係る一方の矩形領域の横幅と縦幅を表し、Ｈ２，Ｖ２は他方の矩形領域の横幅と縦幅を表す。 Specifically, the distance r between the centers of gravity is

If it is, it is determined to be integrated, and the rectangular regions related to this pair are integrated (S6). Here, H1 and V1 represent the horizontal width and vertical width of one rectangular area in the pair, and H2 and V2 represent the horizontal width and vertical width of the other rectangular area.

  Specifically, one rectangular area is defined by its upper left coordinates (xlt1, ylt1) and lower right coordinates (xrb1, yrb1), and the other rectangular area is its upper left coordinates (xlt2, ylt2) and lower right coordinates. When (xrb2, yrb2) is included, the integrated rectangular region has, for example, upper left coordinates (min [xlt1, xlt2], min [ylt1, ylt2]) and lower right coordinates (max [xrb1, xrb2], max [yrb1, yrb2]). Here, min [a, b] represents the smaller one of the values a and b. Thus, when relatively small areas are close to each other, the areas are integrated.

  In step S6, each piece of the design rectangle information corresponding to the integration-related rectangle area is replaced with the newly calculated coordinate information, and a new post-integration defined by the newly generated design rectangle information. Calculate the center of gravity of a rectangular area. Then, it is checked whether there is an unselected pair (S7). That is, it is checked whether there is a combination that has not yet been selected among combinations that extract two different pattern rectangle information from the integration candidates in the integration candidates including the new pattern rectangle information after integration. If there is an unselected pair, the process returns to step S4 to select one of the unselected pairs and repeat the process. If there is no unselected pair, the integration process is terminated.

  On the other hand, when it is determined in the process S5 that they are not integrated, the pattern portion processing unit 26 determines whether or not the rectangular areas related to the selected pair intersect each other (S8). Here, “intersection” refers to the case where the other rectangular area is included in one rectangular area as shown in FIG. 4A, and the case where each rectangular area overlaps as shown in FIG. 4B. If you are.

  In the process S8, when it is determined that they intersect each other, the pattern part processing unit 26 calculates the area of the intersecting part and calculates the area of each rectangular area (S9). Then, whether or not the area of the intersecting portion is a predetermined ratio (for example, 50%) or more with respect to the area of any of the rectangular regions, that is, a portion of the one rectangular region more than a predetermined ratio, It is checked whether or not the other rectangular area is included (S10). If this condition is met, it is determined that the pair of the selected rectangular areas is to be integrated, and the process proceeds to S6. If the condition is not met in process S10, the process proceeds to process S7.

  Further, the pattern portion processing unit 26 determines whether or not each rectangular area after the integration process can be divided. This determination is made according to a predetermined division rule. Here, the division rule is based on, for example, the property of the rectangular area. Specifically, the ratio between the area of the rectangular area and the area occupied by significant pixels in the rectangular area (ratio of the area occupied by significant pixels) ) Is less than or equal to a predetermined value, it is determined that division is possible. That is, when there are many insignificant pixels, it is determined that the pixels can be divided.

  Here, an example of the division method will be specifically described. The pattern portion processing unit 26 counted the number of significant pixels for each line in the horizontal direction (horizontal projection histogram) and the number of significant pixels for each line in the vertical direction for the rectangular area to be divided. The result (vertical projection histogram) is generated (FIG. 5). Then, among the vertex positions of the polygons of each histogram, the change between the adjacent vertices is larger than a predetermined threshold (points P and Q in FIG. 5), and the adjacent vertex is “0” (no significant pixel). (Points R, S, T, U in FIG. 5) are respectively selected as pole points.

  Here, for example, the extreme positions of the vertical projection histogram are x1, x2... Xn (where x1 <x2 <... Xn), and the extreme positions of the horizontal projection histogram are y1, y2. ... <yn), x1, x2 ... xn-1 are extracted as x components of the division start point position coordinates, and y1, y2 ... yn-1 are similarly extracted as y components.

  Then, a combination of the extracted components is generated. That is, (x1, y1), (x1, y2), ... (x1, yn-1), (x2, y1) ... (xn-1, yn-1) are generated. This set becomes the division start point position coordinates.

  Next, with respect to the generated combination, division end position coordinates corresponding to the division start point position coordinates are generated using the coordinates of the extreme points adjacent to the respective components related to the combination. That is, for (x1, y1), the positions of the extreme points adjacent to each component are x2, y2, and therefore the division end position coordinates corresponding to (x1, y1) are determined as (x2, y2). Then, coordinate information defining a divided rectangle is generated by a set of the division start point position coordinates and the division end point position coordinates.

  Specifically, in the example of FIG. 5, (x1, y1)-(x2, y2), (x1, y2)-(x2, y3), (x2, y1)-(x3, y2) as coordinate information , (X2, y2)-(x3, y3) are generated.

  The pattern part processing unit 26 determines whether or not a significant pixel is included in the rectangle defined by the generated coordinate information. If the significant pixel is not included, the coordinate information is removed and the significant pixel is removed. Is stored in the storage unit 12 as a new picture rectangular area. Note that the information of the pattern rectangular area before the division is removed from the storage unit 12.

  Thereby, a cluster of significant pixels is divided into a line segment drawn in the vertical direction from the extreme points of the vertical projection histogram and a line segment drawn in the horizontal direction from the local points of the horizontal projection histogram. Specifically, in the example of FIG. 5, since there is no significant pixel in the rectangle defined by (x1, y2) − (x2, y3) among the generated coordinate information, this coordinate information is removed, It is divided into rectangles (regions A to C in FIG. 5) defined by the other three coordinate information.

  In this embodiment, the integration process and the division process are continuously performed, so that the area where the ratio of significant pixels becomes too small due to the integration is subdivided into areas with higher compression ratios. It will be possible (FIG. 6 (a)). Further, here, the division process is performed after the integration, but as shown in FIG. 6B, the integration process may be performed subsequently after the division process. Also in this case, regions that have been divided too much (regions A and B in FIG. 6B) can be reintegrated to form a region group with a higher compression rate.

  Furthermore, although the integration process and the division process are performed here, only one of them may be performed.

[surface treatment]
Further, the pattern portion processing unit 26 uses the pattern rectangle information after the division / integration and adjustment to remove significant pixels in the region defined by the pattern rectangle information from the image data to be processed. Extract the ground part. Then, based on the value of the pixel included in the background portion, the number of colors of the background portion is calculated. This calculation of the number of colors is obtained, for example, by counting the number of peaks having a predetermined frequency or more from a histogram of pixel values. Then, the compression processing unit 27 checks whether or not the calculated number of colors is one, and if it is one color, information on one representative color (for example, each pixel) determined from the pixel value in the background image data. Is generated as background plane data, and stored in the storage unit 12. At this time, when the distance between the representative color and white in the color space is less than a predetermined threshold, the color of the ground plane data may be “white”. In this case, when the image data generated by the image processing apparatus of the present embodiment is to be decoded on the side where there is no instruction (for example, when PDF or the like is used), The base plane data may not be generated.

[Compression processing]
The compression processing unit 27 reads the design rectangle information (the design rectangle information adjusted by the integration / division processing) stored in the storage unit 12 and is an area surrounded by each design rectangle information in the processing target image data. The partial image data is compressed by a method such as JPEG (Joint Picture Experts Group). Thereby, compressed data is generated corresponding to each picture rectangle information.

  Further, the compression processing unit 27 removes significant pixels of the character / line drawing portion extracted by the character / line drawing extraction processing unit 23 from the image data expressed in the YCbCr color space output from the preprocessing unit 21. Data or pattern image data obtained by replacing the removed portion with the average value of the peripheral pixels is compressed by a method such as JPEG.

  Then, the compression processing unit 27 compares the sum of the data sizes of the compressed data generated corresponding to each design rectangle information with the size of the data generated by compressing the design image data as it is, whichever is smaller Select. For example, when the size of the data generated by compressing the design image data as it is is small, the data generated by compressing the design image data as it is is selected and stored in the storage unit 12 as design plane data.

  If the total data size of the compressed data generated corresponding to each picture rectangle information is smaller, each compressed data generated corresponding to each picture rectangle information is stored in the storage unit 12 as picture plane data. Stored. At this time, since the background portion is removed, the background plane data (if generated) is included in the design plane data and stored in the storage unit 12.

  In this example, the size is compared by actually compressing each of them. However, the number of the generated pattern rectangle information (the number of divisions of the pattern portion) is multiplied by the size of the header information added in the compression process. If the overhead amount is larger than the size determination threshold value by comparing the obtained value (overhead amount) with a predetermined size determination threshold value, the image data expressed in the YCbCr color space output by the preprocessing unit 21 Then, the pattern image data obtained by removing the significant pixels of the character / line drawing part extracted by the character / line drawing extraction processing unit 23 is compressed as it is by a method such as JPEG to generate the pattern plane data, and stored in the storage unit 12. It may be stored.

  When the amount of overhead is smaller than the size determination threshold value, the design rectangle information stored in the storage unit 12 (design rectangle information adjusted by the integration / division process) is read out, and the image data to be processed The partial image data in the region surrounded by each picture rectangle information is compressed by a method such as JPEG, and each compressed data generated corresponding to each picture rectangle information and the base plane data (if generated) Are stored in the storage unit 12 as picture plane data.

  Here, the size of the header information may be approximate when the header information has a variable length. For example, in the case of JPEG, the thumbnail is added to 630 bytes, so it may be estimated that it is 650 bytes.

[Format processing section]
The format processing unit 28 couples the character / line drawing plane data, the pattern plane data, and the coordinate information of the pattern part circumscribed rectangle stored in the storage unit 12, and generates a series of data including these.

  Specifically, the series of data can be PDF (Portable Document Format) data. That is, an instruction to decompress the base plane data (if included) to generate a bitmap, an instruction to decompress the design plane data to generate a bitmap, and each character included in the character / line drawing plane data The PDF data includes an instruction to draw the pixels of the line drawing and an instruction to combine them.

  Here, an instruction to draw each character or line drawing included in the character / line drawing plane data is drawn for the pixels in the binarized and reversibly compressed group by decompressing and drawing, and representatives related to the group. This is an instruction to set the pixel (for example, black pixel) portion after the drawing as the representative color based on the color information. In this case, the white pixel portion is a pixel for which no pixel value is set.

  The format processing unit 28 stores the generated PDF data in the storage unit 12 or outputs the PDF data to the image output unit 14 to send it to an external device.

  Note that when restoring the original image data from the image data generated in this way, the following processing is performed. That is, first, ground plane data is extracted. Then, the base plane data is expanded to generate base image data.

  Specifically, when the background plane data includes the representative color of the background portion, image data of the size filled with the representative color of the background portion is generated based on the size information of the original image data.

  If the base plane data includes compressed data, the compression is expanded to generate image data of the base portion. At this time, when the size of the ground plane data is reduced, a process of enlarging the size to the same size as the original image data is performed.

  Next, the image plane data is extracted and decompressed, and the image data overwritten with the decompressed image data is displayed at the corresponding position on the background image data in accordance with the coordinate information of the pattern part circumscribed rectangle relating to each pattern plane data. Generate.

  Further, the character / line drawing plane data is extracted and decompressed to generate a character / line drawing bit map. Then, the character line drawing bitmap for each group is overwritten at the designated position on the generated image data.

  According to the present embodiment, when compressing a pattern part, the division mode of the pattern part is set to a state suitable for compression processing (the number of significant pixels in the divided area is increased as much as possible and is not excessively divided). So that the compression rate can be improved.

1 is a configuration block diagram illustrating an example of an image processing apparatus according to an embodiment of the present invention. It is a functional block diagram showing the processing content performed by the control part of the image processing apparatus which concerns on embodiment of this invention. FIG. 10 is a flowchart illustrating an operation example of the pattern portion processing unit 26. It is explanatory drawing showing the example of a cross | intersection of an area | region. It is explanatory drawing showing 1 process of a division | segmentation process. It is explanatory drawing showing the operation example in the case of performing integration and a division | segmentation continuously.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Control part, 12 Memory | storage part, 13 Image input part, 14 Image output part, 21 Pre-processing part, 23 Character line drawing extraction processing part, 24 Picture extraction processing part, 25 Character line drawing processing part, 26 Picture partial processing part, 27 Compression Processing unit, 28 Format processing unit.

Claims (7)

A means for demarcating a region including a continuous region of significant pixels in the image data based on the image data to be processed, and generating at least one region information representing the region;
For each of the generated region information, means for generating an integrated region obtained by integrating a plurality of regions according to a predetermined integration rule based on the property for each region defined by each region or the positional relationship between the regions;
An image processing apparatus comprising: A means for demarcating a region including a continuous region of significant pixels in the image data based on the image data to be processed, and generating at least one region information representing the region;
For each of the generated region information, means for dividing a region and generating a plurality of divided regions according to a predetermined division rule based on the property of each region defined by each of the region information,
An image processing apparatus comprising: A means for demarcating a region including a continuous region of significant pixels in the image data based on the image data to be processed, and generating at least one region information representing the region;
For each of the region information, means for integrating a plurality of regions and generating new region information according to a predetermined integration rule based on the property of each region defined by each region or the positional relationship between the regions;
For each of the region information, means for dividing the region according to a predetermined division rule based on the property for each region defined by each region information, and generating new region information;
An image processing apparatus comprising: The image processing apparatus according to claim 1 or 3,
The image processing apparatus according to claim 1, wherein the integration rule includes a condition relating to a size of an area indicated by the area information or an area of an overlapping portion of the area. The image processing apparatus according to claim 2 or 3,
The image processing apparatus according to claim 1, wherein the division rule includes a condition relating to a ratio of an area occupied by the significant pixels in the region indicated by the region information. Using a computer,
Based on the image data to be processed, a region including a continuous region of significant pixels is defined in the image data, and at least one region information representing the region is generated,
For each of the region information, a step of integrating a plurality of regions and generating new region information according to a predetermined integration rule based on the property of each region defined by each region or the positional relationship between the regions, or the region information For each, at least one step of dividing the region and generating new region information according to a predetermined division rule based on the property of each region defined by each region is executed.
An image processing method. On the computer,
A procedure for defining a region including a continuous region of significant pixels in the image data based on the image data to be processed, and generating at least one region information representing the region;
For each of the region information, a process of integrating a plurality of regions and generating new region information according to a predetermined integration rule based on the property of each region defined or the positional relationship between the regions, or the region information A procedure for performing at least one of a process of dividing an area and generating new area information according to a predetermined division rule based on the property of each area defined by each;
An image processing program for executing

Images