JPH10171920A - Character recognition device, its character recognition method and its recording medium - Google Patents

Abstract

(57) [Summary] [Problem] To determine the type of a form in which a ruled line is unclear. SOLUTION: Character recognition is performed by a print character recognition CPU 102 on a read image of a form. The company name indicating the type of the form is detected in the text obtained as a result of the character recognition, and the control
The format information is extracted from the format table 123 of the DD 107.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention reads a form,
The present invention relates to a character recognition device for recognizing a character image described on a form and converting the character image into a character code, a character recognition method thereof, and a recording medium thereof.

[0002]

2. Description of the Related Art Currently, orders, orders, deliveries, bills, payments, and inventories are made. Logistics, reservation, medical treatment, tax payment, notification, transfer,
The use of various designated slips, such as transfer, has become extremely remarkable. In other words, each company or organization has set up an individual form and performs its business. Such a slip is converted by a character recognition device called OCR into a document called read text. This enables various information processing such as saving or printing the contents of the form in a database on an information processing device, for example, a workstation or a personal computer, without the user having to input the contents of the form from the keyboard. Is becoming possible.

[0003] Although the character recognition rate of the character recognition device is improving, it has not reached 100%. In particular, the read document in which the ruled line image and the character image are mixed like a form has a lower character recognition rate. This is because the comparison character pattern stored in the character recognition device does not include a ruled line image, so that if the image to be read includes a ruled line, a perfect match is obtained in pattern recognition. It is due to not having. For this reason, some conventional character recognition devices detect a ruled line in a read image when a document document is read, delete the ruled line image from the read image, or extract a character image. An apparatus for performing character recognition has been proposed.

As a technique for automatically determining the type of a form, a technique for detecting the length, width, and the like of a ruled line frame based on a read image and discriminating the type of the form based on its size is already known. I have.

[0005]

When the type of a form can be automatically determined, the result of character recognition, that is, a text consisting of a character code including a control code such as a line feed code, and a ruled line having a predetermined format (format) It is possible to print the image by overlaying it.

However, when the type of a form is determined based on the ruled lines of the form original, there are the following problems to be solved.

(1) When the width of a ruled line is narrow, a dotted line, or a form original in which a ruled line is not described, dimensional information relating to the ruled line cannot be obtained, so that the type of such a form cannot be determined.

(2) Since the ruled line detection processing is performed from the read image, the ruled line detection processing takes longer as the form document becomes larger. Also, the larger the number of ruled lines, the longer the time required for ruled line detection processing.

In view of the above, it is an object of the present invention to provide a character recognition device which does not depend on the shape and description state of ruled lines of a form, and which can reduce the type determination processing time as compared with the related art.
An object of the present invention is to provide a character recognition method and a recording medium.

[0010]

In order to achieve the above object, according to the first aspect of the present invention, an image of a form read from a document of a form having different characters and having different formats is recognized by a character. In a character recognition device (100) for automatically determining the format type of a form, a unique identification name written on the document of the form and associated with the format type is stored for each type of format in the form of a character code. Identification name storage means (107), and whether or not there is a character code string that matches the identification name stored in the storage means among character code strings obtained as a result of character recognition of the read form image Determining means (S200 in FIG. 6) for determining whether or not the format has been determined based on the association between the type of format and the identification name when an affirmative determination is obtained. And obtaining the type of format corresponding to the character string to be.

According to a second aspect of the present invention, in the character recognition apparatus according to the first aspect, a ruled line is described in the document of the form, and the document is read from the read form image prior to the character recognition. An image processing means (S3 in FIG. 5) for erasing the image of the ruled line is further provided.

According to a third aspect of the present invention, in the character recognition device according to the second aspect, a format information storage means (107) for storing format information indicating a position of a ruled line in association with the format type; Information processing means (S8 in FIG. 5) for acquiring format information corresponding to the acquired format type from the format information storage means based on an association between the format type and the format information. Features.

According to a fourth aspect of the present invention, in the character recognition apparatus according to the third aspect, a cell character recognizing means (103) for recognizing a cell image of a character image sandwiched between ruled lines at positions indicated by the acquired format information. Further features are provided.

According to a fifth aspect of the present invention, in the character recognition apparatus according to the first aspect, an image input means for inputting an image of the form is further provided.

According to a sixth aspect of the present invention, in the character recognition apparatus according to the fifth aspect, the image input means is an image reading device (108) for reading the document of the form.

According to a seventh aspect of the present invention, in the character recognition apparatus according to the fifth aspect, the image input means receives an image of the form transmitted from an image source via a public telephone line. 111).

According to an eighth aspect of the present invention, in the character recognition apparatus according to the fifth aspect, the image input means receives a form image transmitted from an image source via a wide area network. ).

According to a ninth aspect of the present invention, in the character recognition device according to the eighth aspect, the image generation source is an information processing device (2
00) and an image reading device (201) connected to the information processing device.

According to a tenth aspect of the present invention, in the character recognition device according to the first aspect, prior to the determination by the determination means,
A correction means (S205 in FIG. 6) for correcting an erroneously recognized portion in the character string obtained as a result of the recognition is further provided.

According to an eleventh aspect of the present invention, in the character recognition device according to the tenth aspect, the correction means corrects the character string to be corrected by inserting and deleting characters. Character processing means (S in FIG. 5)
7B).

According to a twelfth aspect of the present invention, in the character recognition apparatus according to the tenth aspect, the correcting means includes a character code string of the identifier stored in the storage means and a character code string obtained as a result of the character recognition. Detecting means for detecting an identification name having a predetermined number or more of matching characters between S1 and S2 (S1 in FIG. 6)
20) and character processing means for correcting the character code string of the detected identification name by replacing the character string including the erroneously recognized part (S205 in FIG. 6).

According to a thirteenth aspect of the present invention, in the character recognition device according to the first aspect, the identification name is an organization name.

According to a fourteenth aspect, in the character recognition device according to the first aspect, the identification name is a slip name.

According to a fifteenth aspect of the present invention, there is provided a character recognition method for a character recognition device for recognizing a character of a form read from a document of a form in which characters are described and having different formats and for automatically discriminating the type of the form of the form. In the document of the form, a unique identifier for each type of the format is stored in a storage device in the character recognition device in the form of a character code in association with the type of the format, and It is determined by an information processing device in the character recognition device whether or not there is a character code sequence that matches the identification name stored in the storage means in a character code sequence obtained as a result of character recognition of the image of the created form. When an affirmative determination is obtained, the information processing apparatus determines the matching sentence based on the association between the format type and the identifier. And obtaining the type of format corresponding to the column.

According to a sixteenth aspect of the present invention, in the character recognition method of the character recognition device according to the fifteenth aspect, ruled lines are described in the document of the form, and prior to the character recognition,
The information processing apparatus deletes the image of the ruled line from the read image of the form.

According to a seventeenth aspect of the present invention, in the character recognition method of the character recognition apparatus according to the sixteenth aspect, format information indicating a position of a ruled line is stored in the storage device in association with the format type. The information processing apparatus acquires the format information corresponding to the acquired format type based on the association between the format type and the format information.

According to an eighteenth aspect of the present invention, in the character recognition method of the character recognition apparatus according to the seventeenth aspect, the information processing device recognizes a character image sandwiched between ruled lines at positions indicated by the acquired format information as cell character recognition. It is characterized by doing.

A nineteenth aspect of the present invention is the character recognition method of the character recognition device according to the eighteenth aspect, wherein the character recognition device includes an image input device for inputting an image of the form.

According to a twentieth aspect of the present invention, in the character recognition method according to the nineteenth aspect, the image input device is an image reading device for reading the document of the form.

According to a twenty-first aspect of the present invention, in the character recognition method for the character recognition device according to the nineteenth aspect, the image input device receives an image of the form transmitted from an image source via a public telephone line. Communication device.

According to a twenty-second aspect of the present invention, in the character recognition method according to the nineteenth aspect, the image input device receives an image of the form transmitted from an image source via a wide area network. It is a communication device.

According to a twenty-third aspect of the present invention, in the character recognition method of the character recognition device according to the twenty-second aspect, the image generation information processing apparatus and an image reading apparatus connected to the information processing apparatus. .

According to a twenty-fourth aspect of the present invention, in the character recognition method of the character recognition device according to the fifteenth aspect, the information processing apparatus performs an error in the character string obtained as the recognition result prior to the determination processing. It is characterized in that the recognition part is corrected.

According to a twenty-fifth aspect of the present invention, in the character recognition method of the character recognition apparatus according to the twenty-fourth aspect, the information processing apparatus corrects the character string to be corrected by inserting and deleting characters. Is performed.

According to a twenty-sixth aspect of the present invention, in the character recognition method of the character recognition device according to the twenty-fourth aspect, the information processing device includes a character code string of an identification name stored in the storage device and the character recognition result. An identifier having at least a predetermined number of matching characters is detected between the obtained character code string and the character code string of the detected identifier is corrected by replacing the character string including the erroneously recognized portion. It is characterized by the following.

According to a twenty-seventh aspect of the present invention, in the character recognition method according to the fifteenth aspect, the identification name is an organization name.

According to a twenty-eighth aspect of the present invention, in the character recognition method according to the fifteenth aspect, the identification name is a slip name.

According to a twenty-ninth aspect of the present invention, an image of a form read from a document of a form in which characters are described and having different formats is recognized by an internal information processing device, and the type of the format of the form is automatically determined. On a recording medium loaded with a program code means that is loaded on a recognition device and recorded by the information processing device, an identification name written on the document of the form and unique for each type of the format is set to the type of the format. It is stored in the storage device in the character recognition device in the form of a character code in association with the program code means,
A processing step of determining whether or not there is a character code string that matches the identification name stored in the storage means in a character code string obtained as a result of character recognition of the read form image; Is obtained based on the association between the format type and the identifier,
A processing step of acquiring a format type corresponding to the matching character string.

[0039]

Embodiments of the present invention will be described below in detail with reference to the drawings.

(Embodiment 1) FIG. 1 shows a system configuration of a character recognition system to which the present invention is applied. In FIG.
The character recognition device 100 and the terminal device 200 are connected via a LAN (Wide Area Network) 250 constructed by Ethernet or the like. The terminal device 200 can use an information processing device that can read a form read by a scanner 201 such as a workstation or a personal computer and transfer it to the character recognition device 100.

The character recognition device 100 is a fax (FA)
X) The transceiver 300 is also connected to the public telephone line 350,
A read image of a form can be received from the fax transceiver 300. The character recognition device 100 is a terminal device 2
00 or the read image of the form sent from the FAX transceiver 300 is subjected to character recognition to recognize the contents of the form as characters, and to determine the type of the form according to the present invention. As will be described later, the character recognition device 100 also has a scanner, and the document can be read by the character recognition device alone.

FIG. 2 shows the internal configuration of the character recognition device 100. In FIG. 2, the following components are connected to the bus 113.

The control CPU 101 controls the entire apparatus according to a system control program (hereinafter abbreviated as an operating system or OS). Further, the form type determination processing according to the present invention is executed according to the processing procedure shown in FIG.

The print character recognition CPU 102 uses a digital processor such as a DSP for high-speed processing, and performs character recognition on a read image of a form in which characters are printed in print. A conventionally known method can be used as the character recognition method. In the present embodiment, an image is extracted for each line, and a character string image sandwiched between blanks is extracted from the image. Thereafter, character recognition is performed by comparing the image of each character with the character image of the standard pattern. Each character of the character string recognition result is represented by a character code.

The handwriting recognition CPU 103 uses a digital processor for high-speed processing such as a DSP and recognizes handwritten characters written on a form. As a technique for recognizing handwritten characters, a conventionally known technique can be used.
For example, a handwritten character is recognized by extracting a character feature pattern such as the direction, length, and number of line segments in a character image, and comparing the extracted character pattern with a standard character feature pattern prepared for each character.

The work memory 104 is provided with the CPU 10 described above.
1. Temporarily store various data used for calculations executed by the CPU 102 and the CPU 103. Work memory 104
Can use a RAM. Video memory 10
Reference numeral 5 stores image data for display. The image data is written by the control CPU 101, read out by a DMA (direct memory access) controller (not shown) or the like, and sent to a CRT (cathode ray tube display) 106. The CRT 106 displays the transmitted image data. In this embodiment, a read image of a form, a character recognition result, a form type determination result, and the like can be displayed.

The hard disk storage device (abbreviated as HDD) 107 includes, in addition to the above-described system program for the OS,
Form read image 120, text 12 after character recognition
1. Format dictionary 122 used for type discrimination, form format table 123, cell image (image included in character frame) extracted from form read image 1
24 is stored. These programs and data are stored in the work memory 104 when the CPU is used.
Loaded and used.

The format dictionary 122 includes a character string (character code form, hereinafter referred to as an identification name) described in the form as a character string unique to the company or the corporation, and a format of the form of the company or the corporation. The assigned form codes are described in association with each other. In the present embodiment, an example of the identification name is an organization name such as a company or a corporation.

In the format table 123, format information indicating the start point coordinates and end point coordinates of the ruled line described in the form, the number of characters in the table formed by the ruled line, and the like, and the form code are described in association with each other.

The scanner 108 reads (photoelectrically converts) the document document using an image pickup device such as a CCD, and delivers the document document to the CPU 101 in the form of image data. Scanner 1
08 is typically connected to the bus 113 via a scuzzy (SCSI) cable and interface, but these components are not shown for convenience of explanation.

The keyboard 108 performs operation instructions to the control CPU 101 and inputs information such as characters and numerals. The mouse 110 moves the mouse cursor displayed on the CRT 106, and specifies the position by operating the click button with the mouse cursor. By this position designation, information assigned to this position, for example, information such as a command and a character string is input. The communication interface 111 communicates with an external device and takes in a form image. In the present embodiment, it is possible to input a form image to the character recognition device via the FAX modem 112, a public telephone line, or also via a LAN such as Ethernet.

The form type discriminating process performed in such a system configuration will be described. In the present embodiment, it is assumed that the form document image is read by the scanner 108. The user sets the document on the scanner 109 and instructs the keyboard 109 to start an image reading (OCR) program. In response, a character recognition program having the contents described in FIG. 5 is loaded from the HDD 107 into the program storage area in the work memory 104 and executed by the control CPU 101. Although FIG. 5 and the processing procedure of FIG. 6 showing the details of FIG. 5 are described in a program language that can be executed by the CPU, they are separated into well-known functions for convenience of description. Those skilled in the art will be able to create an execution program based on this description.

The control CPU 101 instructs the scanner 108 to read. In response to this instruction, the scanner 108 reads the form document. The read form image is captured in the work memory 104 in the form of image data for each pixel, more specifically, in the form of binary data when reading in black and white, and in multivalued data when reading in color, and then stored in the HDD 107. (See step 120 in FIG. 2; step S1 → S2). The contents of the form in this state are schematically indicated by reference numeral 201 in FIG. Here, the numbers are handwritten characters, and the other characters are printed typefaces. Note that in the read image 201, characters and ruled lines are in the form of image data.

Next, the CPU 101 erases the ruled line image from the form image in the work memory 104 (firm-out) (step S3). Since the ruled line has black pixels continuous for a certain length or more and a certain width or less, a conventional technique of detecting a set of pixels satisfying such a condition from a form image is used. Figure 3 shows the contents of the form image with the ruled line image deleted
Is schematically indicated by reference numeral 202. The control CPU 101 instructs the print character recognition CPU 102 to perform character recognition on the form image excluding the ruled line image. In response to this instruction, the CPU 102 for character recognition for printing uses the work memory 1
Character recognition is performed on the form-read image that has been erased from the ruled line stored in the document 04 (step S4). Thus, the character string of the printed portion in the content indicated by reference numeral 202 in FIG. 3 is recognized. However, some characters may be erroneously recognized. A blank character code is used as a recognition result for an image in a blank portion, and a forced line feed code is inserted at the end of the line of the recognition result. In the present embodiment, the type of a form is determined using such a recognition result. That is, due to the nature of the form, attention is paid to the fact that the form always includes a proper noun (identification name) indicating the type, for example, a company name, and the identification name registered in the format dictionary 122 in advance is used for character recognition. The control CPU 101 inquires whether the result is included in the result (step S5). In the present embodiment, assuming that there is an erroneous recognition, for example, the first three characters of each character string of the text and the first three characters of each identification name in the format dictionary are compared with each other to find a matching character string. Is detected. If a matched string is obtained, the entire string including the string used for comparison is compared next, and if an exact match is not found, the character string on the text side of the character recognition result is corrected (See FIG. 4), and the procedure shifts to step S8 (step S6). Since this process is related to the present invention,
This will be described later with reference to FIG.

On the other hand, if the identification name registered in the format dictionary 122 is not described, the form to be read is a new form or a misrecognized character among the first three characters used for comparison. , The control CPU 101 reads the text form (see reference numeral 202 in FIG. 3) from the work memory 104, converts the character code into an image, and converts the character code into an image.
Write 5 As a result, the contents of the form are displayed on the CRT 106. At the same time, a message indicating that there is no registered identification name is also displayed. The user looks at this display content and determines whether the form is new or contains a misrecognized character string. In the case of a new name form, the user ends the current program (step S7A → end), inputs a new identification name from the keyboard 109, and enters the format dictionary 12
In step 2, the control CPU 101 additionally registers the input new identification name.

Thereafter, the user executes the program shown in FIG. 5 again to perform character recognition again. On the other hand, if the first three character strings include erroneous recognition, the user moves the mouse cursor on the display screen to the erroneously recognized character by using the mouse 110 to specify the character to be corrected. 10
Enter correct characters from 9. The control CPU 101 deletes the designated character in the text on the work memory, inserts the input correct character, and displays the corrected text. In response to an instruction from the keyboard 109 such as the end of the correction, the control CPU 101 returns to step S5 and re-executes the inquiry processing of the identifier (steps S7A → S7B → S
5).

When the above processing is executed, the identification code of the form described in the format dictionary 122 corresponding to the identification name is obtained from the identification name described in the form in the text form (YES determination in step S6). . As a result, the type content of the form is determined.
The PU 101 retrieves the format data corresponding to the form code from the format table 123 by searching for character recognition of the handwritten characters in the read form image (step S9).

Next, the control CPU 101 passes the format data and an instruction to execute handwritten character recognition to the handwritten character recognition CPU 103. As a result, the handwritten character recognition CPU 103 performs cell character recognition for a character image in a ruled line frame (cell) whose position is indicated by the format data, that is, for a character image sandwiched between vertical and horizontal ruled lines (S1).
0). As a result, a recognition result for the region for which the recognition frame is designated is obtained as indicated by reference numeral 203 in FIG. It goes without saying that typeface characters are also recognized. The character recognition result is displayed on the CRT 106. The user looks at this display and corrects an erroneously recognized part in the recognition result (step S11 → S12).

After the above-described processing, the user operates the keyboard 10
9 or the mouse 110, the control CPU 101 ends the control procedure of FIG. Also, if necessary, the form image (reference numeral 201 in FIG. 3) and the result of handwritten character recognition (reference numeral 203 in FIG. 3) read by the scanner 108 are stored in the HDD 107 by the control CPU 101 according to the automatic or manual instruction. Is stored.

The identification name inquiry processing in step S5 in FIG. 5 will be described with reference to FIG.

The control CPU 101 is a work memory 104
From the text stored in, the first part in the character string group starting after the blank code is detected. More specifically, the pointer is positioned at the head of the text, and the character at the position of the pointer is extracted from the text. It is determined whether this character is a blank code. Next, the pointer is moved to the next character position, and the character at that position is extracted from the text. It is determined whether or not this character is a character that is not a symbol such as a blank character or a punctuation mark. When the above two determination processes are executed while moving the pointer, when the first determination is a blank character and the next determination is not a blank character, the pointer value is changed to the first character string of the second character string. Indicates the th character position. Thus, the position of the first character string can be detected. Hereinafter, by moving the pointer and performing the above-described processing, the second and subsequent character strings can be detected. The control CPU 101 extracts three characters from the head of this character string and temporarily stores them in the work memory 104 (step S100).

Next, the control CPU 101 reads out the character code string of the first identifier by specifying the address of the storage location of the first identifier described in the format dictionary 122, and
It is temporarily stored in the work memory 104. The first three characters of the character code string of the identification name on the work memory 104 are extracted, and the comparison with the three characters extracted from the above-described text is performed (steps S110 → S120).

When a match is obtained, a match comparison is performed on the entire character string stored in the work memory 104 and extracted from the identification name and the text (step S200). If a match is obtained, the form code stored in the same field as the identification name is extracted from the format dictionary 122 (step S210).

On the other hand, if a mismatch is determined, it is determined that there is an erroneously recognized portion in the character string extracted from the text, and the control CPU 101
Work memory 104 with a character string (identification name) extracted from
Modify the corresponding string in the text inside. As a result, the recognized character string is corrected to a correct character string (step S200 → S205). Thereafter, the form code is extracted from the correct character string as described above (S205 → S200 → Step S210).

If a three-character comparison results in a mismatch (NO in step S120),
The character string on the text side to be compared is fixed, and the character string to be compared on the format dictionary 122 is changed to the character string at the next position (step S130 → S135). Hereinafter, it is determined whether the three characters match in the same manner as described above (step S120). When the mismatch determination continues in this way, the character string of the final identifier on the format dictionary 122 side to be compared reaches. Therefore, the control CPU 10
1 is the first character string on the text side is the format dictionary 12
2 is not registered, and the character string to be compared on the text side is changed (steps S130 → S140 → S
145 → S100). In this way, the character string of the identifier in the format dictionary 122 is compared with the character string on the text side, and the comparison process is repeated until the extracted three characters match. If the first three characters are not misrecognized and the form document is not new, a match is always obtained (step S120), and a corresponding form code is obtained from the matched character string (step S210).

If all character strings in the text are not compared with all character strings in the format dictionary 122, a match cannot be obtained for three characters.
1 judges that the form document is new, and displays a message to that effect, a text and a read form image.
Upon seeing this display, the user manually operates, for example,
Correction of a misrecognized character in the character portion and new registration of an identification name in the format dictionary 122 are performed. The process of inserting and deleting characters for correcting characters and the process of registering a dictionary are well known and will not require detailed description.

(Embodiment 2) In Embodiment 1 described above, the type of a form is determined based on an organization name. However, as the types of slips for character recognition increase, the number of registered organizations also increases. Therefore, a second embodiment suitable for character recognition of various slips will be described. It should be noted that the system configuration can be substantially the same as the system configuration shown in FIG. 2, and therefore only the differences will be described. In the second embodiment, the format dictionary 122 has a hierarchical structure as shown in FIG. That is, the character code string indicating the type of the slip is set to the upper layer, and the lower layer is associated with the character code string of the organization name used in the invoice, purchase order, and the like, and the corresponding form code for each type of the slip name. It is described.

In the inquiry process of the identifier in step S5 in FIG. 5, a two-stage inquiry is made to the format dictionary 122. In the first time, the slip name in the format dictionary 122 is compared with the character string of the text whose character has been recognized from the read image of the form. When a matching slip name is obtained, next, it is determined whether the organization name located in the lower layer of the corresponding slip name in the format dictionary 122 matches the character string in the text. Acquires the corresponding form code when it is determined that the organization name matches.

When the above processing is performed, it is not necessary to perform the search processing for all the organization names, and the inquiry processing time can be shortened.

The following embodiment can be carried out in addition to this embodiment.

1) There are various proposals for a character recognition method for correctly correcting the result of character recognition, and it is preferable that a user uses a desired character recognition method.

2) In order to input a form image to the character recognition device 100, a floppy disk,
An input method for reading from a recording medium such as a magneto-optical disk can be used.

3) In the above-described embodiment, the character recognition of the typeface and the character recognition of the handwritten character are performed. However, when the form is composed of only the typeface, only the character recognition of the typeface is performed. May be configured.

4) In the above embodiment, the high-speed processing is aimed at, and the character recognition processing is shared by the three CPUs 101 to 103. However, the character recognition processing may be performed by one CPU.

5) In addition to an organization name, a slip name, and a combination of these names, a character string used to determine the type of a form can use a unique name for the form, and the user can select a different type of name. And use it.

[0076]

As described above, as described above, claims 1 and 1
In the inventions of the fifth and 29th embodiments, since the type of the form is determined based on the character code string of the character recognition result, image processing is required, and the processing time is shortened as compared with the conventional example in which the form is determined by determining the shape of the ruled line. In addition, even if the ruled line is not clear, the ruled line is not affected and the type determination accuracy is improved.

According to the second and 16th aspects of the present invention, the character recognition accuracy is improved by deleting the ruled line image. Thereby, the accuracy of the type determination is also improved.

According to the third, fourth, seventeenth, and eighteenth aspects of the present invention, it is possible to print a form based on the format information indicating the position of the ruled line and the result of character recognition, and to describe the form in a specific ruled line frame. It becomes possible to recognize the handwritten character. Further, since the character recognition area can be specified by the format information, the cutout processing of the character image is easy, and thus the handwritten character recognition processing time can be reduced.

According to the fifth to ninth and nineteenth to twenty-third aspects, it is possible to adopt a suitable form image input form according to the user's request.

According to the tenth to twelfth and twenty-fourth to twenty-sixth aspects, the recognition character can be corrected automatically or manually, so that the type determination accuracy is improved.

According to the invention of claims 13, 14, 27, and 28, it is possible to easily determine the type of a form by using a form-specific character string such as an organization name such as a company name or a corporation name or a form name. Further, by combining such names, it is possible to improve the accuracy of discriminating the type of a form, and to omit an inquiry process using useless names.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a system configuration of a character recognition system according to a first embodiment.

FIG. 2 is a block diagram showing a system configuration of the character recognition device 100.

FIG. 3 is an explanatory diagram showing a read image of a form document, form contents after removing ruled line images, and a handwritten character recognition area.

FIG. 4 is an explanatory diagram for explaining a process of deriving a form code from a result of character recognition.

FIG. 5 is a flowchart illustrating a form type determination processing procedure.

FIG. 6 is a flowchart showing a detailed processing procedure of step S5 in FIG. 5;

FIG. 7 is an explanatory diagram showing the contents of a format dictionary 122 according to the second embodiment.

[Explanation of symbols]

 100 Character Recognition Device 101 Control CPU 102 Print Character Recognition CPU 103 Handwritten Character Recognition CPU 104 Work Memory 105 Video Memory 106 CRT 107 HDD 108 Scanner 109 Keyboard 110 Mouse 111 Communication Interface 122 Format Dictionary 123 Format Table 200 Terminal Device 300 FAX Transceiver

Claims (29)

[Claims] 1. A character recognition device for recognizing characters of a plurality of types of forms read from a document of a form in which characters are described and having different formats, and for automatically discriminating a type of the form of the form. And an identifier storing means for storing a unique identifier for each type of the format in the form of a character code in association with the type of the format, and a character obtained as a result of character recognition of the read image of the form. Determining means for determining whether or not there is a character code string that matches the identification name stored in the storage means in the code string; and, when a positive determination is obtained, the type of the format and the identification name A character recognition device for acquiring a type of a format corresponding to the matching character string based on the association between the character strings. 2. The character recognition device according to claim 1, wherein a ruled line is described in the document of the form, and the image of the ruled line is erased from the read image of the form before the character recognition. A character recognition device further comprising image processing means for performing 3. A character recognition apparatus according to claim 2, wherein: format information storage means for storing format information indicating a position of a ruled line in association with said format type; A character recognition apparatus, further comprising information processing means for acquiring format information corresponding to the acquired format type from the format information storage means based on an association between the formats. 4. The character recognition apparatus according to claim 3, further comprising cell character recognition means for recognizing a cell image of a character image sandwiched between ruled lines indicated by the acquired format information. Character recognition device. 5. The character recognition device according to claim 1, further comprising image input means for inputting an image of the form. 6. The character recognition device according to claim 5, wherein said image input means is an image reading device for reading a document of said form. 7. The character recognition device according to claim 5, wherein said image input means is a communication device for receiving an image of said form transmitted from an image generation source via a public telephone line. Character recognition device. 8. The character recognition device according to claim 5, wherein said image input means is a communication device for receiving an image of said form transmitted from an image generation source via a wide area network. Character recognition device. 9. The character recognition device according to claim 8, wherein the image generation source is an information processing device and an image reading device connected to the information processing device. 10. The character recognition device according to claim 1, further comprising correction means for correcting an erroneously recognized part in the character string obtained as a result of the recognition before the determination by the determination means. Character recognition device. 11. A character processing device according to claim 10, wherein said correction means corrects the character string to be corrected by character editing of insertion and deletion of characters. A character recognition device comprising: 12. The character recognition device according to claim 10, wherein said correction means determines a predetermined value between a character code string of an identification name stored in said storage means and a character code string obtained as a result of the character recognition. Detecting means for detecting an identifier having more than the number of matching characters, and character processing means for correcting the character code string of the detected identifier by replacing the character string containing the erroneously recognized part. Character recognition device characterized by the above-mentioned. 13. The character recognition device according to claim 1, wherein the identification name is an organization name. 14. The character recognition device according to claim 1, wherein the identification name is a slip name. 15. A character recognition method of a character recognition apparatus for recognizing characters of a plurality of types of forms read from a document of a form in which characters are described and having different formats and automatically discriminating the type of the format of the form. The identification form, which is described in the document of the form, is stored in a storage device in the character recognition device in the form of a character code in a form associated with a unique identification name for each type of the format in association with the type of the format, It is determined by the information processing device in the character recognition device whether or not there is a character code sequence that matches the identification name stored in the storage unit in the character code sequence obtained as a result of character recognition of the image, When an affirmative determination is obtained, the information processing device matches the matching character string based on the association between the format type and the identifier. A character recognition method for a character recognition device, characterized by acquiring a corresponding format type. 16. The character recognition method of the character recognition device according to claim 15, wherein a ruled line is described on the document of the form, and prior to the character recognition, the information processing device reads the read data. A character recognizing method for a character recognizing apparatus, wherein an image of a ruled line is deleted from a form image. 17. The character recognition method according to claim 16, wherein format information indicating a position of a ruled line is stored in the storage device in association with the format type, and the information processing device performs Based on the association between format type and format information,
A character recognition method for a character recognition device, comprising: obtaining format information corresponding to the obtained type of the format. 18. The character recognition method for a character recognition device according to claim 17, wherein the information processing device recognizes a cell image of a character image sandwiched between ruled lines at positions indicated by the acquired format information. Character recognition method of the character recognition device to perform. 19. The character recognition method for a character recognition device according to claim 18, wherein said character recognition device includes an image input device for inputting an image of said form. . 20. The character recognition method for a character recognition device according to claim 19, wherein said image input device is an image reading device for reading a document of said form. 21. The character recognition method according to claim 19, wherein the image input device is a communication device that receives an image of the form transmitted from an image generation source via a public telephone line. A character recognition method for a character recognition device, comprising: 22. The character recognition method according to claim 19, wherein the image input device is a communication device that receives an image of the form transmitted from an image source via a wide area network. A character recognition method for a character recognition device, comprising: 23. The character recognition method according to claim 22, wherein the image generation source is an information processing device and an image reading device connected to the information processing device. Character recognition method. 24. The character recognition method for a character recognition device according to claim 15, wherein the information processing device corrects an erroneously recognized portion in the character string obtained as a result of the recognition before the determination process. A character recognition method for a character recognition device, comprising: 25. The character recognition method according to claim 24, wherein the information processing apparatus corrects the character string to be corrected by inserting and deleting characters. A character recognition method of a character recognition device. 26. The character recognition method according to claim 24, wherein the information processing apparatus includes a character code string of an identification name stored in the storage device and a character code string obtained as a result of the character recognition. And detecting a distinguished name having a predetermined number or more of matching characters between the detected distinguished name and correcting the character code string of the detected distinguished name by replacing the character string including the misrecognized part. Character recognition method for character recognition device. 27. The character recognition method according to claim 15, wherein the identification name is an organization name. 28. The character recognition method according to claim 15, wherein the identification name is a slip name. 29. A character recognition device for recognizing a plurality of types of form images read from a document of a form having different characters and having different formats by an internal information processing apparatus and automatically discriminating the type of the form of the form. In the recording medium loaded with the program code means readable by the information processing device, a unique identifier for each type of the format, which is described in the document of the form, is associated with the type of the format. The program code unit is stored in a storage device in the character recognition device in the form of a character code, and the program code unit is stored in the storage unit in a character code string obtained as a character recognition result of the read form image. Processing step of determining whether there is a character code string that matches the identified name, Receiving a format type corresponding to the matching character string based on the association between the format type and the identifier when the recording is performed. Medium.