JP4973536B2 - Image processing apparatus and image processing program - Google Patents

Description

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

There is a character recognition device that recognizes characters in an image.
The processing by the character recognition device accepts an image, cuts out a line from the image, cuts out a character in the line, performs character recognition processing for each cut out character, and finally outputs a character code.

  As a technology related to this, for example, Patent Document 1 has an object to improve the character cutout accuracy of proportionally printed English documents, Japanese-English mixed documents, and the like. Are grouped, and a parameter between groups is calculated. Using this parameter, the possibility of integration of the image in the target group and the image in the reference group is determined. Based on the determination result, a single character image is extracted, and character extraction is performed. This section discloses that an image determined to be a single character image is cut out as a character, and a character is cut out by performing necessary integration / separation with respect to other images.

Further, for example, Patent Document 2 uses a method of extracting a rectangle having the same shape as the calculated certainty factor, and improves matching efficiency at the time of character recognition by caching that temporarily stores and uses an image. Characters with high confidence up to the previous line when the input character image is separated into line units and the character recognition method that recognizes the characters in line order is repeated for the number of lines. It is disclosed that an image is stored and the same character image is searched for a character image with high reliability of the recognition result.
JP 09-297816 A JP 2000-090202 A

  The present invention relates to a case where a plurality of character images in an image are similar, one of the character images is recognized, and the other character image is not recognized. It is an object of the present invention to provide an image processing apparatus and an image processing program that can give a character recognition result to an image.

The gist of the present invention for achieving the object lies in the inventions of the following items.
According to the first aspect of the present invention, there is provided a character image removing unit that removes a first character image that has been recognized in the image from the image, and a second image that has been removed from the first character image by the character image removing unit. A character image extracting means for extracting the character image, and comparing the first character image already recognized by the character with the second character image extracted by the character image extracting means . When the second character image is similar , the image processing apparatus includes a recognition result adding unit that adds a character recognition result of the first character image to the second character image. .

  The invention according to claim 2 further comprises recognition result output means for recognizing a character image in the image and outputting the character image and the character recognition result in correspondence with each other, and the first character used by the character image removal means. The image is a character image output by the recognition result output unit, and the character recognition result of the first character image used by the recognition result adding unit is corresponding to the first character image by the recognition result output unit. The image processing apparatus according to claim 1, wherein the image processing apparatus outputs the character recognition result.

According to a third aspect of the present invention, a black pixel component extracting unit that extracts a black pixel component in an image and the black pixel component included in a character rectangle surrounding a character image that has already been recognized in the image are used. A character image generation recognition result adding unit that generates the character image and assigns the character recognition result to the character image, and a second character using a black pixel component that has not been used by the character image generation recognition result addition unit. The first character image extracted by the character image extraction means for extracting an image, the character image generation recognition result assigning means, and the second character image extracted by the character image extraction means are compared, and the first character image is extracted . When the character image of the second character image is similar to the second character image, there is provided a recognition result adding means for adding the character recognition result of the first character image to the second character image. An image processing apparatus.

  The invention according to claim 4 further comprises recognition result output means for recognizing a character image in the image, and outputting a character rectangle surrounding the character image and a character recognition result in association with each other, and adding the character image generation recognition result The character rectangle used by the means is a character rectangle output by the recognition result output means, and the character recognition result of the first character image used by the recognition result giving means corresponds to the character rectangle of the first character image. The image processing apparatus according to claim 3, wherein the image recognition result is output by the recognition result output unit.

According to a fifth aspect of the present invention, there is provided a computer, a character image removing unit that removes a character-recognized first character image from the image, and an image from which the first character image has been removed by the character image removing unit. A character image extracting means for extracting a second character image from the image, and comparing the first character image recognized by the character with the second character image extracted by the character image extracting means . When the image and the second character image are similar to each other , the image processing is made to function as a recognition result adding unit that adds a character recognition result of the first character image to the second character image. It is a program.

According to a sixth aspect of the present invention, a computer uses a black pixel component extracting means for extracting a black pixel component in an image, and the black pixel component included in a character rectangle surrounding a character image that has already been recognized in the image. The first character image is generated, and the character image generation recognition result adding unit for adding the character recognition result to the character image and the black pixel component not used by the character image generation recognition result adding unit are used. Comparing the first character image extracted by the character image extraction means for extracting the second character image, the second character image extracted by the character image extraction means, and the second character image extracted by the character image extraction means; When the first character image is similar to the second character image, the second character image is caused to function as a recognition result adding unit that adds a character recognition result of the first character image. Features Is an image processing program.

  According to the image processing apparatus of claim 1, a plurality of character images in the image are similar, one of the character images is recognized, and the other character image is not recognized. The character recognition result can be given to the unrecognized character image.

  According to the image processing apparatus of the second aspect, it is possible to cope with a character recognition process in which a character image and its character recognition result are output in association with each other.

  According to the image processing device of claim 3, a plurality of character images in the image are similar, one of the character images is recognized, and the other character image is not recognized. The character recognition result can be given to the unrecognized character image.

  According to the image processing apparatus of the fourth aspect, it is possible to cope with the character recognition processing in which the character rectangle surrounding the character image and the character recognition result are output in association with each other.

  According to the image processing program of claim 5, a plurality of character images in the image are similar, one of the character images is recognized, and the other character image is not recognized. The character recognition result can be given to the unrecognized character image.

  According to the image processing program of claim 6, a plurality of character images in the images are similar, one of the character images is recognized, and the other character image is not recognized. The character recognition result can be given to the unrecognized character image.

First, an outline of the present embodiment will be described.
Character recognition is performed on an image, particularly an image targeted for a document. In addition, a plurality of similar (including the same) characters may appear in the document image. However, some of the similar characters can be recognized and some cannot be recognized.
As a factor, there are a case where “blurred” or “collapsed” or the like has occurred in a character image, or a document with an irregular layout, which makes it difficult to cut out lines.
In particular, the latter case will be described. The character recognition process is performed in the order of three processes, “line cutout”, “character cutout”, and “character recognition”, and each process handles only the result of the preprocessing. In this way, a character code is not assigned to a character image not included in a line. For example, a character image close to or included in a graphic, or a character image at the beginning or end of a line (particularly, symbols such as punctuation marks, parentheses, and middle black) may be determined as a graphic and may not be included in the line.

  The present embodiment mainly uses such character recognition processing (character recognition processing performed in the order of line cutout, character cutout, and character recognition). Then, it is determined whether or not the character-recognized image is similar to the character-recognized character image, and the character code of the character-recognized character image is assigned to the similar image. Is.

Hereinafter, examples of various preferred embodiments for realizing the present invention will be described with reference to the drawings.
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. Hereinafter, “connection” is used not only for physical connection but also for 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.

  In the present embodiment, as shown in FIG. 1, the image receiving module 110, the line cutting module 120, the character cutting module 130, the character recognition module 140, the recognized character image removal module 150, the character image extraction module 160, the pattern A matching processing module 170 and a character recognition result output module 180 are provided.

  The image reception module 110 is connected to the line cutout module 120 and the recognized character image removal module 150, receives an image, and passes the image to the line cutout module 120 and the recognized character image removal module 150. Accepting an image includes reading an image with a scanner, receiving an image with a fax, reading an image from an image database, and the like. The image includes a character image. The image may be a single image or a plurality of images. Further, the content of the image may be a document used for business, a pamphlet for advertisement, or the like. The image may be a multi-valued or color image other than the binary image. However, when the image is other than the binary image, it is desirable to perform binary image processing.

  The line cut-out module 120 is connected to the image reception module 110 and the character cut-out module 130, receives an image from the image reception module 110, cuts out a line composed of characters therein, and cuts out the cut-out line. It is passed to the character extraction module 130. The rows include horizontal writing rows and vertical writing rows (columns). As a method of cutting out rows, for example, there is a method of taking a projection (frequency distribution of black pixels) in the horizontal direction (or vertical direction) of an image and cutting out rows at the valleys.

  The character cutout module 130 is connected to the line cutout module 120 and the character recognition module 140, receives lines from the line cutout module 120, cuts out character images included in the line one by one, and cuts out the character images. The received character image is passed to the character recognition module 140. As a method for extracting a character image, for example, there is a method of taking a projection in a vertical direction (or a horizontal direction in a column) within a row and cutting out a character at a valley portion. Moreover, the method disclosed in Japanese Patent Laid-Open No. 09-274645 may be used, or a plurality of clipping methods may be combined.

  The character recognition module 140 is connected to the character cutout module 130, the recognized character image removal module 150, and the pattern matching processing module 170, receives a character image from the character cutout module 130, and performs character recognition of the character image. . That is, the character image is converted into the character code. Then, the character image is transferred to the recognized character image removal module 150, and a character code that is a character recognition result is assigned to the character image and transferred to the pattern matching processing module 170.

  The recognized character image removal module 150 is connected to the image reception module 110, the character recognition module 140, and the character image extraction module 160. The recognized character image removal module 150 receives images from the image reception module 110 and character recognition images from the character recognition module 140. The character image that has been recognized is removed from the image. That is, the character image cut out as a line by the line cut-out module 120, cut out as a character by the character cut-out module 130, and character-recognized by the character recognition module 140 is removed from the original image. As a method of removing, for example, an XOR (eXclusive OR) operation may be performed on a character image and an image at that position in the image. As a result of the processing by the recognized character image removal module 150, only an image that has not been character-recognized remains in the image. This image is also called a residual image. The residual image is passed to the character image extraction module 160.

  The character image extraction module 160 is connected to the recognized character image removal module 150 and the pattern matching processing module 170, receives a residual image from the recognized character image removal module 150, extracts a character image from the residual image, The extracted character image is passed to the pattern matching processing module 170. The character image extraction processing by the character image extraction module 160 may be the same processing as that of the character cutout module 130, may be a technique disclosed in Patent Document 1, and further using FIGS. Processing described later may be used.

  The pattern matching processing module 170 is connected to the character recognition module 140, the character image extraction module 160, and the character recognition result output module 180, and character images recognized by the character recognition module 140 (character images to which character codes have been assigned). , Hereinafter also referred to as a recognized character image), a character image obtained by extracting a character code corresponding to the recognized character image from the residual image from the character image extraction module 160 (a character image to which no character code is assigned, hereinafter, an unrecognized character). Pattern matching processing between the unrecognized character image and the recognized character image, and if it is determined that they are similar, the character code of the recognized character image is assigned to the unrecognized character image. Character images (both recognized character images and unrecognized character images with character codes) Free), and passes a character code corresponding to the character image to the character recognition result output module 180. More details will be described later with reference to FIG.

  The character recognition result output module 180 is connected to the pattern matching processing module 170, receives a character image and a character code corresponding to the character image from the pattern matching processing module 170, and externally (others such as a storage device and a document processing application). To the information processing system connected via the communication line). That is, more character recognition results than the character recognition results by the character recognition module 140 (which may be the same) are output in the same format as the character recognition module 140.

FIG. 2 is a flowchart illustrating a processing example according to the first exemplary embodiment.
In step S202, the image receiving module 110 receives a target image.
In step S204, the line cutout module 120 analyzes the image received in step S202 and cuts out a line.
In step S206, the character cutout module 130 cuts out the character image in the line cut out in step S204.
In step S208, the character recognition module 140 performs character recognition on the character image cut out in step S206.

In step S210, the recognized character image removal module 150 generates a residual image by removing the character image recognized in step S208 from the image received in step S202.
In step S212, the character image extraction module 160 extracts a character image from the residual image generated in step S210.
In step S214, the pattern matching processing module 170 performs pattern matching processing between the character image extracted in step S212 (unrecognized character image) and the character image recognized in step S208 (recognized character image), and the recognized character image. The character code of the recognized character image is given to the non-recognized character image similar to.
In step S216, the character recognition result output module 180 outputs character images (recognized character image, non-recognized character image similar to the recognized character image) and their character codes.

3 and 4 are explanatory diagrams illustrating an example of processing in which the character image extraction module 160 generates a character image using a circumscribed rectangle.
The character image extraction module 160 extracts 8-neighbor concatenation or 4-neighbor concatenation black pixel components from the residual image. For example, the example shown in FIG. 3A includes two black pixel components.
Next, a circumscribed rectangle of the extracted black pixel component (a rectangle that is in contact with and surrounds the black pixel components at the highest, lowest, rightmost, and leftmost black pixels) is generated. For example, FIG. 3B shows an example in which two circumscribed rectangles (310, 320) are generated for the image shown in the example of FIG. If the size of the circumscribed rectangle is equal to or greater than a predetermined threshold, it is determined that the character image is not a character image.

The character image extraction module 160 takes out the overlapping circumscribed rectangles and integrates them as one circumscribed rectangle. For example, FIG. 3C shows one circumscribed rectangle from the two circumscribed rectangles shown in FIG. 3B.
Then, an image in the integrated circumscribed rectangle is set as a character image.

Further, the character image extraction module 160 may further perform an expansion process (a process of enlarging outside) on the circumscribed rectangle.
For example, the example shown in FIG. 4A includes two black pixel components. The circumscribed rectangles (410, 420) do not overlap as shown in the example of FIG.
The character image extraction module 160 performs expansion processing on each of the circumscribed rectangles 410 and 420 to generate expanded rectangles 430 and 440 as shown in the example of FIG. Take out the overlapped circumscribed rectangle with the expansion process.
Then, the circumscribed rectangles having the overlap are integrated as one circumscribed rectangle. For example, FIG. 4D shows one circumscribed rectangle from the two circumscribed rectangles shown in FIG. 4C.
Then, an image in the integrated circumscribed rectangle is set as a character image.

FIG. 5 is an explanatory diagram showing a processing example of the pattern matching processing module 170.
The pattern matching processing module 170 receives the recognized character image 510 and its character code “4” from the character recognition module 140 and receives the unrecognized character image 520 from the character image extraction module 160. Pattern matching processing is performed on the recognized character image 510 and the unrecognized character image 520.

For example, the following processing is performed.
(1) The centers of black pixels of two character images (recognized character image 510 and non-recognized character image 520) are matched.
(2) An XOR operation is performed on two character images having the same center of gravity.
(3) The number of pixels (different pixels) that become 1 as a result of the XOR operation is counted. If this number is less than or equal to the threshold, it is determined that the character images are similar.
(4) Alternatively, for unrecognized character images, the number of pixels that differ from all recognized character images may be counted, and the smallest may be determined to be a similar character image.
Further, a technique disclosed in Japanese Patent Application Laid-Open No. 07-200755 may be used.

  Then, the character code “4” corresponding to the character image received from the character recognition module 140 is added to the recognized character image 510 and output. The character code “4” corresponding to the similar recognized character image 510 is assigned to the unrecognized character image 520 and output.

FIG. 6 is an explanatory diagram showing an example of processing when the recognition result is used for vertically written characters when the recognition of horizontally written characters is successful.
The example shown in FIG. 6A is a case where there is a horizontally written character “19” in a vertically written line, and character segmentation often fails. The example shown in FIG. 6B is a character image in the same image, and it is assumed that the character recognition is successful. Therefore, the character codes “1”, “9”, “8”, “4”, “year” are assigned to the character images “1”, “9”, “8”, “4”, “year”. Has been.
In the example shown in FIG. 6A, the recognized character image “hour” and “ni” are removed by the recognized character image removal module 150, and the unrecognized characters “1” and “9” are removed by the character image extraction module 160. An image is extracted. Then, the character code “1” is assigned to the unrecognized character image “1” and the character code “9” is assigned to the unrecognized character image “9” by the pattern matching processing module 170.

FIG. 7 is an explanatory diagram showing an example in which processing by the line cutout module 120 fails and an example in which a character code is given by the pattern matching processing module 170.
The example shown in FIG. 7A is a part of an image received by the image receiving module 110. As described above, when there is a character image (black pixel components 711 to 715) in the vicinity of the black pixel component 710 that is a graphic, the line cutout module 120 often extracts the black pixel components 712 to 715 as rows. That is, the example shown in FIG. 7B is an image (black pixel component 710 and black pixel component 711) that is not cut out as a row by the row cutout module 120, and is a residual image by the recognized character image removal module 150. is there. The example shown in FIG. 7C is an image cut out as a line, which is a recognized character image.
On the other hand, if there are character images (black pixel components 721 to 726) as shown in FIG. 7D in the same image, it is assumed that these characters have been successfully recognized. That is, the character code “·” is given to the black pixel component 724 “·” by the character recognition module 140. As shown in FIG. 7E, the character image extraction module 160 assumes that the black pixel component 710 whose circumscribed rectangle is equal to or larger than a predetermined size is not a character, and extracts the black pixel component 711 as an unrecognized character image. The pattern matching processing module 170 determines that the black pixel component 711 is similar to the black pixel component 724 that is a character image, and assigns the character code “·” to the black pixel component 711 as well.

FIG. 8 is a conceptual module configuration diagram of a configuration example according to the second embodiment.
In the first embodiment, the character image and its character code are output as the character recognition process. In the second embodiment, the character rectangle position information (for example, surrounding the character image is enclosed) as the character recognition process. This corresponds to the case of outputting the upper left coordinates of the rectangle and the vertical and horizontal lengths).
In addition, the same code | symbol is attached | subjected to the site | part of the same kind as 1st Embodiment, and the overlapping description is abbreviate | omitted.

  In the second embodiment, as shown in FIG. 8, an image receiving module 110, a line cutting module 120, a character cutting module 130, a character recognition module 840, a connected component extraction module 850, an integration module 860, a character image extraction. A module 870, a pattern matching processing module 880, and a character recognition result output module 890;

  The character recognition module 840 is connected to the character cutout module 130 and the integration module 860, receives a character image from the character cutout module 130, and performs character recognition of the character image. That is, the character image is converted into the character code. Then, the position information of the character rectangle surrounding the character image and the character code which is the character recognition result of the character image in the character rectangle are given and passed to the integration module 860.

  The connected component extraction module 850 is connected to the image reception module 110 and the integration module 860, receives an image from the image reception module 110, extracts black pixel components therein, and extracts the extracted black pixel components to the integration module 860. hand over.

  The integration module 860 is connected to the character recognition module 840, the connected component extraction module 850, the character image extraction module 870, and the pattern matching processing module 880, and the character rectangle position information from the character recognition module 840 and the characters in the character rectangle A character code that is a character recognition result of the image is received, a black pixel component in the image is received from the connected component extraction module 850, and the black pixel component included in the character rectangle is integrated to form a character image (hereinafter, a recognized character image). (Also called). Then, a character code corresponding to the recognized character image is assigned, and the recognized character image and the character code corresponding to the recognized character image are passed to the pattern matching processing module 880. On the other hand, the black pixel component not included in the character rectangle is passed to the character image extraction module 870. Note that an image composed of black pixel components not included in the character rectangle corresponds to the residual image in the first embodiment, and is an image obtained by removing the character image. More details will be described later with reference to FIG.

  The character image extraction module 870 is connected to the integration module 860 and the pattern matching processing module 880, and the black pixel component that is not used by the integration module 860 from the integration module 860, that is, the black pixel component that is not included in the character rectangle. The character image is extracted from the black pixel component, and the character image (hereinafter also referred to as an unrecognized character image) is passed to the pattern matching processing module 880. The extraction method of the unrecognized character image is the same as that of the character image extraction module 160 except that the black pixel component is used. Further, after the black pixel component is converted into an image to generate a residual image, the same processing as that of the character image extraction module 160 may be performed.

  The pattern matching processing module 880 is connected to the integration module 860, the character image extraction module 870, and the character recognition result output module 890. The pattern matching processing module 880 receives the recognized character image and the character code corresponding to the recognized character image from the integration module 860, A non-recognized character image is received from the image extraction module 870, and pattern matching processing between the non-recognized character image and the recognized character image is performed. As a result, when it is determined that they are similar, Is given the character code. Then, a character rectangle of a character image (including both a recognized character image and a non-recognized character image to which a character code is assigned) is generated. The position information of the generated character rectangle and the character code corresponding to the character rectangle are passed to the character recognition result output module 890. Note that the position information of the character rectangle output from the character recognition module 840 may be used as the position information of the character rectangle of the recognized character image.

  The character recognition result output module 890 is connected to the pattern matching processing module 880. The character recognition result output module 890 receives the position information of the character rectangle and the character code corresponding to the character rectangle from the pattern matching processing module 880, and receives the external (storage device, document processing application). To other processing such as an information processing system connected via a communication line). That is, more character recognition results than the character recognition results by the character recognition module 840 (which may be the same) are output in the same format as the character recognition module 840.

FIG. 9 is a flowchart illustrating a processing example according to the second exemplary embodiment.
In step S902, the image reception module 110 receives a target image.
In step S904, the line cutout module 120 analyzes the image received in step S902 and cuts out a line.
In step S906, the character cutout module 130 cuts out the character image in the line cut out in step S904.
In step S908, the character recognition module 840 performs character recognition on the character image cut out in step S906.

In step S910, the connected component extraction module 850 extracts the black pixel component in the image received in step S902.
In step S912, the integration module 860 integrates and recognizes the black pixel component included in the character rectangle using the character rectangle of the character image recognized in step S908 and the black pixel component extracted in step S910. A character image is extracted, and black pixel components that are not included are extracted.
In step S914, the character image extraction module 870 extracts an unrecognized character image using the non-included black pixel component extracted in step S912.
In step S916, the pattern matching processing module 880 performs a pattern matching process between the unrecognized character image extracted in step S914 and the recognized character image extracted in step S912, and generates a non-recognized character image similar to the recognized character image. The character code of the recognized character image is assigned. Then, the position information of the character rectangle of the character image to which the character code is assigned is generated.
In step S918, the character recognition result output module 890 determines the position information of the character rectangle (character rectangle of the recognized character image, character rectangle of the unrecognized character image similar to the recognized character image) generated in step S916, the character Outputs the character code corresponding to the rectangle.

FIG. 10 is an explanatory diagram showing a processing example by the integration module 860.
The integration module 860 receives the position information of the character rectangle from the character recognition module 840 and integrates the black pixel components included therein to generate a character image.
The inclusion determination here includes, for example, a case where a circumscribed rectangle of the black pixel component is included in a character rectangle. This will be described with reference to FIG. The position information of the character rectangle 1010 passed from the character recognition module 840 is an upper left position 1011, a width 1012, and a height 1013 as shown in the example of FIG. As shown in the example of FIG. 10B, the circumscribed rectangle 1021 and circumscribed rectangle 1022 of the black pixel component are included in the character rectangle 1010.
Alternatively, when the area ratio of the black pixel component in the character rectangle is equal to or greater than a predetermined threshold, the character rectangle may be included in the character rectangle.

  A hardware configuration example of the present embodiment will be described with reference to FIG. The configuration shown in FIG. 11 is configured by a personal computer (PC), for example, and shows a hardware configuration example including a data reading unit 1117 such as a scanner and a data output unit 1118 such as a printer.

  A CPU (Central Processing Unit) 1101 includes various modules described in the above-described embodiments, that is, a line extraction module 120, a character extraction module 130, a character recognition module 140, a recognized character image removal module 150, a character image. The control unit executes processing according to a computer program describing an execution sequence of each module such as the extraction module 160, the pattern matching processing module 170, the connected component extraction module 850, and the integration module 860.

  A ROM (Read Only Memory) 1102 stores programs used by the CPU 1101, calculation parameters, and the like. A RAM (Random Access Memory) 1103 stores programs used in the execution of the CPU 1101, parameters that change as appropriate during the execution, and the like. These are connected to each other by a host bus 1104 including a CPU bus.

  The host bus 1104 is connected to an external bus 1106 such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge 1105.

  A keyboard 1108 and a pointing device 1109 such as a mouse are input devices operated by an operator. The display 1110 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) 1111 includes a hard disk, drives the hard disk, and records or reproduces a program executed by the CPU 1101 and information. The hard disk stores accepted images and character recognition results. Further, various computer programs such as various other data processing programs are stored.

  The drive 1112 reads out data or a program recorded in a removable recording medium 1113 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and the data or program is read out as an interface 1107 and an external bus 1106. , A bridge 1105, and a RAM 1103 connected via a host bus 1104. The removable recording medium 1113 can also be used as a data recording area similar to the hard disk.

  The connection port 1114 is a port for connecting the external connection device 1115 and has a connection unit such as USB or IEEE1394. The connection port 1114 is connected to the CPU 1101 and the like via the interface 1107, the external bus 1106, the bridge 1105, the host bus 1104, and the like. The communication unit 1116 is connected to the network and executes data communication processing with the outside. The data reading unit 1117 is a scanner, for example, and executes document reading processing. The data output unit 1118 is a printer, for example, and executes document data output processing.

  Note that the hardware configuration illustrated in FIG. 11 illustrates one configuration example, and the present embodiment is not limited to the configuration illustrated in FIG. 11, and is a configuration capable of executing the modules described in the present embodiment. I just need it. 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. 11 may be connected to each other via a communication line 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 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 a flowchart which shows the process example by 1st Embodiment. It is explanatory drawing which shows the process example which produces | generates a character image using a circumscribed rectangle. It is explanatory drawing which shows the process example which produces | generates a character image using a circumscribed rectangle. It is explanatory drawing which shows the process example of a pattern matching process module. It is explanatory drawing which shows the example of a process in the case of using the recognition result for a vertical writing character, when recognition of a horizontal writing character is successful. It is explanatory drawing which shows the example in which the process by the line extraction module failed, and the example to which the character code was given by the pattern matching process 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 explanatory drawing which shows the process example by an integrated module. It is a block diagram which shows the hardware structural example of the computer which implement | achieves this Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 110 ... Image reception module 120 ... Line extraction module 130 ... Character extraction module 140 ... Character recognition module 150 ... Recognized character image removal module 160 ... Character image extraction module 170 ... Pattern matching processing module 180 ... Character recognition result output module 840 ... Character recognition module 850 ... Connected component extraction module 860 ... Integration module 870 ... Character image extraction module 880 ... Pattern matching processing module 890 ... Character recognition result output module

Claims (6)

A character image removing means for removing a first character image that has been recognized in the image from the image;
A character image extracting means for extracting a second character image from the image from which the first character image has been removed by the character image removing means;
When the first character image recognized by the character is compared with the second character image extracted by the character image extraction means, and the first character image and the second character image are similar an image processing apparatus characterized by comprising a recognition result providing means for providing a character recognition result of the first character image in the second character image. A recognition result output means for recognizing the character image in the image and outputting the character image and the character recognition result in correspondence with each other;
The first character image used by the character image removing unit is a character image output by the recognition result output unit,
The character recognition result of the first character image used by the recognition result adding unit is a character recognition result output by the recognition result output unit corresponding to the first character image. An image processing apparatus according to 1. Black pixel component extraction means for extracting a black pixel component in the image;
A character image generation recognition result for generating a first character image using the black pixel component included in a character rectangle surrounding a character image that has already been recognized in the image, and adding the character recognition result to the character image Granting means;
A character image extraction means for extracting a second character image using a black pixel component that has not been used by the character image generation recognition result giving means;
The first character image extracted by the character image generation recognition result assigning means and the second character image extracted by the character image extracting means are compared, and the first character image and the second character image are compared . An image processing apparatus comprising: a recognition result giving unit that gives a character recognition result of the first character image to the second character image when the two are similar to each other . A recognition result output means for recognizing the character image in the image and outputting the character rectangle surrounding the character image and the character recognition result in correspondence with each other;
The character rectangle used by the character image generation recognition result assigning unit is a character rectangle output by the recognition result output unit,
The character recognition result of the first character image used by the recognition result giving means is a character recognition result output by the recognition result output means corresponding to the character rectangle of the first character image. The image processing apparatus according to claim 3. Computer
A character image removing means for removing a first character image that has been recognized in the image from the image;
A character image extracting means for extracting a second character image from the image from which the first character image has been removed by the character image removing means;
When the first character image recognized by the character is compared with the second character image extracted by the character image extraction means, and the first character image and the second character image are similar an image processing program for causing to function as a recognition result providing means for providing a character recognition result of the first character image in the second character image. Computer
Black pixel component extraction means for extracting a black pixel component in the image;
A character image generation recognition result for generating a first character image using the black pixel component included in a character rectangle surrounding a character image that has already been recognized in the image, and adding the character recognition result to the character image Granting means;
A character image extraction means for extracting a second character image using a black pixel component that has not been used by the character image generation recognition result giving means;
The first character image extracted by the character image generation recognition result assigning means and the second character image extracted by the character image extracting means are compared, and the first character image and the second character image are compared . When the two are similar , an image processing program causing the second character image to function as a recognition result adding unit that adds a character recognition result of the first character image.

Images