JP2000172785A - Device and method for recognizing character and computer readable memory - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a character recognizing device capable of improving character recognition precision and also total throughput and to provide a character recognizing method and a computer readable memory. SOLUTION: A attribute information reading part 4 reads the attribute information of image data inputted from an image inputting part 2. A character recognition controlling part 6 decides a dictionary of character recognition corresponding to the read attribute information. A character recognizing part 10 performs character recognition to the inputted image data by using the decided character recognition dictionary.

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 method for performing character recognition on input image data, and a computer readable memory.

[0002]

2. Description of the Related Art In a conventional character recognition apparatus, particularly in the recognition of handwritten characters, since the characters of an unspecified number of writers are to be recognized, it is virtually impossible to create a character recognition dictionary that satisfies all of them. It was possible. Therefore, the current situation is to create a character recognition dictionary based on a statistically extracted sample of handwritten characters, or to create a character recognition dictionary by limiting the writer, etc. Recognition. Regarding type recognition,
The dictionary for character recognition is created by limiting fonts to some extent.

A post-processing unit that performs post-processing after character recognition checks the connection between words in the character recognition result and corrects character recognition errors. The words used here are registered in the word collation dictionary in advance, and are called for each check. Generally, words are registered in the word collation dictionary so as to correspond to all types of manuscripts.

[0004]

However, in the above-mentioned conventional character recognition apparatus, it is difficult to narrow down a character recognition dictionary optimal for a recognition target, so that it is difficult to obtain a certain level of accuracy. In addition, when the writer or the font is limited, the usability is greatly deteriorated.

In the post-processing unit of the character recognition device, the dictionary for word collation is made versatile, so that the degree of redundancy is high and the processing speed and accuracy are reduced accordingly.

SUMMARY OF THE INVENTION The present invention has been made in consideration of the above problems, and has as its object to provide a character recognition apparatus and method capable of improving character recognition accuracy and total throughput, and a computer readable memory. .

[0007]

A character recognition apparatus according to the present invention for achieving the above object has the following arrangement.
That is, a character recognizing device that performs character recognition on input image data, comprising: reading means for reading attribute information of the input image data; and character recognition corresponding to the attribute information read by the reading means. Determining means for determining a dictionary for use; and character recognition means for performing character recognition on the input image data using the character recognition dictionary determined by the determining means.

[0008] Preferably, the image data is F
This is image data in a flashPix format.

[0009] Preferably, the attribute information is information indicating a creator of the input image data.

[0010] Preferably, the determination means includes a storage means for storing a character recognition dictionary for each creator,
A character recognition dictionary corresponding to the attribute information read by the reading unit is determined with reference to the storage unit.

Preferably, the attribute information is information indicating a type of the input image data.

Preferably, the determining means includes a storage means for storing a character recognition dictionary for each type of the input image data, and the reading means is read by the reading means with reference to the storing means. A character recognition dictionary corresponding to the attribute information is determined.

A character recognition device according to the present invention for achieving the above object has the following configuration. That is, a character recognition device that performs character recognition on input image data, a character recognition unit that performs character recognition on the input image data, and a reading device that reads attribute information of the input image data. Means for specifying the type of the input image data based on the attribute information read by the reading means; and the character recognition means based on the type specified by the specifying means. Post-processing means for performing post-processing on the character recognition result.

Preferably, the image data is F
This is image data in a flashPix format.

Preferably, the post-processing means stores a word collation dictionary for collating words in a character recognition result for each type of image data, and the identification means refers to the storage means. Determining means for determining a word matching dictionary corresponding to the type specified in the step (a), and performing post-processing on the character recognition result using the word matching dictionary determined by the determining means.

A character recognition method according to the present invention for achieving the above object has the following configuration. That is, a character recognition method for performing character recognition on input image data, comprising: a reading step of reading attribute information of the input image data; and a character recognition corresponding to the attribute information read in the reading step. And a character recognition step of performing character recognition on the input image data using the character recognition dictionary determined in the determination step.

A character recognition method according to the present invention for achieving the above object has the following configuration. That is, a character recognition method for performing character recognition on input image data, wherein a character recognition step for performing character recognition on the input image data, and reading for reading attribute information of the input image data A step of specifying the type of the input image data based on the attribute information read in the reading step; and a step of obtaining the character recognition step based on the type specified in the specifying step. And a post-processing step of performing post-processing on the character recognition result.

A computer readable memory according to the present invention for achieving the above object has the following configuration. That is,
A computer-readable memory storing a character recognition program code for performing character recognition on the input image data, wherein a program code for a reading step for reading attribute information of the input image data; and Character recognition for performing character recognition on the input image data by using a program code of a determination step of determining a character recognition dictionary corresponding to the read attribute information and a character recognition dictionary determined in the determination step Process code.

A computer readable memory according to the present invention for achieving the above object has the following configuration. That is,
A computer readable memory storing a character recognition program code for performing character recognition on input image data, wherein a program code for a character recognition step for performing character recognition on the input image data is provided. A program code for a reading step for reading attribute information of the input image data, a program code for a specifying step for specifying the type of the input image data based on the attribute information read in the reading step, and the specifying step And a program code of a post-processing step for performing post-processing on the character recognition result obtained in the character recognition step based on the type specified in the above.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below in detail with reference to the drawings.

First, a FlashPix format which is one of image formats of an image used in the following embodiments will be described.

FlashPix ™ (FlashPix
In the file format (registered trademark of Eastman Kodak Company, USA), attribute information and image data stored in an image header portion are further structured and managed as a file. This structured image file is shown in FIGS.

FIG. 9 and FIG. 10 are diagrams showing the structure of a FlashPix format image file.

Each property and data in the image file is accessed by storage and stream corresponding to the directory and file of MS-DOS. 9 and 10, the shaded portion is the storage and the unshaded portion is the stream. Image data and attribute information
Stored in the stream part. The image data is hierarchized at different resolutions, and the image data of each resolution is called a Subimage, and each is a Resolution.
on0, 1,... n. For image data of each resolution, information necessary for reading out the image data is stored in the Subimage header, and actual image data is stored in the Subimage data.

A property set is defined by classifying attribute information according to its purpose of use and its contents.
Mary info. Property set, I
image info. Property set, I
mage contentsPropertyse
t, Extension list Property
There is a set. <Explanation of each property set> Summery inf
o. Property set is FlashPix
It is not unique and is an essential property set in Microsoft's structured storage. In this property set, the title, title, author, thumbnail image, etc. of the image file are stored.

Image contents Prop
The erty set is an attribute that describes a method of storing image data (FIG. 13). This attribute includes the number of layers of the image data, the width and height of the image data of the maximum resolution, the width, height, color configuration of the image data of each resolution,
Alternatively, the definition of the quantization table / Huffman table when using JPEG compression is described.

[0027] Image info. Property
y set stores various information that can be used when using image data, for example, information on how an image as shown below is captured and how it can be used.

Information on a method of capturing and / or generating digital image data (File Source)
e) Copyright information (Intellectual p
information) (information about the content of the image (person, place, etc. in the image)) (Content description) ・ Information about the camera used for shooting (Camera)
Information) Information on camera settings (exposure, shutter speed, focal length, use of flash, etc.) at the time of shooting (Per Picture camera series)
(tings) ・ Information on digital camera specific resolution and mosaic filter (Digital camera charac)
Information about film manufacturer name, product name, and type (negative / positive, color / black and white) (Film desrip)
Information about the type and size when the original is a book or printed matter (Original document)
(scan description)-In the case of a scanned image, information on a scanner, software, and a person who operated the device (Scan device) Extension list Property s
et stores information not included in the basic specifications of the FlashPix.

FlashPix Image of FIG.
The view object is an image file that stores viewing parameters and image data used when displaying an image. What are viewing parameters?
This is a set of processing coefficients stored for adapting the processing of image rotation, enlargement / reduction, movement, color conversion, and filtering when displaying an image.

Source / Result Flash
Pix image object is FlashPi
x-format image data.
eFlashPix image object is required, Result FlashPix image o
bject is optional. Source Fla
shPix image object converts m original image data to Result FlashPix i
The image object stores image data obtained as a result of image processing using viewing parameters.

Source / Result desc.
The property set is a property set for identifying the image data, and stores an image ID, a property set for which change is prohibited, a last update date and time, and the like.

[0032] Transform property
set is Affin for rotation, scaling / movement
e conversion coefficient, color conversion matrix, contrast adjustment value,
Stores filtering coefficients.

Next, handling of image data will be described.

FIG. 11 is a diagram showing an example of an image file composed of a plurality of image data having different resolutions.

In FIG. 11, the maximum resolution image data has a column × row of X0 × Y0, and the next largest image data has a column × row of X0 / 2 × Y0 / 2. Thereafter, both the column and the row are sequentially reduced by １ ／, and the process is repeated until both the column and the row are equal to or smaller than 64 pixels.

As described above, as a result of hierarchization into a plurality of resolutions, as the attribute information of the image data, “the number of layers in one image file” and the image data of each layer
Header information and image data are required. The number of layers in one image data, the width and height of the image data of the maximum resolution,
Alternatively, information on the width, height, color configuration, compression method, and the like of an image at each resolution can be obtained from Image shown in FIG.
contents Described in the property set.

Further, the image data of each resolution is divided into 64 × 64 tiles as shown in FIG. When the image data is divided into 64 × 64 tiles sequentially from the upper left, blanks may occur in some of the right and lower end tiles depending on the image data. In this case, 64 × 64 pixels are constructed by repeatedly inserting the rightmost image data or the bottommost image data. FlashP
In the ix format, image data in each tile is stored in one of JPEG compression, single color, and non-compression.

JPEG compression is based on ISO / IECJTC1.
/ SC29 is an image compression system internationally standardized, and the description of the system itself is omitted here. The image data divided in this way is described in Subimage data.
ta, the total number of tiles, the size of each tile, the starting position of the data, and the compression method are all
It is stored in the Subimage header (FIG. 14). The single color is a method in which the color of a tile is represented by one color without recording the value of each pixel only when all the tiles are composed of the same color. This method is particularly useful for images generated by computer graphics. [First Embodiment] FIG. 1 is a block diagram showing a configuration of an information processing apparatus applicable to a character recognition apparatus according to a first embodiment of the present invention.

In FIG. 1, a CPU 101 controls the entire information processing apparatus 200 via a main bus 107, and also has an input device 111 (for example, an image scanner, a storage device,
Other information processing devices connected via a network line, a facsimile connected via a telephone line, etc.) are controlled via an input I / F (interface) 104.
In addition, an output device 112 (for example, a printer, a monitor, another information processing device connected via a network line, a facsimile connected via a telephone line, etc.) connected to the outside of the information processing device 200 is output to the output I. / F105. In addition, the CPU 101 determines whether the KBDI / F
A series of processes such as image input, image processing, color conversion processing, image output control, etc., according to an instruction input from an input unit (for example, the pointing device 123, the keyboard 124, or the pen 125) via the (keyboard interface) 106. Execute Further, a display unit 109 that displays image data input from the input device 111 and image data created using the pointing device 123, the keyboard 124, and the pen 125 is controlled via a video I / F (interface) 108. .

The ROM 102 stores various control programs for executing various controls of the CPU 101. RAM
The CPU 103 loads and executes an OS and other control programs including a control program for implementing the present invention by the CPU 101. In addition, it functions as various work areas used for executing the control program and a temporary save area. In addition, image data input from the input device 111, the pointing device 123, the keyboard 1
A VRAM (not shown) is configured to temporarily hold image data created using the pen 24 and the pen 125.

Next, the functional configuration of the character recognition device according to the first embodiment will be described with reference to FIG.

FIG. 2 is a diagram showing a functional configuration of the character recognition device according to the first embodiment of the present invention.

In FIG. 2, reference numeral 2 denotes an image input unit for inputting a FlashPix image. 4 is the input Flash
The attribute information reading unit reads attribute information of the hPix image. Reference numeral 6 denotes a character recognition control unit that controls character recognition based on the attribute information read by the attribute information reading unit 4. Reference numeral 8 denotes an image processing unit that performs processing suitable for character recognition on the image body of the FlashPix image input by the image input unit 2.
Reference numeral 10 denotes a character recognition unit that performs character recognition.

Note that the character recognition control unit 6 has a FlashP
A dictionary for character recognition for each document creator indicated by attribute information of the ix image and a dictionary for general character recognition are stored.

Next, the operation will be described.

FlashP input from the image input unit 2
ix image, the attribute information of which is stored in the attribute information reading unit 4
The entity data of the AshPix image body is input to the image processing unit 8. The attribute information reading section 4 reads document creator information indicating the creator of the document image. The character recognition control section 6 selects a character recognition dictionary based on the read document creator information. On the other hand, in the image processing unit 8,
Binary processing is performed on the entity data of the hPix image body. The binarized image data is input to the character recognition unit 10, and character recognition is performed using the character recognition dictionary selected by the character recognition control unit 6.

Next, the processing executed by the character recognition control unit 6 of the first embodiment will be described with reference to FIG.

FIG. 3 is a flowchart showing processing executed by the character recognition control unit according to the first embodiment of the present invention.

In step S202, the document creator information obtained by the attribute information reading section 4 is read. In step S204, it is determined whether or not the original creator is specified from the original creator information read in step S202. If the creator has been designated (YES in step S204), the flow advances to step S206 to select a handwritten character recognition dictionary set in advance by the document creator. On the other hand, if the creator has not been specified (NO in step S204), the flow advances to step S208 to select a general-purpose handwritten character recognition dictionary.

Next, the processing executed by the character recognition unit 10 will be described with reference to FIG.

FIG. 4 is a flowchart showing the processing executed by the character recognition unit according to the first embodiment of the present invention.

In step S302, the image data binarized by the image processing section 8 is read. Step S304
Then, the dictionary for handwritten character recognition selected by the character recognition control unit 6 is read. In step S306, step S3
02, character recognition is performed using the image data read in step S304 and the handwritten character recognition dictionary.

As described above, according to the first embodiment, by utilizing the attribute information of the FlashPix image, it is possible to select an optimal character recognition dictionary for the text. That is, the character recognition accuracy and usability can be significantly improved. [Second Embodiment] In the first embodiment, the configuration is such that the dictionary for handwritten character recognition is selected by specifying the document creator of the input image data. However, the present invention is not limited to this. For example, Image info. p
Original Docu of the property set
Ment Scan Description Gro
The configuration may be such that the type of the document is specified from up, and a dictionary for character recognition is selected. In this case, it is particularly effective for print character recognition. Third Embodiment FIG. 6 is a diagram showing a functional configuration of a character recognition device according to a third embodiment of the present invention.

In FIG. 6, reference numeral 102 denotes FlashPix.
An image input unit for inputting an image. Reference numeral 104 denotes an attribute information reading unit that reads attribute information of the input FlashPix image. 106 is a post-processing control unit that controls post-processing of character recognition based on the attribute information read by the attribute information reading unit 104. Reference numeral 108 denotes a character recognition unit that performs character recognition on the input entity data of the FlashPix image. 1
Reference numeral 10 denotes a post-processing unit that performs word matching on the character recognition result obtained by the character recognition unit 108.

Note that the post-processing control unit 106 has a
A dictionary for word matching for each document type indicated by the attribute information of the Pix image and a dictionary for general-purpose word matching are stored.

Next, the operation will be described.

Flas acquired from the image input unit 102
The entity data of the hPix image body is input to the character recognition unit 108. The attribute information of the FlashPix image is
The information is input to the attribute information reading unit 104. Post-processing control unit 10
In step 6, the type of the document to be recognized is determined based on the read attribute information, and a word collation dictionary in the character recognition post-processing unit 110 is selected accordingly. On the other hand, in the character recognition unit 108, the FlashPix input from the image input unit 102
Character recognition is performed on the entity data of the image body. The character recognition post-processing unit 110 performs word matching on the character recognition result obtained by the character recognition unit 108 based on the word matching dictionary selected by the post-processing control unit 106, and corrects character recognition errors.

Next, the processing executed by the post-processing control unit 106 in the third embodiment will be described with reference to FIG.

FIG. 7 is a flowchart showing the processing executed by the post-processing control section of the third embodiment of the present invention.

In step S212, the attribute information reading unit 10
The attribute information of the FlashPix image acquired in step 4 is read. In step S214, of the attribute information read in step S212, for example, the I shown in FIG.
image info. O of Property set
original Document Scan Des
The “Cryption Group” is read to determine the type of the document. If it is determined that the type of the original is for general use, the process proceeds to step S218, where a general-purpose word collation dictionary is selected. On the other hand, when the type of the document is determined to be a document in a specific field, the most appropriate word matching dictionary for the specific field is selected.

Next, the processing executed by the character recognition post-processing unit 110 will be described with reference to FIG.

FIG. 8 is a flowchart showing a process executed by the character recognition post-processing unit executed in the third embodiment of the present invention.

In step S312, the character recognition result obtained from character recognition unit 108 is read. Step S314
Then, the dictionary for word matching selected by the post-processing control unit 102 is read. In step S316, step S314
Based on the word collation dictionary selected in step S312,
Corrects the error in the character recognition result read by.

As described above, according to the third embodiment, by utilizing the attribute information of the FlashPix image, it is possible to select the most suitable word matching dictionary for the sentence. Therefore, it is possible to improve post-processing accuracy and, consequently, accuracy of overall character recognition. [Fourth Embodiment] In the third embodiment, in the attribute information of the FlashPix image, Image info. prop
erty set Original Documen
The configuration is such that the t Scan Description Group is read and the type of the document is specified, but the present invention is not limited to this. For example, Image i
nfo. Content of property set
The configuration may be such that the type of the document is inferred from the Description Group. [Embodiment 5] In the third embodiment, the dictionary for word matching in post-processing is divided into a general dictionary and a dictionary for a specific field. However, the present invention is not limited to this. May be prepared.

In each of the embodiments described above, a FlashPix format Flash is used as an image to be processed.
Although the Pix image was used, the present invention is not limited to this. Any image may be used as long as the image has a format that has information that can implement the present invention, such as information of a creator and information that can specify the type of a document.

Even if the present invention is applied to a system including a plurality of devices (for example, a host computer, an interface device, a reader, a printer, etc.), a device including one device (for example, a copying machine, a facsimile machine) Etc.).

Another 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 to provide 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 implements 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 instructions 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.

[0072]

As described above, according to the present invention,
A character recognition device and method capable of improving character recognition accuracy and total throughput, and a computer-readable memory can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an information processing device applicable to a character recognition device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a functional configuration of the character recognition device according to the first embodiment of the present invention.

FIG. 3 is a flowchart illustrating a process executed by a character recognition control unit according to the first embodiment of the present invention.

FIG. 4 is a flowchart illustrating processing executed by a character recognition unit according to the first embodiment of the present invention.

FIG. 5 shows an image info. Image defined in the FlashPix format. Property set
It is a figure showing an item.

FIG. 6 is a diagram illustrating a functional configuration of a character recognition device according to a third embodiment of the present invention.

FIG. 7 is a flowchart illustrating a process executed by a post-processing control unit according to a third embodiment of the present invention.

FIG. 8 is a flowchart illustrating a process executed by a character recognition post-processing unit executed in a third embodiment of the present invention.

FIG. 9 is a diagram illustrating a configuration of a FlashPix format image file.

FIG. 10 is a diagram showing a configuration of a FlashPix format image file.

FIG. 11 is a diagram showing an example of an image file composed of a plurality of image data having different resolutions.

FIG. 12 is a diagram for explaining tile division.

FIG. 13: Image contents Prop
It is a figure for explaining rty Set.

FIG. 14 is a diagram for explaining a Subimage header.

[Explanation of symbols]

 101 CPU 102 ROM 103 RAM 104 Input I / F 105 Output I / F 106 KBDI / F 107 Main bus 108 Video I / F 109 Display unit 111 Input device 112 Output device 123 Pointing device 124 Keyboard 125 Pen 200 Information processing device 2, 102 Image input unit 4, 104 Attribute information reading unit 6 Character recognition control unit 8 Image processing unit 10, 108 Character recognition unit 106 Post-processing control unit 110 Character recognition post-processing unit

Claims (20)

[Claims] 1. A character recognition device for performing character recognition on input image data, comprising: a reading unit that reads attribute information of the input image data; and a character recognition device that corresponds to the attribute information read by the reading unit. Determining means for determining a character recognition dictionary to be performed, and character recognition means for performing character recognition on the input image data using the character recognition dictionary determined by the determining means. Recognition device. 2. The character recognition apparatus according to claim 1, wherein the image data is FlashPix format image data. 3. The character recognition device according to claim 1, wherein the attribute information is information indicating a creator of the input image data. 4. The deciding means comprises a storage means for storing a character recognition dictionary for each creator, and referring to the storage means, a character recognition dictionary corresponding to the attribute information read by the reading means. The character recognition device according to claim 3, wherein a dictionary is determined. 5. The apparatus according to claim 1, wherein the attribute information is information indicating a type of the input image data.
The character recognition device according to 1. 6. The image processing apparatus according to claim 1, wherein the determining unit includes a storage unit that stores a character recognition dictionary for each type of the input image data, and stores the attribute information read by the reading unit with reference to the storage unit. The character recognition device according to claim 5, wherein a corresponding character recognition dictionary is determined. 7. A character recognition device for performing character recognition on input image data, comprising: character recognition means for performing character recognition on the input image data; and attribute information of the input image data. Reading means for reading the image data; specifying means for specifying the type of the input image data based on the attribute information read by the reading means; and the character recognizing means based on the type specified by the specifying means. And a post-processing means for performing post-processing on the character recognition result obtained in (1). 8. The character recognition device according to claim 7, wherein the image data is FlashPix format image data. 9. The post-processing unit includes: a storage unit that stores a word collation dictionary for collating with words in character recognition results for each type of image data; Determining means for determining a word matching dictionary corresponding to the type, and performing post-processing on the character recognition result using the word matching dictionary determined by the determining means. 8. The character recognition device according to 7. 10. A character recognition method for performing character recognition on input image data, comprising: a reading step of reading attribute information of the input image data; and a method corresponding to the attribute information read in the reading step. And a character recognition step of performing character recognition on the input image data using the character recognition dictionary determined in the determination step. Recognition method. 11. The method according to claim 11, wherein the image data is a FlashPix.
The character recognition method according to claim 10, wherein the data is image data in a format. 12. The character recognition method according to claim 10, wherein the attribute information is information indicating a creator of the input image data. 13. The character recognition dictionary according to claim 12, wherein the determining step determines a character recognition dictionary corresponding to the attribute information read in the reading step from the character recognition dictionary for each creator. Character recognition method. 14. The character recognition method according to claim 10, wherein the attribute information is information indicating a type of the input image data. 15. The determining step includes a storing step of storing a character recognition dictionary for each type of the input image data, and referring to the storing step, the attribute information read in the reading step is referred to. The character recognition method according to claim 14, wherein a corresponding character recognition dictionary is determined. 16. A character recognition method for performing character recognition on input image data, comprising: a character recognition step of performing character recognition on the input image data; and attribute information of the input image data. A reading step of reading the image data; a specifying step of specifying a type of the input image data based on the attribute information read in the reading step; and a character recognition step based on the type specified in the specifying step. And a post-processing step of performing post-processing on the character recognition result obtained in (1). 17. The method according to claim 17, wherein the image data is a FlashPix.
17. The character recognition method according to claim 16, wherein the data is image data in a format. 18. The post-processing step determines a word matching dictionary corresponding to the type specified in the specifying step from a word matching dictionary that matches a word in a character recognition result for each type of image data. 17. The character recognition method according to claim 16, further comprising a determining step, wherein post-processing is performed on the character recognition result using the word matching dictionary determined in the determining step. 19. A computer readable memory storing a character recognition program code for performing character recognition on input image data, the program code for a reading step of reading attribute information of the input image data, A program code of a determining step of determining a character recognition dictionary corresponding to the attribute information read in the reading step; and using the character recognition dictionary determined in the determining step, a character is input to the input image data. A program code for a character recognition step for performing recognition. 20. A computer readable memory storing a character recognition program code for performing character recognition on input image data, wherein a program for a character recognition step for performing character recognition on the input image data is provided. A program code for a reading step for reading attribute information of the input image data; and a program code for a specifying step for specifying the type of the input image data based on the attribute information read in the reading step. And a program code of a post-processing step of performing post-processing on a character recognition result obtained in the character recognition step based on the type specified in the specifying step.