JPH11224306A - Device and method for recognizing character - Google Patents

Abstract

PROBLEM TO BE SOLVED: To exactly perform character recognition flexibly corresponding to the difference of a document format when performing the character recognition concerning respective items in a document. SOLUTION: When an image is inputted in S201, that input image is divided into plural block areas based on color information in area identifying processing S202. Based on the color information in the respective block areas provided by the area identifying processing, item information is set to the respective block areas in item information converting processing S203. Based on the item information set by the item information converting processing, a character recognition object type is limited for each block in recognition setting selecting processing S204. Concerning the image in each area block, character recognizing processing is performed in character recognizing processing $205 within the range of the character recognition object type set in the recognition setting selecting processing S204.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a character recognition apparatus, and more particularly to a character recognition apparatus and method for performing character recognition on a document such as a form having a predetermined format.

[0002]

2. Description of the Related Art In a conventional character recognition apparatus, there is an apparatus which performs character recognition by reading an image surrounded by ruled lines such as a form. In this type of apparatus, the format of a document to be read is limited to one format, and only a specific location (that is, a specific item) of the document is recognized.

In order to correspond to a format having different coordinates of a character area, a method has been proposed in which a ruled line is extracted and a portion surrounded by the ruled line is recognized.

[0004]

However, the technique of performing character recognition at a specific location in a document as described above cannot support a plurality of types of formats, and cannot read an item to be read when an image is read by a scanner or the like. Is required to prevent the position from being shifted to a position other than the predetermined position.

In the technique of extracting a ruled line to specify a character recognition position, when processing a document having a ruled line such as a complicated table, even if the format is known in advance, the inclination, the blur, and the displacement of the position are determined. If there are factors such as differences in size, ruled lines cannot be extracted, or text or the like may be erroneously formed as ruled lines, which makes accurate character recognition difficult.

The present invention has been made in view of the above problems, and provides a character recognition apparatus and method capable of flexibly coping with differences in document formats when character recognition is performed for each item in a document. The purpose is to do.

Another object of the present invention is to enable more accurate character recognition by performing appropriate character recognition processing on each item in a document.

[0008]

The character recognition apparatus according to one aspect of the present invention that achieves the above object has, for example, the following configuration. That is, an area dividing unit that divides an image into a plurality of areas based on color information, an attribute setting unit that sets a data attribute to each area based on color information of each area obtained by the area dividing unit, Recognition processing setting means for setting processing conditions of character recognition processing for each of the regions based on the data attribute set by the setting means;
A character recognizing unit that performs a character recognizing process on the image of each area under the processing conditions set by the recognition process setting unit.

A character recognition method according to another embodiment of the present invention for achieving the above object includes, for example, the following steps. That is, an area dividing step of dividing an image into a plurality of areas based on color information, an attribute setting step of setting a data attribute in each area based on the color information of each area obtained in the area dividing step, A recognition processing setting step of setting a processing condition of character recognition processing for each of the areas based on the data attribute set in the setting step; and a processing condition set in the recognition processing setting step. A character recognition step of performing character recognition processing on the image.

[0010]

Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 is a block diagram showing the configuration of the character recognition device of the present embodiment. In FIG. 1, reference numeral 101 denotes a scanner which irradiates an image original with light and reads the reflected light and converts the read light into an electric signal; and 102, converts the electric signal obtained by the scanner 101 into a binary digital electric signal and converts other devices A scanner interface circuit for transmitting to the element. Reference numeral 103 denotes a pointing device (mouse in this example) for inputting desired coordinates on a window of the display, and 104 denotes an interface circuit for receiving a signal from the pointing device 103 and transmitting the signal to another device component. It is. A CPU 105 controls the entire apparatus and executes an item identification process and the like.
Reference numeral 6 denotes an RO storing control programs executed by the CPU 105, various processing programs, font data, and the like.
Reference numerals M and 107 denote R used as a work area for document image development processing and item identification processing by the CPU 105.
AM.

Reference numeral 108 denotes a display for displaying an input image or the like, and reference numeral 109 denotes a display interface circuit. The display 108 has a RAM 1
The image stored in the VRAM area which is the predetermined address area 07 is displayed. 110 is an external storage device such as a hard disk in which registered data is stored, and 111 is its interface. The external storage device 110 stores an item conversion database 110a and a target character type limitation database 110b.
The cholera database will be described later. And 112
Is a bus connecting each device component. In addition, CPU
Control program executed by the external storage device 110
May be stored in the RAM 107 and loaded into the RAM 107 as needed, and the CPU 105 may execute the program.

Next, the processing performed by the character recognition apparatus according to the present embodiment will be described with reference to the flowchart of FIG. 2, FIGS. 3, 4, 5, and 5.
This will be described with reference to FIG. FIG. 2 is a flowchart illustrating an operation procedure of the character recognition device according to the first embodiment. FIG. 3 is a diagram illustrating a structure example of block area data according to the present embodiment. FIG. 4 is a diagram illustrating an example of an item conversion table in the item conversion database 110a according to the present embodiment. FIG. 5 is a diagram illustrating an example of a document to be read by the character recognition device of the present embodiment. FIG. 6 is a diagram illustrating an example of the target character type table 110b of the target character type restriction database.

First, in step S201, a document to be recognized (for example, the document shown in FIG. 5) is read by the scanner 101 and converted into n-bit color image data. An area identification process is performed on the obtained input image data in step S202. In this area identification processing, a block area is extracted for each background color of the document, and its color information and layout information are stored as block data. In the present embodiment, as in the example shown in FIG. 3, “color” is used as color information, and “coordinates” and “attributes” are used as layout information.
“Assembly direction” is stored as block data.

Next, item information conversion processing (step S20)
Perform 3). In the item information conversion process, the insult storage device 11
The item information is obtained for each block area from the color information for each block area with reference to the item conversion table of the item conversion database 110a stored in “0”. In the case of the present embodiment, since the document to be recognized (FIG. 5) is “address book”, it is used for the item conversion database “address book” (FIG. 4).
Is used.

For example, since the color information of the block 502 in FIG. 5 is color D, it is converted into the item "address" by the item conversion database 110a (FIG. 4), and the data "4" is set in the item number of the block data. You. In addition,
In the present embodiment, the setting of the recognition processing order is performed simultaneously by setting the item number also in the recognition order of the block data. That is, the value of the item number becomes the value of the recognition order as it is.

Next, a recognition setting selection process (step S20)
In 4), a setting for limiting the character type to be subjected to character recognition is selected. In the case of this embodiment, the target character type limitation database 110b stored in the external storage device 110 has a target character type limitation table for each document format,
The target character type for character recognition is selected from the item numbers with reference to the target character type limitation table (FIG. 6) for each block data. For example, when the item number “3” is set in the block data, “numeric” is selected as the target character type for character recognition from the target character type limitation table (FIG. 6).

Next, in the character recognition processing (step S205), character recognition is performed for the character type selected as the target character type in each block area in the order of the recognition order of the block data, that is, in the order of the item numbers. .

As described above, according to the first embodiment, when processing is performed on a document such as a form having a different format, if the item information (correspondence between color and item) is the same, The character recognition processing can be performed in the same type of format.

Further, according to the first embodiment, since the character type to be recognized is set in each block area, the character recognition accuracy can be improved.

[Second Embodiment] In the first embodiment, the recognition target character type in the character recognition processing is limited in the recognition setting selection processing (step S204). However, the present invention is not limited to this. In the second embodiment, in the recognition setting selection process (step S204) of FIG. 2, a configuration is adopted in which a recognition dictionary used for character recognition is selected. That is, a recognition dictionary database having a recognition dictionary correspondence table as shown in FIG. 7 for each document format is prepared in the external storage device 110 in advance. Then, referring to the recognition dictionary correspondence table (FIG. 7) for each block area, a dictionary used for character recognition of each block area is selected from the item numbers set in each block area.

Next, the character recognition process of FIG. 2 (step S20)
In 5), the processing is executed in the order of the item numbers set in each block area as in the first embodiment. However, when performing the character recognition processing for each block area, the recognition setting selection processing (step In S204), the recognition dictionary set for each block area is used.

For example, for a block area in which item number "3" is set in the block data, "numerical dictionary" is selected from the recognition dictionary correspondence table (FIG. 7) as a dictionary used for character recognition (step). S204). Then, in the character recognition processing (step S205), the character recognition processing using the “numeric dictionary” is performed in the processing order determined by the item number (3) of the block area.

As described above, according to the second embodiment, since an appropriate dedicated dictionary can be used for each block area, the accuracy of character recognition can be improved.

[Third Embodiment] In the first and second embodiments, in the recognition setting selection processing, the characters to be recognized in the recognition processing are restricted according to the item numbers set in the respective blocks, and the recognition dictionary is not changed. Switching and controlling the recognition process itself. In the third embodiment, in the recognition setting selection process, a word collation dictionary used for correcting the character recognition result is selected.

FIG. 8 is a flowchart for explaining the operation procedure of the character recognition device according to the third embodiment. FIG. 9 is a diagram showing an example of a word matching dictionary correspondence table in the word matching dictionary database according to the third embodiment. In the third embodiment, a word matching dictionary database having a word matching dictionary table as shown in FIG. 9 for each document format is stored in the external storage device 110.

In FIG. 8, steps for performing the same processing as in FIG. 2 are denoted by the same step numbers, and description thereof is omitted here. In the recognition setting selecting process (step S904), the word matching dictionary used in the word matching process (step S906) is selected from the item numbers by referring to the word matching dictionary database (FIG. 8) for each block data. In step S905, character recognition processing is performed on the image in each block area in the order indicated by the item numbers. In step S906, the characters of the recognition result in each block area are subjected to word matching using the selected word matching dictionary in the order of the item numbers, and errors in the character recognition result are corrected.

For example, when "4" is set as the item number of the block data, "place name dictionary" is selected as a dictionary used for word matching from the word matching dictionary database (FIG. 8). Character recognition processing (step S90)
In the word matching process after step 5) (step S906), the result of the character recognition process for the block is verified and corrected by the word matching dictionary selected in the recognition setting selection process (step S904).

Note that step S in the third embodiment is performed.
Obviously, the two processes 905 and S906 may be performed successively for each block region, and the two processes of steps S905 and S906 may be repeated to perform the process for all block regions. .

In addition, the first or second embodiment is combined with the third embodiment to limit the recognition target or switch the recognition dictionary to perform the character recognition process, and further perform the word matching process in step S906. You may do so.

In each of the above embodiments, the item conversion database, the target character type limitation database, the recognition dictionary database, and the word collation dictionary database have a correspondence table for each of various document formats. When the user specifies using the input device such as the mouse 103, the table corresponding to the specified document format is controlled to be used.

As described above, according to the third embodiment, the word collation processing is switched according to the items in the block area, so that a more appropriate correction can be performed on the character recognition result.

As described above, according to the above embodiments, color information is converted into item information, and processing related to character recognition is set from the item information. For this reason, character recognition processing of documents such as forms becomes accurate. Character recognition processing can be performed as the same document format for documents having different shapes such as character area coordinates and ruled lines, that is, documents having different format shapes. Recognize items, so it is possible to deal with item deviations when reading images with a scanner, etc. It is possible to perform accurate character recognition according to item contents or correct character recognition errors according to item contents , Strict alignment is unnecessary when reading documents,
This has the effect of improving operability, shortening the overall processing time, and the like.

The present invention can be applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), but can be applied to a single device (for example, a copier, a facsimile). Device).

An object of the present invention is to provide a storage medium storing a program code of software for realizing the functions of the above-described embodiments to a system or an apparatus, and a computer (or CPU) of the system or apparatus.
And MPU) read and execute the program code stored in the storage medium.

In this case, the program code itself read from the storage medium realizes the functions of the above-described embodiment, and the storage medium storing the program code constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy disk, hard disk, optical disk, magneto-optical disk, CD-ROM, CD
-R, a magnetic tape, a nonvolatile memory card, a ROM, or the like can be used.

When the computer executes the readout program code, not only the functions of the above-described embodiment are realized, but also the OS (Operating System) running on the computer based on the instruction of the program code. ) May perform some or all of the actual processing, and the processing may realize the functions of the above-described embodiments.

Further, after the program code read from the storage medium is written into a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, based on the instruction of the program code, It goes without saying that the CPU included in the function expansion board or the function expansion unit performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0040]

As described above, according to the present invention, when character recognition is performed for each item in a document, character recognition can be accurately performed flexibly in response to differences in document formats. it can.

[0041]

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a character recognition device according to an embodiment.

FIG. 2 is a flowchart illustrating an operation procedure of the character recognition device according to the first embodiment.

FIG. 3 is a diagram illustrating a configuration example of block area data according to the first embodiment.

FIG. 4 is a diagram illustrating an example of an item conversion table in an item conversion database according to the first embodiment.

FIG. 5 is a diagram showing an example of a document to be read by the character recognition device of the embodiment.

FIG. 6 is a diagram illustrating an example of a target character type table of a target character type restriction database.

FIG. 7 is a diagram illustrating an example of a recognition dictionary correspondence table in a recognition dictionary database according to a second embodiment.

FIG. 8 is a flowchart illustrating an operation procedure of the character recognition device according to the third embodiment.

FIG. 9 is a diagram illustrating an example of a word matching dictionary correspondence table in the word matching dictionary database according to the third embodiment.

Claims (15)

[Claims] 1. An area dividing means for dividing an image into a plurality of areas based on color information, and an attribute setting means for setting a data attribute to each area based on the color information of each area obtained by the area dividing means. Recognition processing setting means for setting processing conditions of character recognition processing for each of the areas based on the data attributes set by the setting means; and setting of the image for each area by the recognition processing setting means. And a character recognizing means for performing a character recognizing process under the set processing conditions. 2. The attribute setting means has a conversion table in which color information and data attributes are associated with each other, and converts the data attribute corresponding to the color information of each area obtained by the area dividing means into the conversion table. 2. The character recognition apparatus according to claim 1, wherein the character recognition apparatus acquires the information by referring to and sets the area. 3. The data attribute includes a processing order, and the character recognizing unit executes a character recognizing process for each area according to a processing condition set by the recognition process setting unit in accordance with the processing order indicated by the data attribute. The character recognition device according to claim 1, wherein the character recognition is performed. 4. The character recognition device according to claim 1, wherein the data attribute represents a data item of each area. 5. The apparatus according to claim 1, wherein the recognition processing setting unit limits the character type to be subjected to character recognition for each of the areas based on the data attribute set by the setting unit. Character recognition device. 6. The character recognition apparatus according to claim 1, wherein the recognition processing setting means selects a recognition dictionary to be used for character recognition based on the data attribute set by the setting means. . 7. A word matching means for correcting a result recognized by the character recognition means with reference to a word matching dictionary, wherein the recognition processing setting means is based on a data attribute set by the setting means. 2. The character recognition device according to claim 1, wherein a word matching dictionary used by said word matching means is selected. 8. An area dividing step of dividing an image into a plurality of areas based on color information, and an attribute setting step of setting a data attribute in each area based on the color information of each area obtained in the area dividing step. A recognition processing setting step of setting processing conditions of character recognition processing for each of the regions based on the data attributes set in the setting step; and setting the image of each region in the recognition processing setting step. A character recognition step of performing a character recognition process under the set processing conditions. 9. The attribute setting step includes a conversion table in which color information and data attributes are associated with each other. The data attribute corresponding to the color information of each area obtained in the area dividing step is stored in the conversion table. 9. The character recognition method according to claim 8, wherein the character recognition method is obtained by referring to and setting the respective areas. 10. The data attribute includes a processing order, and the character recognizing step executes a character recognizing process for each area in accordance with the processing order indicated by the data attribute under the processing conditions set in the recognition process setting step. 9. The character recognition method according to claim 8, wherein the character recognition is performed. 11. The character recognition method according to claim 8, wherein the data attribute represents a data item of each area. 12. The character recognition method according to claim 8, wherein in the recognition processing setting step, a character type to be subjected to character recognition is limited for each of the areas based on the data attribute set in the setting step. Character recognition method. 13. The character recognition method according to claim 8, wherein said recognition processing setting step selects a recognition dictionary to be used for character recognition based on the data attribute set in said setting step. . 14. A word matching step for correcting a result recognized in the character recognition step with reference to a word matching dictionary, wherein the recognition processing setting step is based on the data attribute set in the setting step. 9. The character recognition method according to claim 8, wherein a word matching dictionary used in the word matching step is selected. 15. A storage medium for storing a control program for character recognition processing, the control program comprising: a code for an area dividing step of dividing an image into a plurality of areas based on color information; A code of an attribute setting step of setting a data attribute in each area based on the color information of each area obtained in the step; and a character recognition process for each of the areas based on the data attribute set in the setting step. A code of a recognition processing setting step of setting processing conditions; and a code of a character recognition step of performing character recognition processing on the image of each area under the processing conditions set in the recognition processing setting step. Storage media.