JP4935732B2 - Image processing apparatus and image processing program - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing program.

  When enlarging an image (particularly an image including characters), blur and curved irregularities may occur in the multi-value data format, and uneven width / interval and curved irregularities may occur in the binary data format. Further, jaggy is prominent in the nearest neighbor interpolation method, and blurring is remarkable in the linear interpolation method, and unevenness of a curve may occur even in the convolution interpolation method of the sampling function. Each of these methods uses only information based on the resolution of the image (for example, the resolution at the time of scan-in (for example, image reading using a scanner)).

  As a technology related to this, for example, Patent Document 1 aims to improve the reproducibility of an image in which a character image and a photographic image subjected to pseudo halftone processing are mixed. 1 pixel of the input image is divided into two in both the vertical and horizontal directions so as to match the writing density of the recording system and is replaced with a total of 4 pixels, and the character region processing unit and the pixel of interest of the input image and its surrounding 5 × 5 pixels Are matched with a pattern set in advance, and the pixel stroke is converted so as to match the writing density of the recording system so that the hatched character stroke is optimal, and the area determination unit is 5 × 5 pixels The number of four connected pixels in the area is counted, and when the counted value is equal to or larger than a predetermined value, the pixel of interest in the center is determined as a character area pixel, and the output of the character area processing unit is selected. Featured image if less than It is disclosed that an element is determined as a pixel of a photographic region and an output of a photographic region processing unit is selected.

  In addition, for example, in Patent Document 2, when creating a binary image enlarged from low-resolution gradation data, a binary image in which edge portions such as character images are well reproduced in addition to a halftone image. In the difference calculation, the difference in luminance in the vertical and horizontal directions is calculated from the luminance value of the pixel adjacent to the processing pixel of the original image, and the calculation is performed. A luminance gradient is detected from the difference in luminance, and if this luminance gradient is smaller than a certain threshold value, it is determined as a halftone image, and n × according to the luminance of the processing pixel is obtained by binarization halftone processing. When m binarized pixels are generated and the luminance gradient is larger than a certain threshold value, the direction of the luminance gradient is detected, and according to this direction, the number 1 or 0 corresponding to the luminance of the processing pixel is detected. Are arranged, and n × m binarized pixels are generated. The It is disclosed that a binary image is formed by the generated binarized pixels.

In addition, for example, in Patent Document 3, low-resolution image information is converted to high resolution while smoothing out jagged edges of outlines of characters and the like while preventing a reduction in processing capability by a simple additional circuit element. In order to improve the quality of printing, the image information edited in the page memory is multiplied by the video signal generation circuit to double the pixel density in the vertical and horizontal directions for each pixel, and this is multiplied by the line memory. The image replacement circuit sequentially prints the pixel data coming to the center cell of the matrix by forming a predetermined number of pixel data in a matrix in the order of advance, by temporarily storing the data in advance. It is determined whether the pixel data of the central cell corresponds to the black pixel or the white pixel of the stepped portion that causes the jaggedness of the inclined portion of the image outline during printing, and the result of the determination of the jagged factor is determined. It is disclosed that performs pixel substitution (black and white inverted) on the basis of.
Japanese Patent Laid-Open No. 06-225122 Japanese Patent Application Laid-Open No. 06-313151 Japanese Patent Laid-Open No. 08-310057

By the way, if a process for generating a pixel block with an increased resolution using a similar pixel block existing in the target image is used, if there are few similar pixel blocks, a pixel block with an increased resolution is generated. I can't.
An object of the present invention is to provide an image processing apparatus and an image processing program capable of increasing the resolution even when there are few similar pixel clusters in a target image.

The gist of the present invention for achieving the object lies in the inventions of the following items.
The invention according to claim 1 is a pixel block extraction unit that extracts a pixel block from an image, a similar pixel block extraction unit that extracts a similar pixel block extracted by the pixel block extraction unit, and the similar pixel A centroid calculating unit that calculates the centroid of a plurality of similar pixel blocks extracted by the block extracting unit, a plurality of similar pixel blocks extracted by the similar pixel block extracting unit, and a centroid calculated by the centroid calculating unit. A representative pixel block generation unit that generates a representative pixel block of the pixel block, and a representative pixel block storage unit that stores the representative pixel block generated by the representative pixel block generation unit; The means includes the representative pixel block stored by the representative pixel block storage unit in the pixel block cut out by the pixel block cutting unit, and extracts similar pixel blocks. That is an image processing apparatus.

  According to a second aspect of the present invention, the similar pixel block extraction unit counts the number of similar pixel blocks extracted by the pixel block extraction unit, and is based on the number counted by the similar pixel block extraction unit. The image processing apparatus according to claim 1, further comprising a use instruction unit that instructs the similar pixel block extraction unit to use the representative pixel block stored by the representative pixel block storage unit. It is.

  The invention according to claim 3 further includes resolution instruction means for instructing the resolution of the representative pixel block generated by the representative pixel block generation means, and the representative pixel block generation means is the resolution specified by the resolution instruction means. The image processing apparatus according to claim 1, wherein the representative pixel block is generated according to the method.

  The invention according to claim 4 further comprises a rotating means for rotating the representative pixel block generated by the representative pixel block generating unit, and the representative pixel block storage unit stores the representative pixel block rotated by the rotating unit. The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus.

  According to a fifth aspect of the present invention, the similar pixel block extraction unit counts the number of similar pixel blocks extracted by the pixel block extraction unit, and is based on the number counted by the similar pixel block extraction unit. The image processing apparatus according to claim 1, further comprising a storage instruction unit that instructs the representative pixel block storage unit to store the representative pixel block. .

  The invention of claim 6 accepts an instruction of registration, output or erasure by an operator, and when the instruction is registration, processing by the centroid calculation means, the representative pixel block generation means and the representative pixel block storage means is performed. If the instruction is an output, the center-of-gravity calculating unit and the representative pixel block generation unit perform processing. If the instruction is erasure, the representative pixel stored in the representative pixel block storage unit The image processing apparatus according to claim 1, further comprising a control unit that controls to delete the lump.

  The invention according to claim 7 is used when the representative pixel block generation unit generates the representative pixel block when the representative pixel block storage unit stores the representative pixel block generated by the representative pixel block generation unit. 7. The image processing apparatus according to claim 1, wherein the representative pixel block is erased.

  The invention according to claim 8 is characterized in that the representative pixel block storage means stores the number of times the stored representative pixel block is extracted by the similar pixel block extraction means. The image processing device according to item.

  The invention of claim 9 is a computer that extracts a pixel block extracting unit that extracts a pixel block from an image, a similar pixel block extracting unit that extracts a similar pixel block extracted by the pixel block cutting unit, and Calculated by the centroid calculating means for calculating the centroid of a plurality of similar pixel blocks extracted by the similar pixel block extracting means, the similar plurality of pixel blocks extracted by the similar pixel block extracting means and the centroid calculating means. The representative pixel block generating unit that generates a representative pixel block of the pixel block based on the center of gravity, and the representative pixel block storage unit that stores the representative pixel block generated by the representative pixel block generation unit, The pixel block extracting unit includes the representative pixel block stored by the representative pixel block storage unit in the pixel block cut out by the pixel block cutting unit, and a similar pixel block Extracting an image processing program, characterized by.

  According to the image processing apparatus of the first aspect, even if there are few similar pixel clusters in the target image, the resolution can be increased.

  According to the image processing apparatus of the second aspect, when there are not a few pixel clusters similar in the target image, it is possible to generate a high-resolution pixel cluster using only the pixel blocks in the target image.

  According to the image processing apparatus of the third aspect, representative pixel blocks having a plurality of resolutions can be generated and matched with the resolution of the target image.

  According to the image processing apparatus of the fourth aspect, it is possible to deal with a rotated target image by generating a representative pixel block that has been rotated.

  According to the image processing apparatus of the fifth aspect, the number of representative pixel blocks to be stored can be reduced, and the storage capacity can be reduced.

  According to the image processing apparatus of the sixth aspect, various processes can be performed in accordance with an instruction from the operator.

  According to the image processing apparatus of the seventh aspect, it is not necessary to store useless representative pixel blocks, and the storage capacity can be reduced.

  According to the image processing apparatus of the eighth aspect, it is possible to perform processing according to the frequency of similar pixel blocks.

  According to the image processing program of the ninth aspect, even if there are few similar pixel clusters in the target image, the resolution can be increased.

First, an outline of the present embodiment (first to eighth embodiments) will be described.
Note that the pixel block targeted by this embodiment includes at least a pixel region that is continuous in four or eight connections, and also includes a set of these pixel regions. The set of these pixel areas means that there are a plurality of continuous pixel areas such as 4-connected, and the plurality of pixel areas are in the vicinity. Here, what is in the vicinity is, for example, an image area in which the pixel areas are close to each other in distance, an image area that is projected vertically or horizontally so as to cut out one character at a time from a line as a sentence, and cut out at a blank spot Or an image region cut out at regular intervals.
In many cases, an image of one character is formed as one pixel block. However, it is not necessary that the pixel area is actually recognizable as a character by humans. There are a part of a character, a character with an extra pixel area added to the character, a pixel area that does not form a character, and the like, and any block of pixels may be used. For example, it may be a figure such as a circle or a line. Hereinafter, the term “character” or “character image” means “pixel block” unless otherwise specified.

This embodiment is intended for an image having a plurality of characters. Using these characters, a high-resolution character image (representative character image data) is generated. As a technique for generating a high-resolution character image, there is a Super resolution technique.
In normal document images, the same character image often appears multiple times. Furthermore, when a plurality of similar character images (including the same character image) are scanned in, the scan-in (sampling) phase is often shifted from the character image (center of gravity). By using a plurality of character images whose sampling phases are shifted from each other, it is possible to construct a character image having a resolution higher than the scan-in resolution (in addition, “structuring a high-resolution character image”). Also referred to as “creating a representative image”). At this time, it is not always necessary to increase the image size, and even if the image size is the same as the original size, only the effective resolution represented by the data can be increased. Alternatively, random noise can be removed.

However, it is not effective for a character image that is used only once for one document image, and is less effective for a character image that is used less frequently.
Therefore, it is effective to prepare various character image data in advance and search for a character image having the same shape including them. For example, if character code data and corresponding character image data are prepared as font data, such as document creation software and printers, they can be used, but if not, they must be newly prepared There is. Further, for example, the processing load for searching for characters having the same shape including the character shapes of 2000 types of common Kanji characters × 4 types of fonts (Mincho, Gothic, etc.) = 8000 types becomes large.

On the other hand, instead of processing one document image, a plurality of document images may be processed together. By processing a plurality of document images, the probability that the same character image is used multiple times increases.
However, as the number of document images increases, the search processing load for character images having the same shape increases.
In this embodiment, representative character image data generated from target image data is registered as character group image data (so-called dictionary), and representative character image data is generated from other target image data using the character group image data. To do. Even if a character image appears only once or a small number of times in other target image data, representative character image data is generated by using character group image data.

Hereinafter, examples of various preferred embodiments for realizing the present invention will be described with reference to the drawings.
<First Embodiment>
FIG. 1 is a conceptual module configuration diagram of a configuration example according to the first embodiment.
The module generally refers to components such as software (computer program) and hardware that can be logically separated. Therefore, the module in the present embodiment indicates not only a module in a computer program but also a module in a hardware configuration. Therefore, the present embodiment also serves as an explanation of a computer program, a system, and a method. However, for the sake of explanation, the words “store”, “store”, and equivalents thereof are used. However, when the embodiment is a computer program, these words are stored in a storage device or stored in memory. It is the control to be stored in the device. In addition, the modules correspond almost one-to-one with the functions. However, in mounting, one module may be composed of one program, or a plurality of modules may be composed of one program. A plurality of programs may be used. The plurality of modules may be executed by one computer, or one module may be executed by a plurality of computers in a distributed or parallel environment. Note that one module may include other modules. In the following, “connection” includes not only physical connection but also logical connection (data exchange, instruction, reference relationship between data, etc.).
In addition, the system or device is configured by connecting a plurality of computers, hardware, devices, and the like by communication means such as a network (including one-to-one correspondence communication connection), etc., and one computer, hardware, device. The case where it implement | achieves by etc. is also included. “Apparatus” and “system” are used as synonymous terms. Further, the term “predetermined” is used in addition to a predetermined meaning, depending on the situation / state at that time or including the meaning according to the situation / state until then.

  As shown in FIG. 1, the first embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110 and a character group image data reduction module 112.

The character extraction module 102 is connected to the similar character image data extraction module 104 and the character group image data reduction module 112, receives a document image, and extracts characters from the document image. The cut character (character image data) is transferred to the similar character image data extraction module 104. Further, the character cutout module 102 receives the representative character reduced from the character group image data reduction module 112 and passes it to the similar character image data extraction module 104 together with the character cut out from the document image.
Examples of character cut-out processing performed by the character cut-out module 102 include, for example, Japanese Patent Laid-Open No. 5-28301, Japanese Patent Laid-Open No. 5-120481, Japanese Patent Laid-Open No. 5-143776, Japanese Patent Laid-Open No. 5-174185, and Japanese Patent Laid-Open No. No. 44406, JP-A-6-187489, JP-A-6-348911, JP-A-7-13994, JP-A-7-160810, JP-A-8-161432, JP-A-8-297718. JP, 10-69524, JP 10-134145, JP 10-261047, JP 2000-57261, JP 2001-43314, JP 2004-78531, etc. Can be used.

  The similar character image data extraction module 104 is connected to the character cutout module 102, the character image data centroid calculation module 106, and the character image data configuration module 108. A similar character is extracted from the characters extracted by the character extraction module 102. Then, the representative characters stored by the character group image data storage module 110 are included in the characters extracted by the character extraction module 102, and similar characters are extracted. The representative characters stored by the character group image data storage module 110 may be received directly from the character group image data storage module 110, may be received via the character group image data reduction module 112, and You may receive via the character extraction module 102. FIG. The extracted similar characters (similar character image data) are passed to the character image data centroid calculation module 106 and the character image data composition module 108.

There are various methods for determining whether or not the similar character image data extraction module 104 is similar. Hereinafter, an example of a method for determining the similarity of characters will be described.
The following is an example of a method for verifying the similarity of characters.
(A1) Two binary images that are clipped character images are input.
(A2) The centers of gravity of the black pixels of the two binary images are matched.
(A3) An XOR (exclusive OR) operation is performed on two binary images with the center of gravity.
(A4) Further, it is moved slightly (several pixels vertically and horizontally), and the number of pixels (different pixels) that become 1 as a result of the XOR operation is counted. As a result of various XOR operations after movement including non-movement, if the minimum value of the number of pixels that are 1 (different pixels) is equal to or less than a predetermined threshold, it is determined that the character images are similar. The threshold value may be a numerical value proportional to the number of pixels of the character image.
(A5) The cut-out character images are calculated for the similarity with other character images by the above-described method, and character images that are similar to each other are grouped together (grouped).

In addition to the above, as a determination of whether or not the character images are similar, for example, the cut character image is treated as one vector data, clustered, and a character image similar to the character image is extracted. Also good. In this case, when the distance (for example, the Euclidean distance) between the vector data representing the character image and the vector data representing the character image to be determined is equal to or less than a predetermined value (that is, when the distance between the two vector data is short) Then, it is determined that the character image is similar to the character image to be determined.
Furthermore, an expanded image may be generated based on the result image of the logical operation of the two character image patterns, and it may be determined whether or not they are similar based on the ratio of overlapping with the expanded image. That is, the degree of difference may be determined based on the degree of expansion (expansion radius) when the two match.
In addition, a method described in Japanese Patent Application Laid-Open No. 07-200755, “Managing Gigabytes” by IH Witten, A. Moffat, and TC Bell, Morgan Kaufmmann Publishers, pp. 320-332.

Note that the similarity is the amount that is maximized when two images are congruent and decreases according to the degree of difference.
Instead of similarity, an amount that is minimized when two images are congruent and increases according to the degree of difference may be used. In this case, it is called “distance” or “degree of difference”. The distance is an image represented by a vector (such as a pixel value itself as a vector or an image feature amount as a vector). This is the distance (distance) between the images. For example, there are Euclidean distance, Manhattan distance, Hausdorff distance, Mahalanobis distance, angles between vectors θ, cos θ, and the square of cos θ.
That is, in order to search for similar characters, “distance” or “difference” may be used in addition to the similarity. Therefore, “similar” is a concept including high similarity, close distance, and the like.

  In the above-described similar character search process, the degree of similarity of one character with all other characters is measured. However, in order to increase the speed, the size of the character rectangle is somewhat the same. Only for the above, similarity calculation with a large amount of calculation may be performed.

  The character image data centroid calculation module 106 is connected to the similar character image data extraction module 104 and the character image data configuration module 108, and calculates centroids of a plurality of similar characters extracted by the similar character image data extraction module 104. . The character centroid (the centroid value of the character image data) is passed to the character image data composition module 108. Note that the center of gravity is obtained with an accuracy of one pixel or less (one pixel in the document image). The calculation process of the center of gravity will be described later with reference to FIGS.

  The character image data composition module 108 is connected to the similar character image data extraction module 104, the character image data centroid calculation module 106, and the character group image data storage module 110, and is similar to those extracted by the similar character image data extraction module 104. Based on the plurality of characters and the center of gravity calculated by the character image data center of gravity calculation module 106, a representative character of the character is generated. The representative character is a character with a resolution higher than that of the character cut out by the character cut-out module 102 and is generated from a plurality of characters. The representative character is generated by estimating the phase of the similar character in the document image from the relative position of the center of gravity, and generating a representative character having a resolution equal to or higher than the resolution of the document image from the estimated phase and the similar character. is there. The generation of the representative character will be described later with reference to FIGS. The character image data composition module 108 outputs the generated representative character to the outside and passes it to the character group image data storage module 110.

The character group image data storage module 110 is connected to the character image data configuration module 108 and the character group image data reduction module 112. The representative character generated by the character image data composition module 108 is stored. Then, it is accessed from the character group image data reduction module 112, and the representative character (character group image data) is passed to the character cutout module 102. Alternatively, the character may be accessed from the character cutout module 102 or the similar character image data extraction module 104, and the representative character may be passed to the character cutout module 102 or the similar character image data extraction module 104. For example, as shown in FIG. 2, the character group image data storage module 110 stores image data 200 composed of representative characters.
The character group image data reduction module 112 is connected to the character group image data storage module 110 and the character cutout module 102, and in response to a request from the character cutout module 102 or the similar character image data extraction module 104, A representative character is extracted from the image data storage module 110, and a reduction process (a process for reducing the resolution) is performed in order to match the resolution of the representative character with the resolution of the document image received by the character cutout module 102. The reduced representative character (reduced image data) is transferred to the character cutout module 102 or the similar character image data extraction module 104 which has made the request. When the representative character and the document image have the same resolution, the character group image data reduction module 112 does not need to perform processing. Therefore, the reduction process here includes a process of reducing (doing nothing) one time.

FIG. 3 is a flowchart illustrating a processing example according to the first exemplary embodiment.
In step S302, the character cutout module 102 receives a document image and cuts out a character from the document image.
In step S304, the character group image data reduction module 112 extracts representative characters from the character group image data storage module 110.
In step S306, the character group image data reduction module 112 performs a reduction process on the representative character extracted in step S304.
In step S308, the similar character image data extraction module 104 extracts a character similar to the cut out character from the character cut out in step S302 and the representative character reduced in step S306.

In step S310, the character image data center-of-gravity calculation module 106 calculates the centers of gravity of the plurality of similar characters extracted in step S308.
In step S312, the character image data composition module 108 generates a representative character based on the character extracted in step S308 and the center of gravity calculated in step S310.
In step S314, the character group image data storage module 110 stores the character generated in step S312.
In step S316, the character image data composition module 108 outputs the representative character to the outside.
Note that any of the processes in steps S314 and S316 may be performed first or simultaneously.

An example of processing by the character cutout module 102, the similar character image data extraction module 104, the character image data centroid calculation module 106, and the character image data configuration module 108 will be described with reference to FIG. The character image data construction module 108 performs enlargement processing.
The character cutting module 102 sets the character image 411, the character image 412, and the character image 413 in the target document image 410 in which a plurality of characters “2” are described as the target images. Then, each character image is cut out with the resolution of the character image 411. Further, the character size / character position data of each character image may be extracted and passed to the character image data configuration module 108.
The similar character image data extraction module 104 searches for similar character images and generates a plurality of character image groups “2”.
Then, the character image data center-of-gravity calculation module 106 obtains the emphasis (intersection of the center of gravity line 411A) of the character image 411, the character image 412, and the character image 413, and the character image data composition module 108 matches the emphasis. The high resolution character image 420 is generated by shifting the phase.
Further, the character image data composition module 108 may assign character code data of a character image “2”, for example. The character code assignment processing may be performed by character recognition processing.
Further, the character image data composition module 108 may generate font data 430 that is outline information, for example, from the high resolution character image 420. The character size / character position data of the character image may be output in association with each other.

An example of gravity center calculation processing by the character image data gravity center calculation module 106, generation of representative characters by the character image data configuration module 108, that is, an example of enlargement processing for constructing a high-resolution character image will be described with reference to FIG.
FIG. 5A illustrates a sampling grid (first sampling grid 501, first sampling grid 502, first sampling grid 503, first resolution at the resolution (first resolution) of the target document image 410. Sampling center 504) and barycentric positions (centroid 501A, centroid 502A, centroid 503A, centroid 504A) of the character image. The character image data centroid calculation module 106 calculates the centroid of each character image.
Next, the character image data construction module 108 first moves the phases of the four sampling grids based on the center of gravity of the character image, as shown in FIG.

FIGS. 5C and 5D are diagrams illustrating an example of a technique for setting a sampling grid having a second resolution higher than the first resolution. The circled numbers (1, 2, 3, 4) shown in FIG. 5C exemplify the value of the character image at the first resolution. Here, the character image is plotted so that the circled numbers are represented on the grid points of the sampling grid at the first resolution.
In FIG. 5C, a first sampling grid 505 is a first resolution sampling grid. In FIG. 5D, the second sampling grid 506 is a sampling grid of the second resolution (high resolution image).
When the phases of the four sampling grids at the first resolution are moved, the character image data construction module 108 sets the sampling grid at the second resolution as shown in FIG. The phase of the sampling grating at the second resolution is moved so that the centroids of the two coincide.

FIG. 5E is a diagram for explaining an example of a method for calculating the value of a character image at the second resolution. The circled numbers in the center of the second sampling grid 506A, the second sampling grid 506B, the second sampling grid 506C, and the second sampling grid 506D indicate the value of the character image at the second resolution. This is just an example. Here, the character image in the second resolution is shown such that the circled numbers at the center are represented on the grid points of the sampling grid in the second resolution.
Then, the character image data construction module 108 interpolates the pixel value of the character image at the second resolution from the pixel value of each character image based on the phase of each character image at the first resolution. In this example, the character image data construction module 108 applies the nearest neighbor interpolation method to interpolate the pixel values of the character image at the second resolution. That is, the character image data composition module 108 samples the character image at the second resolution among the four values of the character image at the first resolution (circle numbers 1, 2, 3, 4 in FIG. 5E). The value closest to the grid point is selected as the character image value at the second resolution. Specifically, in the second sampling grid 506A, the value closest to the center is “1”, and “1” is adopted (the circled number is 1). The interpolation method is not limited to this method, and other methods (for example, linear interpolation method) may be applied.
The processing by the character image data configuration module 108 is not limited to the processing described above, and linear interpolation, cubic convolution, or the like may be used.

FIG. 6 is an explanatory diagram showing an example of processing by the character image data gravity center calculation module 106 and the character image data composition module 108 including the character group image data storage module 110.
The character cutout module 102 cuts out the character image 411 from the target document image data 610. The similar character image data extraction module 104 extracts a character image 621 similar to the character image 411 from the previously prepared character group image data 620 stored in the character group image data storage module 110. In the case of the example in FIG. 6, since there is no character similar to the character image 411 in the target document image data 610, the similar character image data extraction module 104 performs only the character image 621 as a character similar to the character image 411. Is extracted.
The character image data center-of-gravity calculation module 106 calculates the center of gravity of each of the character image 411 and the character image 621 (the center of the cross line in the character images 411 and 621 in FIG. 6).
The character image data composition module 108 generates a high resolution character image 420 that is a representative character. Then, the character group image data storage module 110 stores the generated high resolution character image 420 in the character group image data 620 prepared previously.

<Second Embodiment>
As shown in FIG. 7, the second embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110, a character group image data reduction module 112, and a character group image data use instruction module 701. A character group image data use instruction module 701 is added to the first embodiment. Note that the same type of modules as those in the first embodiment are denoted by the same reference numerals, and redundant description is omitted. Further, even modules with the same reference numerals are described below, which are added to or replaced with the functions and functions of the corresponding modules in the first embodiment.

The similar character image data extraction module 104 is connected to the character cutout module 102, the character image data centroid calculation module 106, the character image data configuration module 108, and the character group image data use instruction module 701. Count the number of similar characters among the extracted characters. Then, the number of similar characters is passed to the character group image data use instruction module 701. Further, the similar character image data extraction module 104 extracts similar characters using the representative characters stored by the character group image data storage module 110 based on an instruction from the character group image data use instruction module 701.
The character group image data storage module 110 is connected to the character image data configuration module 108, the character group image data reduction module 112, and the character group image data use instruction module 701, and receives instructions from the character group image data use instruction module 701. Based on this, the representative character may be passed to the character group image data reduction module 112 (the character cutout module 102 or the similar character image data extraction module 104 may be referred to as the character group image data reduction module 112). .

The character group image data use instruction module 701 is connected to the similar character image data extraction module 104 and the character group image data storage module 110. Based on the number counted by the similar character image data extraction module 104, the similar character image data extraction module 104 is instructed to use the representative character stored by the character group image data storage module 110. This may be realized by instructing the character group image data storage module 110 to actively pass representative characters from the character group image data storage module 110 to the character group image data reduction module 112 or the like. Alternatively, an instruction may be given to the character cutout module 102 or the like, and the representative character may be used from the character group image data storage module 110.
In addition, as “based on the number counted by the similar character image data extraction module 104”, for example, there is no similar character in the document image (in the case where there is one character) or a predetermined threshold (to generate a representative character). In some cases, the number is less than the number of necessary similar characters. Therefore, when there are many similar characters in the document image, representative characters are generated only with the similar characters in the document image.

FIG. 8 is a flowchart illustrating a processing example according to the second exemplary embodiment.
In step S802, the character cutout module 102 receives a document image and cuts out a character from the document image.
In step S804, the similar character image data extraction module 104 extracts similar characters from the characters cut out in step S802.
In step S806, it is determined whether the number of similar characters extracted in step S804 is less than a predetermined threshold. In this determination, if it is determined that the number is small, the process proceeds to step S808, and otherwise, the process proceeds to step S814.
In step S <b> 808, the character group image data reduction module 112 extracts representative characters from the character group image data storage module 110 based on the use instruction of the character group image data use instruction module 701.

In step S810, the character group image data reduction module 112 performs a reduction process on the representative character extracted in step S808.
In step S812, the similar character image data extraction module 104 extracts a character similar to the cut out character from the character cut out in step S802 and the representative character reduced in step S810.
In step S814, the character image data center-of-gravity calculation module 106 calculates the centers of gravity of a plurality of similar characters extracted in step S804 or step S812.
In step S816, the character image data construction module 108 generates a representative character based on the character extracted in step S804 or step S812 and the center of gravity calculated in step S814.
In step S818, the character group image data storage module 110 stores the character generated in step S816.
In step S820, the character image data composition module 108 outputs the representative character to the outside.

<Third Embodiment>
As shown in FIG. 9, the third embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110, a character group image data reduction module 112, and a representative character resolution instruction module 901. A representative character resolution instruction module 901 is added to the first embodiment. The same type of modules as those in the first and second embodiments are denoted by the same reference numerals, and redundant description is omitted. Further, even modules having the same reference numerals are added or replaced with the functions and functions of the corresponding modules in the first and second embodiments.

  The representative character resolution instruction module 901 is connected to the character image data configuration module 108. The resolution of the representative character generated by the character image data composition module 108 is indicated. The designated resolution may be the resolution of the character cut out by the character cut-out module 102. The resolution may be plural. When there are a plurality of characters, the resolution of the representative characters output to the outside by the character image data composition module 108 and the representative characters stored in the character group image data storage module 110 may be different. Furthermore, when different, the representative character output to the outside may have a higher resolution than the representative character stored in the character group image data storage module 110. Also, the resolution of representative characters stored in the character group image data storage module 110 may be plural. For example, the resolution of the representative character output to the outside and the resolution of the character cut out by the character cutout module 102.

The character image data configuration module 108 is connected to the similar character image data extraction module 104, the character image data centroid calculation module 106, the character group image data storage module 110, and the representative character resolution instruction module 901. A representative character is generated according to the resolution instructed by the representative character resolution instruction module 901. When there are a plurality of designated resolutions, there are a plurality of representative characters to be generated.
When there are a plurality of representative characters generated by the character image data composition module 108, the character group image data storage module 110 stores the representative characters having a plurality of resolutions. For example, as shown in FIG. 10, the image data 1000 stores character group image data 1010 and character group image data 1020 composed of representative characters.
The character group image data reduction module 112 selects and reduces a representative character stored in the character group image data storage module 110 that has a resolution (including the same) that is close to the character extracted by the character extraction module 102. May be. Further, when the resolution of the representative character stored in the character group image data storage module 110 is the same as the resolution of the character extracted by the character cutout module 102, the character group image data reduction module 112 may be omitted, and the character group image data storage The module 110 may be connected to the character cutout module 102 or the similar character image data extraction module 104.

FIG. 11 is a flowchart illustrating a processing example according to the third exemplary embodiment.
Steps S1102 to S1110 are the same as steps S302 to S310 in the flowchart shown in FIG. 3 of the first embodiment.
In step S <b> 1112, the character image data configuration module 108 generates a representative character having the resolution specified by the representative character resolution instruction module 901. The resolution instruction by the representative character resolution instruction module 901 may be any time before step S1112. For example, it may be specified only once before the process according to the third embodiment starts.
In step S1114, the character group image data storage module 110 stores a plurality of representative characters generated in step S1112.
Step S1116 is equivalent to step S316 in the flowchart shown in FIG. 3 of the first embodiment.

<Fourth embodiment>
As shown in FIG. 12, the fourth embodiment includes a character extraction module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110, a character group image data reduction module 112, and a representative character image data rotation module 1201. A representative character image data rotation module 1201 is added to the first embodiment. The same type of modules as those in the first to third embodiments are denoted by the same reference numerals, and redundant description is omitted. Further, even modules having the same reference numerals are described below, which are added to or replaced with the functions and functions of the corresponding modules in the first to third embodiments.

The character image data composition module 108 is connected to the similar character image data extraction module 104, the character image data centroid calculation module 106, and the representative character image data rotation module 1201. The generated representative characters are sent to the representative character image data rotation module 1201. hand over.
The representative character image data rotation module 1201 is connected to the character image data composition module 108 and the character group image data storage module 110 and rotates the representative character generated by the character image data composition module 108. The rotated representative character is stored in the character group image data storage module 110. You may rotate by several rotation angles. The rotation angle may be an angle in units of 90 degrees. That is, the representative character may be rotated at four angles of 0 degrees (not rotated), 90 degrees, 180 degrees, and 270 degrees.
The character group image data storage module 110 is connected to the representative character image data rotation module 1201 and the character group image data reduction module 112. The representative character rotated by the representative character image data rotation module 1201 is stored. Representative characters rotated at a plurality of rotation angles may be stored. For example, as shown in FIG. 13, image data 1300 includes character group image data 1301, character group image data 1302 rotated 180 degrees, character group image data 1303 rotated 270 degrees to the right, and character groups rotated 90 degrees to the right. It has image data 1304.
The character group image data reduction module 112 may reduce all the representative characters stored in the character group image data storage module 110. If the rotation angle of the target document image is known, the character group image data reduction module 112 The representative characters of the corresponding rotation angle stored in the group image data storage module 110 may be reduced.

FIG. 14 is a flowchart illustrating a processing example according to the fourth exemplary embodiment.
Steps S1402 to S1412 are the same as steps S302 to S312 in the flowchart shown in FIG. 3 of the first embodiment.
In step S1414, the representative character image data rotation module 1201 rotates the representative character generated in step S1412.
In step S1416, the character group image data storage module 110 stores a plurality of representative characters rotated in step S1414.
Step S1418 is equivalent to step S316 of the flowchart shown in FIG. 3 of the first embodiment.

<Fifth embodiment>
As shown in FIG. 15, the fifth embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110, a character group image data reduction module 112, and a representative character image data storage instruction module 1501. A representative character image data storage instruction module 1501 is added to the first embodiment. The same type of modules as those in the first to fourth embodiments are denoted by the same reference numerals, and redundant description is omitted. Further, even modules with the same reference numerals are described below, which are added to or replaced with the functions and functions of the corresponding modules in the first to fourth embodiments.

The similar character image data extraction module 104 is connected to the character cutout module 102, the character image data centroid calculation module 106, the character image data configuration module 108, and the representative character image data storage instruction module 1501. Count the number of similar characters among the extracted characters. Then, the number of similar characters is passed to the representative character image data storage instruction module 1501.
The representative character image data storage instruction module 1501 is connected to the similar character image data extraction module 104 and the character group image data storage module 110, and based on the number counted by the similar character image data extraction module 104, the character group image Instructing the data storage module 110 to store representative characters. In addition, as “based on the number counted by the similar character image data extraction module 104”, for example, there is no similar character in the document image (in the case where there is one character) or a predetermined threshold (to generate a representative character). In some cases, the number is less than the number of necessary similar characters. Therefore, when there are many similar characters in the document image, the character group image data storage module 110 does not need to store the representative characters of the characters.
The character group image data storage module 110 is connected to the character image data configuration module 108, the character group image data reduction module 112, and the representative character image data storage instruction module 1501, and is based on instructions from the representative character image data storage instruction module 1501. And memorize the representative characters.

FIG. 16 is a flowchart illustrating a processing example according to the fifth exemplary embodiment.
Steps S1602 to S1612 are equivalent to steps S302 to S312 of the flowchart shown in FIG. 3 of the first embodiment.
In step S1614, the representative character image data storage instruction module 1501 determines whether the number of similar characters counted by the similar character image data extraction module 104 is less than a predetermined threshold. In this determination, if it is determined that the number is small, the process proceeds to step S1616, otherwise the process proceeds to step S1618.
In step S1616, the character group image data storage module 110 stores the representative character generated in step S1612.
Step S1618 is equivalent to step S316 of the flowchart shown in FIG. 3 of the first embodiment.

  As shown in FIG. 17, the sixth embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110, a character group image data reduction module 112, and a data output / storage / erasure control module 1701. A data output / storage / erasure control module 1701 is added to the first embodiment. The same type of modules as those in the first to fifth embodiments are denoted by the same reference numerals, and redundant description is omitted. Further, even modules having the same reference numerals are added or replaced with the functions and functions of the corresponding modules in the first to fifth embodiments.

The sixth embodiment is a learning mode in which the processing according to the first embodiment is performed, and learning dedicated to only storing the representative characters in the character group image data storage module 110 without outputting the representative characters. 1 of a mode, a non-learning mode in which only representative characters are output without storing the representative characters, and a learning initialization mode in which the representative characters stored in the character group image data storage module 110 are deleted. Two or more are provided, and the operator gives an instruction for switching them.
The learning mode is for storing and outputting representative characters to the character group image data storage module 110, and is performed at an early stage of use (use of outputting representative characters while generating a learning dictionary). It is. The dedicated learning mode is for storing only representative characters (use for generating only a learning dictionary). The non-learning mode is used after the representative character is stored in the learning mode or the learning dedicated mode, and only the representative character is output without storing the representative character in order to increase the speed. The learning initialization mode is for erasing the memory of representative characters (erasing the learning dictionary), and is used for dictionary maintenance.

The character image data center-of-gravity calculation module 106 is connected to the similar character image data extraction module 104, the character image data composition module 108, and the data output / storage / erasure control module 1701. Based on the above, the center of gravity calculation processing is performed.
The character image data composition module 108 is connected to the similar character image data extraction module 104, the character image data centroid calculation module 106, the character group image data storage module 110, and the data output / storage / erasure control module 1701. Based on an instruction from the storage / erasure control module 1701, representative character generation processing is performed.
The character group image data storage module 110 is connected to the character image data configuration module 108, the character group image data reduction module 112, and the data output / storage / erasure control module 1701. The representative character erasure process is performed based on the above.

The data output / storage / erasure control module 1701 is connected to the character image data gravity center calculation module 106, the character image data configuration module 108, and the character group image data storage module 110. An instruction for registration, output or deletion by the operator is accepted.
If the instruction is registration, control is performed so that processing by the character image data center-of-gravity calculation module 106, the character image data composition module 108, and the character group image data storage module 110 is performed. However, the character image data composition module 108 is controlled not to output representative characters to the outside.
When the instruction is an output, control is performed so that the character image data center-of-gravity calculation module 106 and the character image data composition module 108 perform processing. However, the character image data composition module 108 controls the character group image data storage module 110 not to store the representative character.
When the instruction is erasure, control is performed so as to erase the representative character stored in the character group image data storage module 110.
Moreover, the instruction | indication from the operator who receives may be one or more.
When the instruction is registration and output, control is performed so that processing by the character image data centroid calculation module 106, the character image data composition module 108, and the character group image data storage module 110 is performed. That is, it is equivalent to the processing according to the first embodiment.
When the instruction is registration and output, it corresponds to the learning mode described above, when the instruction is registration, it corresponds to the learning dedicated mode, when the instruction is output, it corresponds to the non-learning mode, and when the instruction is erased This corresponds to the learning initialization mode.

FIG. 18 is a flowchart illustrating a processing example according to the sixth exemplary embodiment.
Steps S1802 to S1808 are the same as steps S302 to S308 in the flowchart shown in FIG. 3 according to the first embodiment.
In step S1810, the data output / storage / erase control module 1701 accepts an instruction from the operator. If the instruction is registration, the process proceeds to step S1812. If output, the process proceeds to step S1818. If erased, the process proceeds to step S1824. .
Steps S1812 to S1816 are the same as steps S310 to S314 in the flowchart shown in FIG. 3 of the first embodiment.
Steps S1818 to S1822 are the same as steps S310, S312 and S316 in the flowchart shown in FIG. 3 according to the first embodiment.
In step S1824, the representative character stored in the character group image data storage module 110 is deleted.
In the case of registration and output, the processing from step S1812 to step S1816 and the processing of step S1822 are performed.

<Seventh embodiment>
As shown in FIG. 19, the seventh embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data storage module. 110 and a character group image data reduction module 112. In addition to the first embodiment, a similar character image data extraction module 104 and a character group image data storage module 110 are connected. The same type of modules as those in the first to sixth embodiments are denoted by the same reference numerals, and redundant description is omitted. Further, even modules with the same reference numerals are described below, which are added to or replaced with the functions and functions of the corresponding modules in the first to sixth embodiments.

The similar character image data extraction module 104 is connected to the character cutout module 102, the character image data centroid calculation module 106, the character image data configuration module 108, and the character group image data storage module 110. It is also passed to the image data storage module 110. The similar character to be passed to the character group image data storage module 110 may be a character stored in the character group image data storage module 110.
The character group image data storage module 110 is connected to the similar character image data extraction module 104, the character image data composition module 108, and the character group image data reduction module 112. The character group image data storage module 110 stores the representative characters generated by the character image data composition module 108. In the case of storing, the representative character used when the character image data construction module 108 generates the representative character is deleted. That is, when a representative character newly generated by the character image data composition module 108 is stored, the representative character stored in the character group image data storage module 110 corresponding to the representative character is deleted. In the character group image data storage module 110, only one character is determined as a similar character.

FIG. 20 is a flowchart illustrating a processing example according to the seventh exemplary embodiment.
Steps S2002 to S2014 are the same as steps S302 to S314 in the flowchart shown in FIG. 3 according to the first embodiment.
In step S2016, the character group image data storage module 110 erases the same characters as the similar characters received from the similar character image data extraction module 104 among the representative characters already stored in the character group image data storage module 110.
Step S2018 is equivalent to step S316 of the flowchart shown in FIG. 3 of the first embodiment.

<Eighth Embodiment>
As shown in FIG. 21, the eighth embodiment includes a character cutout module 102, a similar character image data extraction module 104, a character image data barycenter calculation module 106, a character image data configuration module 108, and a character group image data reduction module. 112, a frequency storage / data erasure control module 2101, a character group image data and frequency storage module 2102. A frequency storage / data deletion control module 2101, character group image data, and a frequency storage module 2102 are added to the first embodiment. The same type of modules as those in the first to seventh embodiments are denoted by the same reference numerals, and redundant description is omitted. Further, even modules with the same reference numerals are described below, which are added to or replaced with the functions and functions of the corresponding modules in the first to seventh embodiments.

The similar character image data extraction module 104 is connected to the character cutout module 102, the character image data centroid calculation module 106, the character image data configuration module 108, and the character group image data and frequency storage module 2102. It is also passed to the character group image data and frequency storage module 2102. The similar characters passed to the character group image data and the frequency storage module 2102 may be the characters stored in the character group image data and the frequency storage module 2102.
The character group image data and frequency storage module 2102 is connected to the similar character image data extraction module 104, the character image data configuration module 108, the character group image data reduction module 112, and the frequency storage / data erasure control module 2101. The representative character generated by the data composition module 108 and the number of times the representative character is extracted by the similar character image data extraction module 104 are stored. That is, the representative character and the number of times that the representative character is extracted as a similar character (the number of times used to generate a new representative character) are stored in association with each other. Then, it is accessed from the character group image data reduction module 112 or the like, and the representative character (character group image data) is passed to the character cutout module 102 or the like.
The frequency storage / data erasure control module 2101 is connected to the character group image data and the frequency storage module 2102, and deletes the representative characters stored in the character group image data and the frequency storage module 2102. At that time, the representative character is erased according to the number of representative characters. For example, the representative character is deleted a number of times less than a predetermined threshold. That is, representative characters that are used less frequently are deleted.

FIG. 22 is a flowchart illustrating an example of processing according to the eighth embodiment.
Steps S2202 to S2212 are equivalent to steps S302 to S312 in the flowchart shown in FIG. 3 of the first embodiment.
In step S2214, the character group image data and frequency storage module 2102 stores the representative character and the usage frequency of the representative character.
Step S2216 is equivalent to step S316 in the flowchart shown in FIG. 3 of the first embodiment.

FIG. 23 is a flowchart illustrating an example of a representative character erasing process according to the eighth embodiment.
In step S2302, the frequency storage / data deletion control module 2101 determines whether or not the use frequency stored in the character group image data and the frequency storage module 2102 is less than a predetermined threshold. In this determination, if it is determined that the number is small, the process proceeds to step S2304; otherwise, the process ends (step S2306).
In step S <b> 2304, the frequency storage / data deletion control module 2101 deletes the representative character that is the target stored in the character group image data and the frequency storage module 2102.

  With reference to FIG. 24, a hardware configuration example of the above-described embodiment will be described. The configuration illustrated in FIG. 24 is configured by, for example, a personal computer (PC), and illustrates a hardware configuration example including a data reading unit 2417 such as a scanner and a data output unit 2418 such as a printer.

  The CPU (Central Processing Unit) 2401 includes various modules described in the above-described embodiments, that is, the character extraction module 102, the similar character image data extraction module 104, the character image data centroid calculation module 106, and the character image data configuration module. Reference numeral 108 denotes a control unit that executes processing according to a computer program describing an execution sequence of each module such as the character group image data reduction module 112 and the character group image data use instruction module 701.

  A ROM (Read Only Memory) 2402 stores programs used by the CPU 2401, operation parameters, and the like. A RAM (Random Access Memory) 2403 stores programs used in the execution of the CPU 2401, parameters that change as appropriate during the execution, and the like. These are connected to each other by a host bus 2404 including a CPU bus.

  The host bus 2404 is connected to an external bus 2406 such as a peripheral component interconnect / interface (PCI) bus via a bridge 2405.

  A keyboard 2408 and a pointing device 2409 such as a mouse are input devices operated by an operator. The display 2410 includes a liquid crystal display device or a CRT (Cathode Ray Tube), and displays various types of information as text and image information.

  An HDD (Hard Disk Drive) 2411 includes a hard disk, drives the hard disk, and records or reproduces a program executed by the CPU 2401 and information. The hard disk stores input image data, character group image data, and the like. Further, various computer programs such as various other data processing programs are stored.

  The drive 2412 reads data or a program recorded on a removable recording medium 2413 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and the data or program is read out to the interface 2407 and the external bus 2406. , The bridge 2405, and the RAM 2403 connected via the host bus 2404. The removable recording medium 2413 can also be used as a data recording area similar to the hard disk.

  The connection port 2414 is a port for connecting the external connection device 2415 and has a connection unit such as USB, IEEE1394. The connection port 2414 is connected to the CPU 2401 and the like via the interface 2407, the external bus 2406, the bridge 2405, the host bus 2404, and the like. A communication unit 2416 is connected to the network and executes data communication processing with the outside. The data reading unit 2417 is, for example, a scanner, and executes document reading processing. The data output unit 2418 is, for example, a printer, and executes document data output processing.

  The hardware configuration shown in FIG. 24 shows one configuration example, and the above-described embodiment is not limited to the configuration shown in FIG. 24, and the modules described in the above-described embodiment can be executed. Any configuration may be used. For example, some modules may be configured with dedicated hardware (for example, Application Specific Integrated Circuit (ASIC), etc.), and some modules are in an external system and connected via a communication line In addition, a plurality of systems shown in FIG. 24 may be connected to each other via communication lines so as to cooperate with each other. Further, it may be incorporated in a copying machine, a fax machine, a scanner, a printer, a multifunction machine (an image processing apparatus having any two or more functions of a scanner, a printer, a copying machine, a fax machine, etc.).

  The first to eighth embodiments described above may be combined. For example, the representative character resolution instruction module 901 of the third embodiment may be combined with the second embodiment.

The program described above may be provided by being stored in a recording medium, or the program may be provided by communication means. In that case, for example, the above-described program may be regarded as an invention of a “computer-readable recording medium recording the program”.
The “computer-readable recording medium on which a program is recorded” refers to a computer-readable recording medium on which a program is recorded, which is used for program installation, execution, program distribution, and the like.
The recording medium is, for example, a digital versatile disc (DVD), which is a standard established by the DVD Forum, such as “DVD-R, DVD-RW, DVD-RAM,” and DVD + RW. Standards such as “DVD + R, DVD + RW, etc.”, compact discs (CDs), read-only memory (CD-ROM), CD recordable (CD-R), CD rewritable (CD-RW), etc. MO), flexible disk (FD), magnetic tape, hard disk, read only memory (ROM), electrically erasable and rewritable read only memory (EEPROM), flash memory, random access memory (RAM), etc. It is.
The program or a part of the program may be recorded on the recording medium for storage or distribution. Also, by communication, for example, a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a wired network used for the Internet, an intranet, an extranet, etc., or wireless communication It may be transmitted using a transmission medium such as a network or a combination of these, or may be carried on a carrier wave.
Furthermore, the program may be a part of another program, or may be recorded on a recording medium together with a separate program. Moreover, it may be divided and recorded on a plurality of recording media. Further, it may be recorded in any manner as long as it can be restored, such as compression or encryption.

It is a conceptual module block diagram about the structural example of 1st Embodiment. It is explanatory drawing which shows the example of representative character image data. It is a flowchart which shows the process example by 1st Embodiment. It is explanatory drawing which shows the process example by a character cut-out module, a similar character image data extraction module, a character image data gravity center calculation module, and a character image data structure module. It is explanatory drawing which shows the example of a more detailed process by a character image data gravity center calculation module and a character image data structure module. It is explanatory drawing which shows the process example by the character image data gravity center calculation module including a character group image data storage module, and a character image data structure module. It is a conceptual module block diagram about the structural example of 2nd Embodiment. It is a flowchart which shows the process example by 2nd Embodiment. It is a conceptual module block diagram about the structural example of 3rd Embodiment. It is explanatory drawing which shows the example of representative character image data. It is a flowchart which shows the process example by 3rd Embodiment. It is a notional module block diagram about the structural example of 4th Embodiment. It is explanatory drawing which shows the example of representative character image data. It is a flowchart which shows the process example by 4th Embodiment. It is a notional module block diagram about the structural example of 5th Embodiment. It is a flowchart which shows the process example by 5th Embodiment. It is a notional module block diagram about the structural example of 6th Embodiment. It is a flowchart which shows the process example by 6th Embodiment. It is a notional module block diagram about the structural example of 7th Embodiment. It is a flowchart which shows the process example by 7th Embodiment. It is a notional module block diagram about the structural example of 8th Embodiment. It is a flowchart which shows the process example by 8th Embodiment. It is a flowchart which shows the example of the deletion process of the representative character by 8th Embodiment. It is a block diagram which shows the hardware structural example of the computer which implement | achieves 1st-8th embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 102 ... Character cutout module 104 ... Similar character image data extraction module 106 ... Character image data gravity center calculation module 108 ... Character image data structure module 110 ... Character group image data storage module 112 ... Character group image data reduction module 701 ... Character group image Data use instruction module 901 ... representative character resolution instruction module 1201 ... representative character image data rotation module 1501 ... representative character image data storage instruction module 1701 ... data output / storage / erase control module 2101 ... frequency storage / data erasure control module 2102 ... character Group image data and frequency storage module

Claims (9)

Pixel block cutting means for cutting out a pixel block from an image;
Similar pixel block extraction means for extracting similar ones of the pixel blocks cut out by the pixel block cutting means;
Centroid calculating means for calculating centroids of a plurality of similar pixel chunks extracted by the similar pixel chunk extracting means;
Representative pixel block generation means for generating a representative pixel block of the pixel block based on a plurality of similar pixel blocks extracted by the similar pixel block extraction unit and the centroid calculated by the centroid calculation unit;
Comprising representative pixel block storage means for storing the representative pixel block generated by the representative pixel block generation means;
The similar pixel block extraction unit includes the representative pixel block stored by the representative pixel block storage unit in the pixel block extracted by the pixel block extraction unit, and extracts similar pixel blocks. Image processing device. The similar pixel block extraction unit counts the number of similar pixel blocks extracted by the pixel block extraction unit,
Use instruction means for instructing the similar pixel block extraction unit to use the representative pixel block stored in the representative pixel block storage unit based on the number counted by the similar pixel block extraction unit; The image processing apparatus according to claim 1. Resolution instruction means for instructing the resolution of the representative pixel block generated by the representative pixel block generation unit;
The image processing apparatus according to claim 1, wherein the representative pixel block generation unit generates the representative pixel block according to the resolution instructed by the resolution instruction unit. A rotation unit for rotating the representative pixel block generated by the representative pixel block generation unit;
The image processing apparatus according to claim 1, wherein the representative pixel block storage unit stores the representative pixel block rotated by the rotation unit. The similar pixel block extraction unit counts the number of similar pixel blocks extracted by the pixel block extraction unit,
5. A storage instructing unit for instructing the representative pixel block storage unit to store the representative pixel block based on the number counted by the similar pixel block extracting unit. The image processing apparatus according to any one of the above. Accepts instructions for registration, output or deletion by the operator,
If the instruction is registration, the center of gravity calculation unit, the representative pixel block generation unit, and the representative pixel block storage unit perform processing,
If the instruction is an output, let the center of gravity calculation means and the representative pixel block generation means perform processing,
6. The apparatus according to claim 1, further comprising: a control unit that controls to erase the representative pixel block stored in the representative pixel block storage unit when the instruction is erasure. An image processing apparatus according to 1. When the representative pixel block storage unit stores the representative pixel block generated by the representative pixel block generation unit, the representative pixel block storage unit erases the representative pixel block used when the representative pixel block generation unit generates the representative pixel block. The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. The image processing apparatus according to any one of claims 1 to 7, wherein the representative pixel block storage unit stores the number of times the stored representative pixel block is extracted by the similar pixel block extraction unit. . Computer
Pixel block cutting means for cutting out a pixel block from an image;
Similar pixel block extraction means for extracting similar ones of the pixel blocks cut out by the pixel block cutting means;
Centroid calculating means for calculating centroids of a plurality of similar pixel chunks extracted by the similar pixel chunk extracting means;
Representative pixel block generation means for generating a representative pixel block of the pixel block based on a plurality of similar pixel blocks extracted by the similar pixel block extraction unit and the centroid calculated by the centroid calculation unit;
Function as a representative pixel block storage unit that stores the representative pixel block generated by the representative pixel block generation unit;
The similar pixel block extraction unit includes the representative pixel block stored by the representative pixel block storage unit in the pixel block extracted by the pixel block extraction unit, and extracts similar pixel blocks. Image processing program.

Images