JP2004164376A - Identification-code-attached form, form reading program, and form creation program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a form reading program that can read a new form while maintaining similar reading precision for an old form without changes in setting information in a device even when the layout of the form is changed. <P>SOLUTION: The form reading program for recognizing characters in a form commands a computer to execute: a step of extracting a field on the form; a step of detecting an identification code of the field described in a given position to the field; a step of reading out prestored field attribute information corresponding to the identification code; and a step of recognizing contents of information entered in the field and an item designated for the field according to the field attribute information. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a form reading method that does not depend on a form format by enabling field data and field attributes of the form to be recognized.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, there are the following methods for reading characters and data described in a form format. For example, as disclosed in Japanese Patent Application Laid-Open No. 9-259221 (Patent Document 1), the position of a field of a form is extracted as a ruled line pattern, and the extracted ruled line pattern and a form format previously indicated in the apparatus by the ruled line pattern The type of the form format is identified by collating with the dictionary data stored as a feature of the form. Based on the identification result of the form format, the characters described on the data form are extracted, and the extracted characters are collated with the dictionary data held in advance as a feature of the character pattern in the apparatus, and the characters on the form are compared. Was aware of. That is, by identifying the type of the form, information for reading a character written on the form is obtained, and the character described on the form is recognized.
[0003]
In addition, there are the following methods for reading form data that can cope with a difference in form format. For example, as shown in Japanese Patent Application Laid-Open No. 2000-29986 (Patent Document 2), an item name described in a form is recognized by a character, and based on the format information of the form registered in advance in the apparatus, The item contents were read according to the type of the recognized item name. That is, by reading the item name itself, the attribute of the item (field) is determined, and the content of the item is read according to the attribute.
[0004]
In addition, as a technique related to the present invention, a method of extracting ruled lines is disclosed in Japanese Patent Application Laid-Open No. 2000-339406.
[0005]
[Patent Document 1]
JP-A-9-259221 (paragraph 0059, paragraph 0154, FIG. 8)
[Patent Document 2]
JP-A-2000-29986 (paragraph 0005)
[Patent Document 3]
JP-A-2000-339406 (paragraph 0032, paragraph 0075 to paragraph 0078)
[0006]
[Problems to be solved by the invention]
However, the conventional form reading method has the following problems. For example, in order to recognize a character written on a form by identifying a format of the form from a ruled line pattern, it is necessary to set position information of a ruled line (field) of the form in advance for each field. Therefore, when the layout on the form is changed (when the field position on the form is changed), it is necessary to change the setting information in the apparatus each time.
[0007]
In addition, in order to read the item contents according to the type of the item name by reading the item name described in the form, when reading complicated characters such as kanji in the character recognition of the item name, the character It was difficult to stabilize the reading accuracy.
[0008]
The present invention solves such a problem, and even if the layout on the form is changed (even if the field position on the form is changed), the old information is not changed without changing the setting information in the apparatus. It is an object of the present invention to provide a form reading method capable of reading a new form while maintaining the same reading accuracy as a form.
[0009]
[Means for Solving the Problems]
In order to solve the above problem, the present invention has the following configuration. The present invention is a form reading program for recognizing characters in a form,
Extracting fields on the form;
Detecting an identification code of the field described at a predetermined position with respect to the field;
Reading field attribute information corresponding to the identification code stored in advance,
Recognizing the contents of the information entered in the field and the items defined for the field based on the field attribute information;
On a computer.
[0010]
According to the present invention, first, each field on the form is extracted. An identification code for identifying the field described in a predetermined position with respect to the extracted field is detected. Field attribute information for specifying a field is stored in the device in advance. Based on the stored field attribute information, field attribute information corresponding to the detected identification code is read. Based on the read field attribute information, the contents of the information entered in the field and the items specified for this field are recognized. Therefore, even if the layout on the form is changed by detecting the identification code for identifying the information existing in the area from the field on the form, the same setting as the old form is performed without changing the setting in the apparatus. Can read the data in the form.
[0011]
Also, the form used in the present invention has a field for information entry, and a computer identifies the field and stores the information entered in the field and the item specified for the field at a predetermined position with respect to this field. A field identification code for reading field attribute information for recognition from the storage device is described.
[0012]
According to the present invention, an identifier for identifying a field (item) on a form is uniformly recognized as a code by adding an identification code (for example, a character code or a bar code) to each field present on the form. Can be. For example, items such as “name” and “name”, which are displayed differently depending on the form, can be uniformly recognized as “name”.
[0013]
The present invention is also a form creation program for creating the form,
Extracting fields on the form;
Reading an item of information written in the field stored in advance and an identification code of the field for recognizing the content of the information;
Adding the read identification code to a predetermined position with respect to the field;
On a computer.
[0014]
According to the present invention, first, each field on the form is extracted. In the apparatus, an item of information to be entered in a field and an identification code for each field for recognizing the content of the information are stored in advance. The identification code corresponding to the extracted field is read using the stored information. The read identification code is added to a predetermined position for each field. This makes it possible to create a form to which an identification code for identifying each field on the form is added.
[0015]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
(Embodiment 1)
FIG. 1 is a block diagram illustrating an example of a system of a form reading apparatus that provides a form reading method according to an embodiment of the present invention. The form reader 1 is connected to an image reader 10 as a peripheral device via an interface. The image reader 10 includes an image information input device (for example, a scanner) that can optically read a form. The interface is configured by a USB (Universal Serial Bus). The form reading device 1 functions by taking in a form from the image reader 10 as image data. Hereinafter, a configuration example of the form reading device 1 will be described with reference to FIG.
[0016]
<Form reader>
The form reading device 1 is configured by a computer such as a personal computer (PC) or a workstation (WS), and includes a CPU 11, a storage unit 12, an input / output unit (not shown), and a communication control unit. The form reader 1 is realized by executing various programs stored in the storage unit 12 on the CPU 11. The CPU 11 includes a ruled line extraction unit 11a, a field ID extraction unit 11b, an attribute determination unit 11c, and a character recognition unit 11d. The storage unit 12 includes a field ID position information file 12a and a field attribute information file 12b. Hereinafter, each component included in the form reading apparatus 1 will be described.
[0017]
《Rules extraction section》
The ruled line extracting unit 11a extracts a field in a form taken in as image data as a ruled line. Here, a square area (field frame) surrounded by ruled lines and content data described in the field are extracted as ruled lines. Ruled lines are extracted as coordinate data, and vertical ruled lines and horizontal ruled lines are indicated as start coordinates and end point coordinates. As a method for extracting the ruled line, for example, a method disclosed in Japanese Patent Application Laid-Open No. 2000-339406 can be applied.
[0018]
<< Field ID extractor >>
The field ID extraction unit 11b cuts out an area at a position separated by a certain distance from the lower right of the square area (field frame) extracted by the ruled line extraction unit 11a, recognizes a number in the cut out area, and sets it as a field ID. I do. That is, based on the ruled line coordinate data of the extracted field frame and the information stored in the field ID position information file 12a, the character (number) indicated at the lower right of the field frame is recognized as a character. Is set as an identifier for identifying the field.
[0019]
《Attribute determining unit》
The attribute determining unit 11c determines the attribute of the character string in the field from information associated with the field ID stored in the field attribute information file 12b based on the field ID extracted by the field ID extracting unit 11b. I do.
[0020]
《Character recognition section》
The character recognizing unit 11d recognizes, as a character, the content data described in the field recognized by the ruled line extracting unit 11a based on the attribute of the character string in the field determined by the attribute determining unit 11c.
[0021]
<Format of form>
FIG. 2 is a diagram showing an example of a form format used in the embodiment of the present invention. This form is a new application form used when opening a new account at a bank. The form includes fields (items) for entering the name, address, deposit amount, and account type of a customer who opens a new account. The type of account is indicated by general, ordinary, or regular. In the vicinity of each field in the form, for example, at the lower right, an ID is added as an identifier for identifying the contents of the field. For example, as shown in FIG. 2, this ID is indicated by a combination of a plurality of digits (three digits in this example), and functions as a field ID indicating a field attribute.
[0022]
<Data structure>
Next, the data structure of the field ID position information file 12a and the field attribute information file 12b provided in the storage unit 12 will be described.
[0023]
FIG. 3 is an example of a data structure indicating field ID position information. Field ID position information is defined in the field position information file 12a. The field ID position information indicates the offset amounts of the start point coordinates (X, Y) and the end point coordinates (X, Y) with respect to the ruled line coordinate data (X, Y) of the extracted field frame. That is, in the field ID position information, information for determining at which position the ID is located for each field ruled line is defined. Therefore, the field ID position information is fixedly defined for all fields, and only one field ID needs to be defined for the entire system. Also, a plurality of pieces of field position information may be set.
[0024]
Further, unlike the conventional field attribute information shown in FIG. 4B, it is not necessary to provide a parameter indicating a position for each field.
[0025]
FIG. 4A is an example of a data structure showing field attribute information, and FIG. 4B is a diagram showing conventional field attribute information as a comparative example. Field attribute information is defined in the field attribute information file 12b. In the field attribute information, as shown in FIG. 4, information such as an item ID, a character attribute, the number of characters, knowledge conversion, and a knowledge dictionary are defined in association with each field ID. The item ID is an identification number for the program to recognize the data. The character attribute and the number of characters are information indicating the attribute of the character string entered in the field. For example, the character attribute indicates information indicating whether the character is a character (NCHR) or a number (MUN). The number of characters indicates the number of characters in a column. The knowledge conversion indicates whether or not there is information (knowledge dictionary) that is a clue when recognizing characters in the field. In the knowledge conversion, the presence or absence of a knowledge dictionary is indicated by Y and N. As shown in FIG. 4, the knowledge dictionary indicates information indicating the corresponding knowledge dictionary when there is a knowledge dictionary (Y). Here, an example in which the content (item) of the field is NAME (name) is shown. The type of the knowledge dictionary is determined by the type of item entered in the field. For example, ADDRESS (address) can be used as another knowledge dictionary. For example, in the case where the contents (items) of the field correspond to ADDRESS, and the field contents are recognized in character recognition, the entry data of "Ota-ku, Tokyo" is recognized as a regular place name of "Ota-ku, Tokyo". Function.
[0026]
<Processing flow>
Next, the processing of the form reading apparatus 1 will be described with reference to FIGS.
[0027]
《Form reading processing》
FIG. 5 is an example of a flowchart showing a form reading process in the form reading apparatus. This processing is performed by the CPU 11 in the form reading apparatus 1 executing various programs for implementing the ruled line extracting unit 11a, the field ID extracting unit 11b, the attribute determining unit 11c, and the character recognizing unit 11d. .
[0028]
First, a form format is read from the image reader 10 as image data, and is taken into the storage unit 12 of the form reading apparatus 1 (S1).
[0029]
In the form reading device 1, the ruled line extracting unit 11a extracts ruled lines defining each field from the image data of the fetched form (S2). That is, a square area (field frame) surrounded by ruled lines and content data written in the field are extracted as ruled lines. At this time, the starting point and the ending point of the ruled line are extracted as coordinate data (X, Y). The extracted field frame and the content data in the field are stored in the storage unit 12 in pairs.
[0030]
Next, the field ID extraction unit 11b extracts the field ID added to the lower right of each field (item) based on the ruled line coordinate data of the field extracted in S2 and the field ID position information ( S3). That is, the position of the field ID added to each field (item) is extracted, the number indicated at the extracted position is recognized as a character, and the recognized character (number) is used as the field ID for the field. Set.
[0031]
Next, the attribute determining unit 11c determines the attribute of the character (character string) corresponding to the field ID set in S3 based on the field attribute information (S4). Here, information (character attribute, number of characters, knowledge conversion, knowledge dictionary) necessary for character recognition associated with each field ID defined as field attribute information and an item ID are read. These pieces of information necessary for character recognition are passed to the character recognition unit 11d together with an item ID that serves as identification information for the processing program (character recognition unit) to recognize the information. The character recognizing unit 11d recognizes the content data in the field as a character by associating the passed information necessary for character recognition with the content data in the field recognized by the ruled line extracting unit 11a (S5).
[0032]
Note that the above-described extraction of the ruled line in S2 corresponds to the extraction step of the present invention. The extraction of the field ID in S3 corresponds to the detecting step of the present invention. Further, the determination of the attribute of the character (character string) in S4 corresponds to the step of reading out the field information of the present invention. The character recognition of the content data in the field in S5 corresponds to the step of recognizing the item of the present invention.
[0033]
As a result, the form reading apparatus 1 extracts the field ID from the ruled line data of each field in the form, thereby recognizing the attribute of the character string for the field based on the correspondence with the field ID defined in advance. By using the attribute of the recognized character string, the character data of the content data in the field can be recognized. Further, the item of the character string recognized by the item ID can be identified. Therefore, even when the position of a field is different depending on the form (when the layout is changed), it is possible to recognize the character string written in each field and its item regardless of the field position in the form. it can. Hereinafter, the process of extracting the field ID in S3 will be described in detail.
[0034]
<< Field ID extraction processing >>
FIG. 6 is an example of a flowchart showing the extraction processing of the field ID. This processing is mainly performed by the field ID extraction unit 11b. In S3 shown in FIG. 5, when the rectangular area (field) surrounded by the ruled line is extracted as coordinate data in S2, only the end point coordinates (X, Y) constituting the ruled line are acquired (S10). . That is, the lower right of the square area (field) surrounded by the ruled line is obtained as the end point coordinates (X, Y). The acquired end point coordinates (X, Y) serve as a reference for specifying the position of the field ID. Based on the acquired end point coordinates (X, Y) and the data defined in the field ID position information, the starting point of the area where the field ID exists for each field is calculated (S11). Here, the offset amounts of the coordinates X and Y with respect to the starting point defined in the field ID position information are added to the acquired end point coordinates (X, Y). Next, based on the end point coordinates (X, Y) acquired in S10 and the data defined in the field ID position information, the end point of the area where the field ID exists for each field is calculated (S12). Here, the offset amounts of the coordinates X, Y with respect to the end point defined in the field ID position information are respectively added to the obtained end point coordinates (X, Y). In S11 and S12, the field ID existing in the area composed of the calculated starting point (X) of the field ID and the ending point (Y) of the field ID is recognized as a combination of numbers (S13). The combination of the recognized numbers is set as a field ID for identifying the field (S14).
[0035]
As a result, the field ID can be extracted as characters from the coordinate data of the ruled lines forming the field frame. Further, since the field ID is indicated by combining specified numbers of 0 to 9, the field ID can be recognized at a higher recognition rate than characters (such as hiragana and kanji).
[0036]
<Modification>
In the above embodiment, as shown in FIG. 2, the field ID for the field in the form format is a combination of numbers. However, in the present invention, as shown in FIG. 7, the field ID may be a bar code instead of a combination of numbers. The position where the barcode, which is the field ID, is added may be, for example, on a ruled line as shown in FIG. As a result, the field ID for each field can be recognized as a barcode, and the field ID recognition rate can be further increased.
[0037]
<How to create a form>
Next, a method of creating a form in which a field ID is entered will be described. FIG. 8 is a diagram illustrating a configuration example of a form creation device. The form creation device 2 can use the configuration of the form reading device as shown in FIG. The differences from the form reader 1 are as follows.
[0038]
(A) The storage unit 12 stores image data of a form in which no field ID is entered. However, the image data can be obtained by downloading, reading from a portable recording medium, or the like, regardless of the obtaining route.
[0039]
(B) The field attribute information file 12c stored in the storage unit 12 has a structure as a correspondence table between field IDs and field attribute information as shown in FIG. 8 unlike FIG. 4A. However, the field ID position information file 12a and the field attribute information file 12b may have a separate table 12d of field ID and field attribute information as shown in FIG.
[0040]
(C) The CPU 11 realizes a field extraction unit 11a, a field ID determination unit 11e, a field ID addition unit 11f, and a print control unit 11g by executing various programs, and controls a printer connected to the form creation device. The form image can be printed on a sheet.
[0041]
An operation example of the form creation device as described above is as follows. FIG. 10 is a flowchart illustrating an operation example.
[0042]
(S21) When the image data of the form is read from the storage unit 12, the field extraction unit 11a extracts each field (ruled line frame) in the image data by the same method as described above.
[0043]
(S22) The field ID determination unit 11e refers to the field attribute information file 12c and reads out the field ID corresponding to the attribute information of each field extracted in the field attribute information S21.
[0044]
(S23) The field ID addition unit sets the field ID corresponding to each field at the position specified by the information of the field ID position information file 12a with respect to the read image data of the form. The image data of the read form is edited and a field ID is added to the form. The edited image data may be stored in the storage unit 12.
[0045]
(S24) The print control unit 11g controls the printer, and causes the printer to print out the edited image data.
[0046]
Note that the image data created in S23 may be stored in the storage unit 12, and the processing in S24 may be performed as needed. Further, the processing of S21 to S23 may be repeatedly performed for each field. Further, the creation device may add a field ID to a form by additionally printing a field ID on a form (not including the field ID) printed on the sheet. In this case, S23 for editing the image data may be omitted.
[0047]
Note that the extraction of the fields in S21 described above corresponds to the extraction step of the present invention. The reading of the field ID in S22 corresponds to the step of reading the identification code of the present invention. The addition of the field ID in S23 corresponds to the adding step of the present invention.
[0048]
【The invention's effect】
According to the present invention, even if the layout on the form is changed (even if the position of the field on the form is changed), the device is extracted by extracting the field ID for each field from the field ruled line in the form. It is possible to read the new form without changing the setting information in the section, while maintaining the same reading accuracy as the old form.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating an example of a system of a form reading apparatus that provides a form reading method according to an embodiment of the present invention.
FIG. 2 is a diagram showing a form format example (1) used in an embodiment of the present invention.
FIG. 3 is an example of a data structure showing field ID position information.
FIG. 4 is an example of a data structure showing field attribute information.
FIG. 5 is an example of a flowchart showing a form reading process in the form reading apparatus.
FIG. 6 is a flowchart example showing a field ID extraction process.
FIG. 7 is a diagram showing a form format example (2) used in the embodiment of the present invention.
FIG. 8 is a diagram illustrating a configuration example of a form creation device.
FIG. 9 is a diagram illustrating an example of a correspondence table between a field ID and field attribute information.
FIG. 10 is a flowchart example showing an operation example in the form creation device.
[Explanation of symbols]
DESCRIPTION OF REFERENCE NUMERALS 1 form reading device 2 form creating device 10 image reader 11 CPU
11a Ruled line extraction unit (field extraction unit)
11b Field ID extraction unit 11c Attribute determination unit 11d Character recognition unit 11e Field ID determination unit 11f Field ID addition unit 11g Print control unit 12 Storage unit 12a Field ID position information file 12b Field attribute information file (form reader)
12c Field attribute information file (form creation device)
12d Correspondence table between field ID and field attribute information

Claims (3)

In a form reading program for recognizing characters in a form,
Extracting fields on the form;
Detecting an identification code of the field described in a predetermined position with respect to the field;
Reading field attribute information corresponding to the identification code stored in advance,
Recognizing the contents of the information entered in the field and the items defined for the field based on the field attribute information;
And a computer for executing the program. A storage device having a field for information entry and storing, at a predetermined position with respect to this field, a computer for identifying the field and for recognizing the information entered in the field and the field attribute information for recognizing an item specified for the field. A form in which the identification code of the field to be read from is described. In a form creation program that creates a form,
Extracting fields on the form;
Reading an item of information written in the field stored in advance and an identification code of the field for recognizing the content of the information;
Adding the read identification code to a predetermined position with respect to the field;
Which is executed by a computer.

Images