JP2007087021A - Electronic documentation device for paper document, electronic documentation method for paper document, and electronic documentation program for paper document - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent fetching of a different kind of electronic document (e.g., estimation sheet) to a scheduled kind of electronic document file (e.g., delivery slip) formed by reading a paper document by a scanner or the like. <P>SOLUTION: Position information of each specific area (a text segment area 12, a table segment area 13, and a drawing segment area 14) of a paper document model form of a computerization object file and format information showing contents (a text, a table column composition of information for rows and lines, and a pointer to a storage area of text data) are preliminarily set. Paper document read data is stored in a bit map memory, and a character, pattern, image or the like of the memory data is recognized for each specific area, and the result is compared with a corresponding part of the format information to determine whether the paper document is of the scheduled kind or not. In the format information, the determination object range of the paper document read data can be selectively instructed in a specific area unit-basis of the original paper document or in a component unit-basis of each specific area. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an electronic documentation device, an electronic documentation method, an electronic documentation program, and the like for reading information written on a paper document with a scanner or the like into electronic document data.

  In particular, when an existing paper document such as various file units is converted into an electronic document, a document that should not be included in the paper document is mistakenly included (for example, a document file of a resident in Shinjuku ward lives in Shibuya ward) If the person's document is mixed), the misunderstood document is detected and prevented from being captured.

  Currently, many companies have a large amount of paper documents. With the enforcement of the Electronic Document Law, it is assumed that the work of converting these paper documents into electronic documents with a scanner or the like will be started in many companies and offices.

  At this time, the problem is the validity of the digitized document content. Ideally, although it is better for the user to visually check each digitized document image, this is not practical when there are hundreds of thousands of paper documents. The “document image” includes a binary image, a grayscale image, and a color image.

  Actually, a large amount of paper documents are read and stored in the form of electronic data with a high-speed scanner, etc., but the scan operation parameter settings are incorrect, and the paper document file to be digitized is not related to the file ( For example, it is possible that the validity of the electronic document file cannot be ensured due to a human error such as that a misdocumented document is originally included.

  Originally, the biggest merit of digitizing paper documents is to reduce the storage cost of paper, so we want to discard the paper after digitization, but there are human errors as mentioned above Given that, there remains concern about the disposal of digitized paper documents. However, if all the electronic documents are visually confirmed individually, the cost is rather high.

  In order to dispel such anxiety, it is determined whether or not the digitization of the paper document has been normally performed only on the scheduled type of paper document (for example, a delivery note for May 2004). When the possibility of a malfunction is detected, it is notified to the user, and the user only has to visually confirm only the sheet.

  Conventionally, when a paper document is optically read and converted into electronic data, the so-called accuracy index of the reading operation is stored in a register, and the output value is compared with a predetermined reference value to thereby compare the read data of the paper document. The quality is judged (for example, refer to Patent Document 1).

The determination target here is whether or not the read data (image data) of the entire paper document has an image quality within an allowable range.
Japanese Patent Laid-Open No. 4-37964

  Thus, the quality check in the conventional electronic conversion of paper documents is the overall validity of the image data itself output by the optical reading means.

  In other words, the validity of the type of document content, such as whether or not other types of paper documents are mixed in a number of types of paper documents to be converted into electronic data (for example, other than the monthly report in 2003) Is not automatically detected).

  For this reason, when a paper document file that has been organized by type is read with a scanner, etc., and converted into electronic data, if another type of paper document is mistakenly included in the file, the electronic document is stored in the imported file state. It will be converted.

  Therefore, in the present invention, the specific area portion of the read data of the paper document is compared with the corresponding portion of the format information of the paper document to determine whether the paper document is of a scheduled type (for example, contract). This prevents electronic documents of another type from being taken into an electronic document file of an arbitrary scheduled type, so that electronic data conversion processing can be performed efficiently and with high accuracy for each type of many paper documents. The purpose is to.

The present invention solves the above problems as follows.
(1) In an electronic documentation device that converts a paper document into an electronic document,
Means (for example, a format information table 3c described later) for holding format information in a form in which the document content is extracted for each specific area in the paper document;
Means for reading the paper document (for example, reading means 1 described later);
Means for determining whether the paper document being processed is of a scheduled type by comparing the read data of the specific area with the corresponding portion of the format information extracted from the holding means (for example, a character A pattern determination unit 3a and an image determination unit 3b).
(2) In (1) above,
As the format information, information including an instruction content indicating whether or not to make the determination processing target in units of individual specific areas is used.
(3) In (1) above,
As the format information, information including an instruction content indicating whether or not to make it a target of the determination process is used in units of components of the document content in the specific area.
(4) In an electronic documenting method in which a computer executes a determination process of a specific area portion of read data of a paper document to check whether the paper document is of a scheduled type,
By the computer
Recognizing the content of the read data of the specific area portion;
Based on the format information holding means in which the document contents are extracted for each specific area in the paper document, the scheduled type is determined by comparing the recognition result with the corresponding portion of the format information. Steps,
Execute.
(5) In an electronic documentation program for executing a determination process for a specific area portion of read data of a paper document and confirming whether the paper document is of a scheduled type,
Recognizing the content of the read data of the specific area portion;
Based on the format information holding means in which the document contents are extracted for each specific area in the paper document, the scheduled type is determined by comparing the recognition result with the corresponding portion of the format information. ,
Let the computer realize the function.

  As described above, according to the present invention, the read data of a specific area in a paper document (for example, an area to which a predetermined logo mark is attached in a predetermined type of paper document) is used as the content of the format information of the paper document (for example, the specific area). In the electronic document file of an arbitrary scheduled type (for example, a delivery note of Company A). Documents (for example, delivery notes from company B) are prevented from being taken in.

  Further, in the format information, the determination target range of the paper document read data can be selectively instructed in units of specific areas of the original paper document or constituent elements of the specific areas (see FIG. 3). Therefore, the setting key for specifying the scheduled type to be electronically documented is given a degree of freedom to enhance the convenience on the user side.

  The best mode for carrying out the present invention will be described with reference to FIGS.

  FIG. 1 is an explanatory diagram showing the overall configuration of the electronic document processing apparatus.

In FIG.
1 is a high-performance scanner that reads a paper document of an electronic document file in the form of image data, a reading means such as a digital camera,
2 is the input / output interface,
3, based on the paper document output data (= electronic document data) of the reading unit 1, executes a determination process to determine whether the content of the paper document matches the format information (see FIG. 3) of the template file of the target file. Checking means to do,
3a processes the data elements (= corresponding pixel data of the bitmap memory 4a) of the text segment area and the table segment area (see FIG. 2) in the paper document output data by various known methods, A character / pattern judging unit for extracting table column information (see FIG. 3);
3b processes image data elements (= corresponding pixel data of the bitmap memory 4) in the figure segment area (see FIG. 2) in the paper document output data by various known methods, and extracts, for example, contour information there. Image judgment unit,
3c is a format information table (see FIG. 3) that holds the format information of the template form of the target file.
Reference numeral 4 denotes reference data for checking the paper document output data, storage means for holding data being processed and processing result data of the check means 3, electronic document data that has been checked, etc.
4a is a bitmap memory for holding the paper document output data,
5 is an input means such as a mouse or a keyboard used when the user specifies the format information (see FIG. 3) of the template form,
6 is a display means for displaying the paper document output data and paper document judgment result on the screen,
6a is a frame memory for storing screen display data,
6b is a display screen for displaying data stored in the frame memory 6a.
Respectively.

  In the electronic document system of FIG. 1, for example, each paper document managed in file units for each month in 2004 is read by the reading means 1, and the output data is sent to the checking means 3 to be bitmap memory. 4a.

  The check means 3 determines whether or not the paper document output data matches the contents of the format information table 3c (see FIG. 3).

  Then, since the determination result is output to the display screen 6b, the user can confirm whether or not anything other than the target file is mixed in the bundle of digitized paper documents according to the output contents. it can.

For example, in an electronic document paper document,
(11) Was a paper document without a specific logo mark inserted in it?
(12) For example, it is possible to check whether “paper documents other than contracts” and “contracts other than May 2004” were mixed in the electronic document of the contract file for May 2004 It is.

As will be described later, the checking means 3 is
In the case of the check processing of (11) above, the image determination unit 3b performs the paper document output image data of the figure segment area (see FIG. 2) designated by the format information table 3c and the template image stored in the storage means 4. Check the difference with the data,
In the case of the check process (12) above, the character / pattern determining unit 3a is different from the paper document output text data in the text segment area (see FIG. 2) that is also designated and the template text data in the format information table 3c. Check out.

  FIG. 2 is an explanatory diagram showing a template form of a paper document to be electronically documented.

In FIG.
11 is a template form of a paper document (for example, a contract, an order form, a table of contents page of a magazine, a product description, etc.) filed for each type,
12 is the text segment area of the form,
13 is the table segment area of the form,
14 is the figure segment area of the form,
Respectively.

  Here, each of the segment areas 12 to 14 has a rectangular shape, and as shown in FIG. 3, for example, the position coordinates (start coordinates and end coordinates) of the lower left point s and the upper right point e of the segment areas, respectively. ).

  FIG. 3 is an explanatory diagram showing a format information table of the template form shown in FIG.

Each record of this format information
・ Start coordinates (corresponding x and y coordinates on bitmap memory)
-End coordinates (corresponding x and y coordinates on the bitmap memory)
-Type (type of reference data for judgment such as text, table, and image)
-Text in the text segment area 12-Pointer to the image data storage section (storage means 4) in the figure segment area 14-Table column information in the table segment area 13 (number of rows and columns in the table)
-It consists of items (fields) such as mandatory and judgment flags.

  Each item data of the text, the pointer, and the table column information (the data in the upper part of the drawing) indicates the reference contents in the comparison determination process in step (s9) of FIG. The item data of “pointer” indicates the storage location of the reference content.

  The “required” item of each record indicates whether or not each record is a target of the check process of FIG. “Yes” is set for the check target record, and “No” is set for the non-check target record.

  Furthermore, each of the essential information “YYYNYYY” in the text item of the record of number 1 corresponds to each element of the upper text data “contract (for May)”, and the fourth “N” in the essential information. "Indicates that the fourth element" 5 "of the text data is not the target of the check process of FIG. “Y” is a target of the check process.

  Similarly, “N” in the essential information in the table column information item of the record of the number 2 is not set to the column element “1” of the table column information (3 rows and 1 column) as the check processing target in FIG. It is shown that. “Y” is a target of the check process.

  Further, the essential information “Y” in the pointer item of the record of number 3 indicates that the image data is the target of the check process of FIG.

That is, when the electronic document processing is executed based on the format information of FIG.
(21) With respect to the text segment area 12, the “N (contract) (May)” of the original display content such as “Contract (April)” is set, for example, by setting “N” in the text item of the record of number 1. ", The paper documents that differ only in the number of months are also imported as the scheduled file documents,
(22) For the table segment area 13, by setting the above “N” in the table column information item of the record with the number 2, for example, a table with 3 rows and 3 columns other than “1”, such as 3 rows and 2 columns. A paper document in a form is also taken in as a scheduled file document.

FIG. 4 is an explanatory diagram showing a processing procedure in the checking means 3 for the output data (electronic document data) of the reading means 1, and the contents are as follows.
(s1) The determination flag of each record in the format information table 3c is initialized to “0: not determined”, and the process proceeds to the next step.
(s2) The electronic document data (one paper document) read by the reading means 1 is held in the bitmap memory 4a, and the process proceeds to the next step.
(s3) It is determined whether or not a record of “determination flag = 0” and “required item = Y” exists in the format information table 3c. If “yes”, the process proceeds to the next step, and “no” In this case, the process proceeds to the next electronic document data processing.
(s4) Select one arbitrary record corresponding to “yes” in the previous step, set “1: determined” for the determination flag, and proceed to the next step.
(s5) If the type of the selected record is “text” or “table”, the process proceeds to the next step, and if it is “image”, the process proceeds to step (s7).
(s6) The character / pattern determination unit 3a executes pattern recognition processing for each pixel data of the bitmap memory 4a corresponding to the segment area specified by the start / end coordinates and end coordinates of the selected record, Proceed to s8).
(s7) Image recognition processing for each pixel data in the bitmap memory 4a corresponding to the segment area specified by the start coordinate and the end coordinate of the selected record is executed by the image determination unit 3b, and the process proceeds to the next step. .
(s8) It is determined whether or not the predetermined character / pattern recognition process or image recognition process has been executed, that is, whether or not a process result exceeding the predetermined standard content has been obtained. If “no”, proceed to step (s10).
(s9) Whether each processing result of character / pattern recognition / image recognition is matched with the text data, table column information or image data (of the pointer destination storage means 4) of the selected record If “yes”, the process returns to step (s3). If “no”, the process proceeds to the next step. Note that the elements for which the essential information in the selected record is set to “N” are not determined here.
(s10) A screen display is made to the effect that the electronic document data in step (s2) is of a paper document other than the scheduled file, and the process proceeds to the next electronic document data processing.

  In the case of pattern recognition processing in step (s6), for example, the number of linear portions in each of the x-axis direction and the y-axis direction is obtained, and then the table column information (row, column) corresponding thereto is specified.

  In the case of the comparison process between the image recognition result in step (s9) and the image data (in the pointer destination storage means 4), for example, the degree of coincidence of the binary data between the corresponding pixels is obtained, and this is a predetermined value. When the above is indicated, it is determined as “match”.

  As described above, when the check process of FIG. 4 is executed using the format information of FIG. 3, the essential information of the fourth “5” of the text item of the record “1” is set to “N”. In step (s9), the fourth corresponding data element is not determined. That is, when the text segment corresponding area (identified by the record) in the bitmap memory 4a is recognized as text information such as “contract (for January)” and “contract (for June)” other than May. The determination result at this step is “yes”.

  Similarly, since the essential information of the “column” of the table column information item of the record “2” is set to “N”, the data element corresponding to this column is not determined in step (s9). That is, when the table segment corresponding area (identified by the record) in the bitmap memory 4a is recognized as table column information such as “3 rows 2 columns” or “3 rows 3 columns” other than 3 rows 1 columns The determination result at this step is “yes”.

  As described above, when a paper document collected in a file unit of a certain content is converted into electronic data by a scanner or the like, for each record of the format information table 3c of the template form of the unit and for each data element in the record 4 can be set as a determination target in step (s9) of FIG.

  For example, if “No” is described in the required item of the record “1” in FIG. 3 (instead of “Yes” in the figure), the electronic document data in the text segment area 12 is stored regardless of the description content of the required information in the record. It is excluded from the judgment in step (s9). In this case, the contents of the text segment area of the paper document to be converted into electronic data are so-called “don't care”.

  Further, if “Yes” is described in the mandatory item of the record “1” and “YYNNNNNN” is described as the essential information of the record, “contract” of the text item “contract (for May)” of the record. Only becomes the reference data in the determination process of step (s9).

  Note that when identifying the bitmap memory area to be recognized in steps (s6) and (s7) in FIG. 4, the user recognizes the recognition process in the display screen 6b while viewing the first read document image, for example. The range may be indicated by the input means 5 (such as a mouse).

  If the determination result in step (s8) is “no”, it is suspected that the error is caused by a parameter setting error of the reading means 1, so that the fact is displayed on the screen and a scanner or the like is displayed on the user side. A parameter check may be prompted.

And, as a recording medium for storing the above electronic document processing program,
・ Program provider's database (DASD etc.)
Any of various types of portable recording media, such as a RAM or a hard disk on the computer main body side. The program is loaded into the computer main body and executed on the main memory.
(Supplementary note 1) In an electronic document converting apparatus for converting a paper document into an electronic document, means for holding format information in a form obtained by extracting the document contents of each specific area in the paper document; The content of the specific area portion of the read data of the paper document and the read data of the paper document is recognized, and the recognition result is compared with the corresponding portion of the format information of the holding means, so that the paper document is of a predetermined type. And a means for determining whether or not there is a paper document electronic documenting device.
(Supplementary note 2) The electronic information of the paper document according to supplementary note 1, wherein the format information includes an instruction content as to whether or not each of the specific areas is a target of the determination process. Device.
(Supplementary note 3) The paper according to supplementary note 1, wherein the format information is a unit of a component of the document content of the specific area and includes an instruction content indicating whether or not to make the determination processing target. An electronic documentation device for documents.
(Supplementary note 4) In an electronic documenting method in which a computer executes a determination process of a specific area portion of read data of a paper document to check whether the paper document is of a scheduled type, the computer Based on the step of recognizing the contents of the read data of the area portion and the format information holding means in which the document contents are extracted for each specific area in the paper document, the corresponding portion of the recognition result and the format information And a step of making a judgment on the scheduled type by collating with each other.
(Supplementary note 5) The electronic information of the paper document according to supplementary note 4, wherein the format information includes an instruction content as to whether or not each of the specific areas is a target of the determination process. Method.
(Supplementary note 6) The paper according to supplementary note 4, wherein the format information is a unit of a component of the document content of the specific area and includes an instruction content indicating whether or not to make the determination processing target. A method for electronic documentation of documents.
(Supplementary note 7) In an electronic documentation program for executing a determination process for a specific area portion of read data of a paper document and confirming whether the paper document is of a predetermined type, the read data of the specific area portion And recognizing the schedule by comparing the recognition result with the corresponding portion of the format information based on the format information holding means in which the content of the document is extracted for each specific area in the paper document. A program for electronic documentization of a paper document, characterized in that it is for making a computer realize the function of making a judgment about the type.
(Supplementary note 8) The electronic document of the paper document according to supplementary note 7, wherein the format information includes an instruction content as to whether or not to make the determination process an individual unit of the specific area. Program.
(Supplementary note 9) The paper according to supplementary note 7, wherein the format information is a unit of a component of the document content of the specific area and includes an instruction content indicating whether or not to make the determination processing target. A program for electronic documentation of documents.

It is explanatory drawing which shows the whole structure of the electronic documentation apparatus of this invention. It is explanatory drawing which shows the model form of the paper document of electronic document object of this invention. It is explanatory drawing which shows the format information table of the model form of FIG. 2 of this invention. It is explanatory drawing which shows the process sequence in the check means 3 with respect to the output data (electronic document data) of the reading means 1 of this invention.

Explanation of symbols

1: Reading means 2 such as a scanner or digital camera 2: Input / output interface 3: Check means 3a: Character / pattern judging section 3b: Image judging section 3c: Format information table (see FIG. 3)
4: storage means 4a: bitmap memory 5: input means such as a mouse or keyboard 6: display means 6a: frame memory 6b: display screen 11: template form 12 of paper document: text segment area 13: table segment area 14: figure Segment area

Claims (5)

In an electronic documentation device that converts paper documents into electronic documents,
Means for holding format information in a form in which the document content is extracted for each specific area in the paper document;
Means for reading the paper document;
Whether the paper document is of a scheduled type by recognizing the contents of the specific area portion of the read data of the paper document and comparing the recognition result with the corresponding portion of the format information of the holding means Means to judge,
An electronic documentation apparatus for paper documents, comprising: The format information includes an instruction content as to whether or not to make it a target of the determination process in units of individual specific areas.
The paper document electronic document forming apparatus according to claim 1. The format information is a unit of a component of document content in the specific area and includes an instruction content indicating whether or not to make it a target of the determination process.
The paper document electronic document forming apparatus according to claim 1. In an electronic documenting method in which a computer executes a determination process of a specific area portion of read data of a paper document to check whether the paper document is of a scheduled type,
The computer
Recognizing the content of the read data of the specific area portion;
Based on the format information holding means in which the document contents are extracted for each specific area in the paper document, the scheduled type is determined by comparing the recognition result with the corresponding portion of the format information. Steps,
An electronic documenting method for a paper document, characterized in that: In an electronic documentation program for executing a determination process for a specific area portion of read data of a paper document and confirming whether the paper document is of a scheduled type,
Recognizing the content of the read data of the specific area portion;
Based on the format information holding means in which the document contents are extracted for each specific area in the paper document, the scheduled type is determined by comparing the recognition result with the corresponding portion of the format information. ,
A program for electronic documentation of a paper document, characterized in that the function is realized by a computer.

Images