JP2005322083A - Document attribute input device and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily verify contents of initially applied attributes of a document while browsing the same document. <P>SOLUTION: Image data or character recognition result data are displayed in a frame F1, and a verification time attribute input table 154 is displayed in a frame F2. Words and phrases or typical words and phrases highlighted in a document are clicked, and transferred to an input field 154a. A keyboard input may be operated. An initially applied attribute value is displayed in display columns 154b and 154c, and when there is no variance in initial several times of attribute assignment, it is set as an input value. When there is variance, attribute names of the pertinent attributes are highlighted. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a document attribute input technique for assigning attributes to a document, and in particular, enables easy verification of the results of attribute assignment work by a plurality of persons or a plurality of times.

  In order to use the office space effectively, it has been proposed to store paper documents in a relatively low-cost warehouse at a relatively low cost, or to discard them and digitize the contents for use. The digitization of paper documents is preferable for reducing office costs and sharing information. By the way, in order to manage the digitized document, it is preferable to extract and assign important matters for each document. Therefore, in order to extract such important matters (document attributes), it is necessary for the operator to read the document, extract the important matters, and further input them. If such attribute assignment work is wrong, it will be difficult to search for the target document, so if multiple people give duplicate attribute assignment work and the assignment contents differ among workers, verification It is important for the person to verify further.

As a patent document related to the present invention, there is a technique for displaying clinical data in other clinical departments and checking whether there is an error in input contents when inputting medical data (Patent Document 1). However, there is no description about changing the display mode according to the difference between a plurality of attribute input values so that the verification is easy.
JP-A-7-262276

  The present invention has been made in view of the above circumstances, and an object thereof is to enable easy verification of the contents of the initial attribute assignment.

  According to this invention, in order to achieve the above-mentioned object, the configuration as described in the claims is adopted. Here, prior to describing the invention in detail, supplementary explanations of the claims will be given.

  That is, according to one aspect of the present invention, in order to achieve the above object, the document attribute input device includes: an attribute storage unit that stores an attribute value assigned to a document for each attribute assignment process; A document designating unit for designating a document; an attribute value given by the attribute giving process for the document designated by the document designating unit and an attribute value input field for the verifier to enter the attribute value in an attribute input form display area A verification attribute input form display means for displaying; a document display means for displaying a document designated by the document designation means in a document display area; a user's entry in the attribute value input field displayed in the attribute input form display area; Attribute value input means for inputting an attribute value based on an operation is provided.

  In this configuration, the content of the initial attribute assignment can be easily verified while browsing the document.

  The attribute assignment process to be verified is usually performed twice or a plurality of times. Each attribute assignment process is typically performed by different workers, but the same worker may repeatedly perform the processes.

  The document attribute input device may be an independent device accommodated in a single housing, or a plurality of devices accommodated in different housings may constitute a document attribute input device.

  In this configuration, when attribute values assigned by different attribute assigning processes match, it is preferable to input the matching attribute value in the attribute value input field. In this way, input during verification can be reduced. Of course, overwriting may be performed on the matched attribute value.

  When attribute values assigned by different attribute assigning processes do not match, it is preferable to highlight (highlight) the attribute value input field. For example, the attribute value input field accompanies the attribute name of the attribute value to be input to the attribute value input field, and highlights the attribute value input field by making the attribute name different from the display attribute.

  The document may be a document composed of image data obtained by optically reading a page of a paper document. A document generated as a character code from the beginning may be used.

  The document includes a first document composed of image data obtained by optically reading a page of a paper document, and a second document composed of character recognition result data obtained by character recognition of the image data, The first document and the second document may be switched and displayed in the document display area.

  The second document composed of the character recognition result data is displayed in a manner in which a predetermined word is highlighted in the document display area, a click operation on the highlighted word, and the attribute value input field in the attribute input form display area The above phrase may be transferred to the attribute value input field based on a click operation on.

  The attribute input form display area and the document display area are arranged side by side, and the attribute value input field in the attribute input form display area is closer to the document display area than the attribute value assigned by the attribute assignment process. You may make it arrange | position to.

  When the original attribute assignment is transferred to the input field of the posting destination by clicking on the posting source phrase, the posting source phrase is highlighted in the displayed document at the time of verification, so that the posting source can be immediately grasped. It is preferable that the verification work can be easily performed.

  Whether the attribute value can be accepted as an attribute value when the attribute value input means inputs the attribute value based on a user operation in the attribute value input field displayed in the attribute input form display area. You may make it discriminate | determine. Based on the determination result, input is prohibited or a prompt is given.

  The present invention can be realized not only as an apparatus or a system but also as a method. Of course, a part of the invention can be configured as software. Of course, software products used to cause a computer to execute such software are also included in the technical scope of the present invention.

  These and other aspects of the invention are set forth in the appended claims and will be described in detail below with reference to examples.

  According to this invention, it is possible to easily verify the contents of the initial attribute assignment performed while browsing a document while browsing a similar document.

  Examples of the present invention will be described below.

  FIG. 1 schematically shows a document attribute input apparatus according to the first embodiment of the present invention. In this embodiment, a plurality of paper documents are digitized to obtain image data of each page of each document, and further, character recognition result data of each page (in this example, HTML and code that are marked up and highlighted with a specific expression) Data or text data) is acquired and an attribute is given to the document while referring to highlight information or the like. The initial document attribute assignment is performed by a plurality of workers. The process of assigning document attributes performed by individual workers is sometimes called a document attribute assignment session. A plurality of document attribute assigning sessions are performed for each document, and the contents of each document attribute assigning session can be verified by a verifier. The verification process performed by the verifier is sometimes called a verification session. In this example, for convenience of explanation, it is assumed that images of three documents are read. However, the number of documents is not limited to this.

  In the following, the configuration of the document attribute input device will be described first, then the operation of the document attribute assignment session by the worker will be described, and finally the operation of the verification session will be described. In the first embodiment, a plurality of workers log in to the server device (second embodiment described later), each worker inputs an attribute to each document, and then the verifier logs in to the server device. Since it is assumed to be verified, each processing is called a “session”, but it is clear that this generally refers to “processing” in general.

  In FIG. 1, the document attribute input device according to the first embodiment includes a paper feeding device 20, an image reading device 30, a character recognition device 31, a paper discrimination unit 40, a document management unit 50, a storage device 60, a display control unit 70, and a display device. 80, a pointing device 90, and the like.

  The paper feeding device 20 and the image reading device 30 digitize a paper document. First, documents D1, D2, and D3 are separated into sheet bundles d1, d2, and d3, and management sheets (also referred to as banner sheets) c1, c2, and c3 are placed on the respective sheet bundles d1, d2, and d3. Inserted. These groups of sheets are placed on the paper feeding device 20 as an object to be batch-processed (hereinafter also referred to as a batch sheet bundle, indicated by reference numeral 10). The paper feeding device 20 sequentially feeds the sheets of the batch paper bundle 10 from the top or bottom and supplies them to the image reading device 30. The image reading device 30 is typically a normal scanner, and optically scans the printing surface of the paper that is sent to acquire electronic image data on the printing surface.

  The character recognition device 31 executes a well-known character recognition process on the electronic image data.

  The paper determination unit 40 detects the management papers c1, c2, and c3 from the batch paper bundle 10. For the management papers c1, c2, and c3, for example, special color papers (some specific areas may be special colors) are used to distinguish them from other papers. Alternatively, a distinguishing character string such as “BANNER” followed by a continuous number (for example, a number gradually increasing from “0001”) may be printed, or a one-dimensional or two-dimensional barcode or RFID tag is provided. It is good also as a paper for document separation. The color of the paper can be determined from the attribute of the electronic image data read by the image reading device 30. The paper feeding device 20 or the like may be provided with a color discrimination mechanism (for example, a detector that detects light of a predetermined color, not shown). When using the description contents of the management sheets c1, c2, and c3, the optical image is converted into a character code by the character recognition device 31, and the description contents, for example, “BANNER” characters and serial numbers are converted into continuity. Detect based on. When using a bar code or RFID tag, a dedicated reader is required. The RFID tag reader causes malfunction when papers are overlapped and a plurality of RFID tags are arranged in the vicinity of each other, so that a state where papers do not overlap such as a conveyance path of the paper feeding device 20 is guaranteed. It is preferable to provide in the place. Moreover, it is preferable to provide an electromagnetic shielding part to enhance resistance to external noise.

  The document management unit 50 uses a management unit 50a as shown in FIG. 5 to display an image of a sheet between the management sheet and the next management sheet (in the case of the bottom sheet bundle, the next sheet of the management sheet). (From the image to the image on the last sheet). Furthermore, in the example of FIG. 5, the image data of each page constituting the document and the character recognition result data (HTML document) obtained by character recognition of the image data are managed under one management unit 50a. ing. More specifically, an identifier (ID) having a continuous number in page order is assigned to the image data of each page for each document. Further, consecutive numbers are assigned to the character recognition result data of each page for each document in page order. The identifier of the corresponding image data and the identifier of the character recognition result data are the same in the main part, and the part indicating the data type is different. For example, the image data is represented by “0001-001 img”, and the character recognition result is represented by “0001-001 html”. The data type may be displayed as an extension. The data identifier may be a path name or another identifier independent of the path name. In a more specific example, image data of one document is arranged in one folder, and consecutive file names are assigned to each image data as one file. Similarly, for the character recognition result, one folder is prepared for each document, and consecutive file names are assigned with the character recognition result data of each page as one file. The mutually corresponding image data file and character recognition result data file have corresponding file names, are associated with each other, and can easily determine one from the other. For example, the folder name may be changed as appropriate, and the extension may be changed as appropriate. The broken lines in FIG. 5 indicate that image data and text data are managed by different folders 50b and 50c. In the example of FIG. 5, image data of a plurality of documents is arranged in one folder 50b, and character recognition result data of a plurality of corresponding documents is arranged in another folder 50c. Of course, the document to which the image data and the character recognition result data belong can be determined by using, for example, a character string indicating which document is included in a part of a data identifier (or a file name) without using the individual folders 50b and 50c. It may be included.

  The above-described image data and character recognition result data are stored in the storage device 60 (the image document storage unit 602 and the HTML document storage unit 601 in FIG. 2).

  The document management unit 50 further manages data management information. FIG. 6 shows an example of data management information. As shown in this figure, for each page ID of image data or character recognition result data, the previous page ID (the previous page ID), the next page ID (immediately after the page ID). Page ID), document separation flag, top page ID of the next document, top page ID of the previous document, and the like. If there is no corresponding page, “null” is assigned. When the document separation flag is “TRUE”, it indicates that the next page is a page of another document, that is, document separation. When the document separation flag is “FALSE”, the next page is also the same document. This indicates that the page is a document, that is, not a document break. Individual documents can be separated and managed using a document separation flag.

  For example, when this page ID is “0001-001” (in this example, indicates page 001 of document 0001), the next page ID is “0001-002” and the previous page ID is “null” (document 001 is First document). The document separation flag is “FALSE”, and the top page ID of the next document is “0002-001”. The top page ID of the previous document is “null”.

  Similarly, when the page ID is “0001-005”, the previous page ID is “0001-004”, the next page ID is “0002-001”, the document separation flag is “TRUE”, and the next page ID is “0001-005”. The top page ID of the document is “0003-001”, and the top page ID of the previous document is “0001-001”.

  By referring to the record (tuple) of each page, the page ID of the previous page, the next page, the top page of the previous document, and the top page of the next document is acquired, and each page is extracted from the storage device 60 and displayed. Can do.

  In this example, the image data folder and page ID are designated to retrieve image data, and the character recognition result data folder and page ID are designated to retrieve character recognition result data. That is, image data and character recognition result data are distinguished by designating a folder. Of course, the image data and the character recognition result data may be distinguished based on the extension, or an additional character string for distinction may be added to the page ID to distinguish the image data or the character recognition result data.

  The data management information in FIG. 6 can adopt various formats. For example, the data management information may be managed as a database record, may be managed as a CSV file, or may be managed in any other format.

  The data management information is stored and held in the storage device 60 (data management information storage unit 605 in FIG. 2).

  Further, the document management unit 50 presents a user interface for initial attribute input as shown in FIG. 3 to the user who wants to give the original document attribute to the document, and inputs the attribute from the user. Accept. That is, the document management unit 50 sends data for displaying a user interface for initial attribute input as shown in FIG. 3 to the display control unit 70. The display control unit 70 displays the display device 80 such as a CRT or a liquid crystal display device. A user interface as shown in FIG. 3 is displayed on the screen. The pointing device 90 is a mouse device, for example, and instructs the coordinates on the display screen of the display device 80.

  In the display example shown in FIG. 3, character recognition result data (HTML document) 151 is displayed in the left frame F1. An original image document (image data) obtained by scanning in a paper document is subjected to character recognition processing, and then a specific expression is extracted and the specific expression is highlighted by a tag. In this example, specific expressions such as “Family allowance application” (document name), “Ai Ueo” (person name), “2004/2/9” (date), “01111” (employee number) are highlighted. It shows.

  Original image data can be displayed in the frame F1 instead of the character recognition result data 151. For example, in the example of FIG. 3, various [Image Document] (original image data), [HTML Document] (character recognition result data), [Previous Page], [Next Page], [Previous Document] are displayed on the command bar C at the top. ] And [Next Document] command buttons are provided, and the document can be switched by clicking the [Image Document] or [HTML Document].

  Also, by clicking [Previous page], [Next page], [Previous document], [Next document], the previous page data (image data or character recognition result data), the next page data, the top page of the previous document The data and the top page data of the next document can be displayed in the frame F1.

  In the example of FIG. 3, the initial attribute input table 152 and the fixed phrase table 153 are displayed in the frames F2 and F3. The document management unit 50 clicks on the highlighted portion of the character recognition result data (HTML document) and clicks on the field of the initial attribute input table 152, and then the word / phrase of the highlighted portion is stored in the initial attribute input table 152. Post to the field. Similarly, the document management unit 50 selects a fixed phrase from the field of the initial attribute input table 152 based on the click of a predetermined fixed phrase of the fixed phrase table 153 and the click of the field of the initial attribute input table 152. Post to.

  In this way, the attribute value can be easily input by clicking the field of the initial attribute input table 152 after clicking the highlighted part or the fixed phrase field of the character recognition result data (HTML document).

  The attribute value may be directly input using a keyboard instead of transferring the attribute value by the operation as described above.

  Since there are misrecognitions in character recognition results, when transferring from the character recognition result data to the initial attribute input table, it may be necessary to view the original image data and verify that it is correct. is there. In some cases, it is necessary to browse the previous and next pages in order to determine the attributes of the document. In addition, there are cases where it is necessary to browse a document to which an attribute has been assigned or a document to which an attribute is to be assigned and determine the attribute from the overall viewpoint.

  In this embodiment, command buttons of [Image Document] (original image data), [HTML Document] (character recognition result data), [Previous Page], [Next Page], [Previous Document], and [Next Document] are displayed. By clicking, it is possible to immediately move to any page of any type of data and display it.

  Attributes input using the initial attribute input table 152 are stored in the attribute data storage unit 603. As shown in FIG. 8, the attribute data includes a document ID, a worker ID, and an attribute name-attribute value. The number of attribute name-attribute value pairs is predetermined according to the type of document.

  Further, the document management unit 50 presents a user interface for attribute input at the time of verification as shown in FIG. 4 to the user who wants to verify the originally assigned document attribute and inputs the attribute from the user. Accept. That is, the document management unit 50 sends data for displaying a verification-use attribute input user interface as shown in FIG. 4 to the display control unit 70. The display control unit 70 is a display device such as a CRT or a liquid crystal display device. A user interface as shown in FIG. 4 is displayed on the 80 screen. The pointing device 90 is a mouse device, for example, and instructs the coordinates on the display screen of the display device 80.

  The display example shown in FIG. 4 is basically the same as the display example of FIG. 3 except for the display of the right frame F2, and functions in the same manner.

  In the example shown in FIG. 4, the verification time attribute input table 154 is displayed in the right frame F2. The verification attribute input table 154 includes an input field 154a in which the verifier inputs an attribute value of the attribute, and display columns 154b and 154c of attribute values input in the initial attribute assignment. In this example, worker A and worker B log in to the system and assign attributes in the initial attribute assignment session. Worker A assigns “family allowance”, “01111”, and “2004/2/9” to the attribute names “application name”, “employee number”, and “application date”, respectively. “Allowance”, “01111”, and “2004/2/8” are assigned. The document management unit 50 includes, for example, an attribute value comparison unit 500 (indicated by a broken line in FIG. 1), compares attribute values in different sessions that were originally input, and if they match, the verifier The attribute value is directly transferred to the input field. In this display example, the employee number “01111” is directly transferred to the input field. On the other hand, when the attribute values in the different sessions initially input are not all the same value, the display of the corresponding attribute name is highlighted. In this display example, “application name” and “application date” are highlighted. The attribute name highlighting process and the transfer of the attribute value originally given are compared with the attribute data in FIG. 8 within the same range of document IDs, and an HTML tag is attached based on the comparison result, or a form tag is used. This can be done by setting default values for the described input form.

  As described with reference to FIG. 3, the verifier can input the highlighted word in the document or the phrase in the fixed phrase table by clicking on the input field of the verifier. Moreover, an attribute value can be directly input using a keyboard. Even when the same attribute value is input between workers in the original attribute input and the attribute value is also posted in the input field of the verifier, the verifier corrects this and inputs another attribute value. You may do it.

  When the verifier determines and inputs the correct attribute value, the attribute value is confirmed by clicking the “Confirm” button. This data is stored in the attribute data storage unit 603 as verified attribute data. The attribute data to which the verification value is input is basically the same as that shown in FIG. However, it is appropriately displayed that the attribute data is final attribute data by the verifier.

  Original image data can be displayed in the frame F1 instead of the character recognition result data 151. For example, in the example of FIG. 4, various [Image Document] (original image data), [HTML Document] (character recognition result data), [Previous Page], [Next Page], [Previous Document] are displayed on the command bar C at the top. ] And [Next Document] command buttons are provided, and the document can be switched by clicking the [Image Document] or [HTML Document].

  Also, by clicking [Previous page], [Next page], [Previous document], [Next document], the previous page data (image data or character recognition result data), the next page data, the top page of the previous document The data and the top page data of the next document can be displayed in the frame F1.

  The data management information shown in FIG. 6 may be used as it is. However, as shown in FIG. 7, data management information for only documents having attribute values between workers is prepared. Also good. In the example of FIG. 7, the document “0002” has no attribute value and is omitted.

  In this way, it is possible to easily verify the contents of the attribute assignment while browsing the document (image data and character recognition result data) when the attribute was initially assigned.

  In this example, since the verifier input field 154a is on the display area side of the character recognition result data 151 and image data, the highlighted phrase in the document can be transcribed with the shortest path. Further, even if the number of workers (the number of initial attribute assignment sessions) increases, an additional column added to the initial attribute value display fields 154b and 154c only extends to the right in the figure, and the verifier input field 154a Since the position itself does not change, stable input can be performed regardless of the number of workers.

  FIG. 2 shows a configuration example of a main part of the document management unit 50 of the document attribute input apparatus according to the first embodiment. In this figure, the document management unit 50 includes an HTML document display unit 501, an image document display unit 502, an attribute. An input table display unit 503, a fixed phrase table display unit 504, an event processing unit 505, and a buffer memory 506 are configured. The document management unit 50 transfers the display data to the display control unit 70 and displays it on the display device 80. An event based on an operation of the pointing device 90 or the like is processed by the event processing unit 505. In this example, the buffer memory 506 is used to transfer the attribute value to the input field. The storage device 60 includes an HTML document storage unit 601, an image document storage unit 602, an attribute data storage unit 603, a fixed phrase storage unit 604, a data management information storage unit 605, and the like.

  The display device 80 includes a normal display device such as a CRT or a liquid crystal display device and a display control unit, and displays character codes and images. The pointing device 90 is a mouse device, for example, and instructs the coordinates on the display screen of the display device 80.

  The HTML document display unit 501 displays the HTML document stored in the HTML document storage unit 601. This HTML document is a document into which attributes are to be input. In this example, the HTML document is obtained by character recognition processing of an original image document (image data) obtained by scanning in a paper document, extracting a unique expression, and highlighting the specific expression with a tag. is there.

  The image document display unit 502 displays the original image document stored in the image document storage unit 602. The HTML document and the image document are switched and displayed. For example, in the example of FIG. 2, various commands are provided in the command bar C at the top, and the document can be switched by clicking the “image document” or “HTML document”.

  The attribute input table display unit 503 and the fixed phrase table display unit 504 display the initial attribute input table 152 and the fixed phrase table 153 in the frames F2 and F3, respectively, in the example of FIG. The attribute data storage unit 603 stores input attribute data, and the fixed phrase storage unit 604 stores fixed phrases.

  As described above, the data management information storage unit 605 stores data management information (FIGS. 6 and 7) for immediately displaying image data and character recognition result data.

  In FIG. 1, the HTML document display unit 501, the image document display unit 502, the attribute input table display unit 503, and the fixed phrase table display unit 504 are separately shown. However, one HTML document is displayed by an HTML frame method or the like. Can be configured as a single display unit. Plug-in software may be used to display the image document.

  The event processing unit 505 processes various events. In association with this example, the event processing unit 505 detects a click operation on the highlight display unit of the HTML document and writes the text of the highlight unit in the buffer memory 506. The buffer memory 506 is typically a clipboard. In addition, when the field of the attribute input table is clicked, the event processing unit 505 reads the text stored in the buffer memory 506 and performs processing for inputting the text into the field.

  Further, the event processing unit 505 writes the text in the buffer memory 506 when a specific field of the fixed phrase is clicked. Thereafter, when a field in the attribute input table is clicked, the text stored in the buffer memory 506 is read and the text is input to the field.

  In this way, an attribute value can be easily input by clicking a highlighted part or a fixed phrase field in an HTML document and then clicking a field in the attribute input table.

  Next, the operation of assigning attributes at the beginning will be described.

  FIG. 9 shows an example of an operation for assigning initial attributes. In this example, a unit for collectively processing a plurality of documents is called a task. The operator specifies a task, specifies multiple documents at once, and operates each command button of [Previous Document], [Next Document], [Previous Page], and [Next Page] to select individual documents, A page is designated, a desired page of a desired document is displayed in the document display area, and attributes are assigned, and the task is selected from a list as shown in FIG.

  In FIG. 9, the worker logs in to the system and selects a task (S01, S02). Documents included in the selected task are sequentially browsed and attributes are assigned (S03). As shown in FIG. 11 or FIG. 12, the attribute may be assigned by transferring a highlighted phrase or fixed phrase in the HTML document to the input field, or directly entering the input field using a keyboard or the like. You may do it. After assigning attributes to all documents in the task, if another task is to be processed, the process returns to step S02 to repeat the process. If no other task is processed, the user logs out (S04).

  FIG. 11 shows an outline of an operation example when the highlighted phrase in the HTML document is transferred to the corresponding field in the attribute input table in step S03 of FIG. HTML document (or image document), attribute input table, and fixed phrase table are displayed (S11). When inputting an attribute by clicking on the HTML document, the HTML document is displayed. When the image document is displayed, the HTML document is displayed by clicking the “HTML” button on the command bar. Next, if there is an appropriate attribute among the highlighted words / phrases, the highlighted portion is clicked using the pointing device 90. The event processing unit 505 discriminates this and takes out the text of the highlighted part (if linked, the linked text) and writes it in the buffer memory 506 (S12). After that, when the corresponding field in the attribute input table is clicked, the event processing unit 505 discriminates this, writes the text in the buffer memory 506 into the field, and then clears the contents of the buffer memory 506 (S13). .

  FIG. 12 shows an outline of an operation example when the fixed phrase of the fixed phrase table is transferred to the corresponding field of the attribute input table in step S03 of FIG. 9, and in this figure, the portion corresponding to FIG. Are marked with corresponding symbols. In FIG. 12, an HTML document (or image document), an attribute input table, and a fixed phrase table are displayed by designating an attribute input target document (S11). If there is an appropriate attribute in the fixed phrase table, the fixed phrase is clicked using the pointing device 90. The event processing unit 505 discriminates this and takes out the text of the fixed phrase (if linked, the linked text) and writes it in the buffer memory 506 (S14). After that, when the corresponding field in the attribute input table is clicked, the event processing unit 505 discriminates this and writes the text in the buffer memory 506 to the field, and then clears the contents of the buffer memory 506 (S13). .

  When the attribute input is complete, press the “Confirm” button to confirm.

  Next, an operation for verifying the attribute value assigned initially will be described. FIG. 13 shows an example of the operation at this time. In this figure, the portions corresponding to those in FIG.

  In FIG. 13, the worker logs in to the system and selects a task (S01, S02). Documents included in the selected task are sequentially selected, and the attribute value initially assigned is verified using the attribute input screen for verification in FIG. 4 (S06). When necessary, as shown in FIG. 11 or FIG. 12, the highlighted phrase or fixed phrase in the HTML document is transferred to the input field to enter the attribute value, or directly input using a keyboard or the like. Fill in the field. If attribute assignment of all documents in the task is verified and other tasks are to be processed, the process returns to step S02 to repeat the process. If no other task is processed, the user logs out (S04).

  This is the end of the description of the first embodiment.

  According to this embodiment, it is possible to easily verify the document attribute value initially assigned.

  In this embodiment, as shown in FIG. 14, a specific expression extraction unit 507 and an HTML document generation unit 508 are provided to generate an HTML document from an original image document and store it in the HTML document storage unit 601. Also good. The character recognition device 31 executes a normal character recognition method that analyzes an image and outputs a character code. The specific expression extraction unit 507 extracts important expressions such as name, name, place name, organization name, date, currency, etc., and usually includes a lexical analysis unit and an extraction rule execution unit. It is not limited. The specific expression extraction unit 507 can also specify the type of the specific expression, thereby selecting the highlight attribute individually. A desired HTML document can be generated by marking up with a tag based on the extracted unique expression and its type. Of course, an HTML document prepared in advance may be used.

  Furthermore, a rule table 509 describing a value that cannot be used as an attribute value may be provided, and the event processing unit 505 may refer to this to prohibit transcription. For example, when the founding time is relatively new, for example, when a company document in the 1970s is digitized, and “56/01/01” and the year is “56”, the year “1956” However, since “Showa 56” is a failure, the input as “56” and the input of “1956” using the complete function are prohibited. In addition, by providing such an attribute value determination rule that is impossible, a character recognition error can be dealt with.

  Next, a second embodiment of the present invention will be described. In the second embodiment, a document attribute input system is constructed using a server device and a client device. Of course, the server application and the client application may be executed on the same computer.

  FIG. 2 shows the entire document attribute input system according to the second embodiment. In this figure, the document attribute input system includes a server device 100 and a client device 200. The server device 100 is typically a web server, and appropriately includes an application server and a database system. The server apparatus 100 receives an HTTP request (POST, GET request, etc.) from the client apparatus 200 and transfers a predetermined document to the client apparatus 200. Alternatively, an argument included in HTTP is passed to a predetermined program, and the return value is transferred to the client device 200. In this example, the program execution unit 110 includes an HTML document output program 111 (corresponding to the HTML document display unit 501 in FIG. 1), an image document output program 112 (image document display unit 502 in FIG. 1), an attribute input table output program ( 1), the fixed phrase table output program 114 (corresponding to the fixed phrase table display unit 504) is executed.

  The storage device 120 stores HTML documents, image documents, and the like.

  The client device 200 executes browsing software 201 such as a web browser or plug-in software. In this example, the browsing software 201 receives and displays an HTML document for attribute input (all displays including the attribute input table and fixed phrase table in FIG. 2). The HTML document includes a script written in a predetermined programming language such as JavaScript (trademark). This script is executed on the browsing software 201. An event processing program 203 composed of this script is executed by the script execution unit 202. This event processing program 203 corresponds to the event processing unit 505 in FIG.

  Also in this embodiment, the initial attribute can be input using the initial attribute input screen shown in FIG. 3, and the input contents of the initial attribute value can be verified using the verification attribute input screen shown in FIG.

  FIG. 16 explains the outline of the phrase transcription operation in this embodiment. In FIG. 16, the browsing software 201 of the client device 200 displays, for example, a list of processing target tasks, and selects and designates a processing target from this list (S21). The server apparatus 100 appropriately extracts an HTML document from the task to be processed and targets and transmits the frame F1 in FIG. 3 or FIG. 4 (S22). Note that the attribute input table and the fixed phrase table may be appropriately transferred. In the client apparatus 200, the highlighted part of the HTML document is clicked and the text of the important phrase is written on the clipboard (S23). At this time, the highlight is tagged as shown in FIG. 17A, for example, and the script used at this time is as shown in FIG.

  Next, when a field in the attribute input table is clicked, the field value is passed to the attribute table output program 113 of the server apparatus 100 as an argument so as to update the field value with the clipboard text (S24). The script executed by the click operation at this time is as shown in FIG. This example also clears the clipboard contents. The attribute table output program 113 of the server device 100 generates an attribute input table in which the text is written in the field, and transfers the attribute input table to the client device 200.

  Also in this embodiment, the contents of the initial attribute assignment can be easily verified.

  The present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the invention. For example, in the above example, the image document and the HTML document are alternatively displayed as a pair, but both may be displayed simultaneously. Further, the image document may not be handled. Also, other markup languages such as XML and XHTML may be used instead of the HTML language.

It is a figure which shows typically the document attribute input device of Example 1 of this invention. 3 is a block diagram illustrating an exemplary configuration of a main part of a document management unit 50 according to Embodiment 1. FIG. It is a figure explaining the example of the attribute input screen at the time of the initial attribute provision of Example 1. FIG. FIG. 10 is a diagram illustrating an example of an attribute input screen at the time of verification according to the first embodiment. It is a figure explaining the management aspect of the image data of Example 1, and character recognition result data. It is a figure explaining the data management information at the time of initial attribute provision of Example 1. FIG. It is a figure explaining the data management information at the time of the verification of Example 1. FIG. It is a figure explaining the example of the attribute data of Example 1. FIG. 6 is a flowchart illustrating an operation example of initial attribute assignment for each task according to the first embodiment. It is a figure explaining the task selection screen of Example 1. FIG. 10 is a flowchart for explaining an example (part 1) of attribute assignment operation according to the first embodiment. 10 is a flowchart for explaining an attribute assignment operation example (No. 2) according to the first embodiment; 6 is a flowchart illustrating an example of a verification operation for each task according to the first exemplary embodiment. FIG. 10 is a block diagram illustrating another configuration example of the document management unit 50 according to the first embodiment. It is a figure explaining the structure of Example 2 of this invention. FIG. 10 is a diagram for explaining the operation of the second embodiment. 10 is a diagram illustrating an example of an HTML document according to Embodiment 2. FIG. FIG. 10 is a diagram for explaining copying from a transcription source in Example 2. FIG. 10 is a diagram illustrating copying to a transfer destination in the second embodiment.

Explanation of symbols

10 batch paper bundle 20 paper feeding device 30 image reading device 31 character recognition device 40 paper discrimination unit 50a management unit 50b folder 50c folder 50 document management unit 60 storage device 70 display control unit 80 display device 90 pointing device 100 server device 110 program execution Unit 111 HTML document output program 112 image document output program 113 attribute table output program 114 fixed phrase table output program 120 storage device 151 character recognition result data 152 initial attribute input table 153 fixed phrase table 154 verification attribute input table 154a input field 154b, 154c Attribute value display column 200 Client device 201 Browsing software 202 Script execution unit 203 Event processing program 500 Attribute value comparison unit 501 TML document display unit 502 Image document display unit 503 Attribute input table display unit 504 Fixed phrase table display unit 505 Event processing unit 506 Buffer memory 507 Specific expression extraction unit 508 Document generation unit 509 Rule table 601 HTML document storage unit 602 Image document storage unit 603 attribute data storage unit 604 fixed phrase storage unit 605 data management information storage unit

Claims (12)

Grant attribute storage means for storing attribute values given to the document for each attribute grant process;
A document designating means for designating a document to be verified;
An attribute input form display unit at the time of verification for displaying in the attribute input form display area the attribute value given by the attribute assigning process for the document specified by the document specifying unit and the attribute value input field for the verifier to input the attribute value When,
Document display means for displaying a document designated by the document designation means in a document display area;
A document attribute input device, comprising: attribute value input means for inputting an attribute value into the attribute value input field displayed in the attribute input form display area based on a user operation.   The document attribute input device according to claim 1, further comprising a comparison unit that compares attribute values assigned by different attribute assigning processes.   3. The document attribute input device according to claim 2, wherein when the attribute values assigned by different attribute assignment processes match, the matching attribute value is input to the attribute value input field.   4. The document attribute input device according to claim 2, wherein the attribute value input field is highlighted when the attribute values assigned by different attribute assigning processes do not match.   5. The attribute value input field is accompanied by an attribute name of an attribute value to be input to the attribute value input field, and the attribute name is displayed differently from other parts to highlight the attribute value input field. Document attribute input device.   The document attribute input device according to claim 1, wherein the document is a document including image data obtained by optically reading a page of a paper document.   The document includes a first document composed of image data obtained by optically reading a page of a paper document, and a second document composed of character recognition result data obtained by character recognition of the image data. The document attribute input device according to claim 1, wherein one document and the second document are switched and displayed in the document display area.   In the second document composed of the character recognition result data, a predetermined word is highlighted in the document display area, and a click operation on the highlighted word and a click operation on the attribute value input field in the attribute input form display area are performed. 7. The document attribute input device according to claim 6, wherein the word / phrase is transferred to the attribute value input field based on the word / phrase.   The attribute input form display area and the document display area are arranged side by side, and the attribute value input field in the attribute input form display area is closer to the document display area than the attribute value assigned by the attribute assignment process. The document attribute input device according to claim 1, which is disposed in   The attribute value input means determines whether or not the attribute value is acceptable as an attribute value when the attribute value is input to the attribute value input field displayed in the attribute input form display area based on a user operation. The document attribute input device according to claim 1. Storing the attribute value assigned to the document for each attribute assignment process by the assignment attribute storage means;
Designating a document to be verified by a document designating means;
The attribute value assigned by the attribute assigning process for the document specified by the document specifying means and the attribute value input field for inputting the attribute value by the verifier are displayed in the attribute input form display area by the attribute input form display means at the time of verification. And steps to
Displaying the document designated by the document designation means in the document display area by the document display means;
And a step of inputting an attribute value based on a user operation into the attribute value input field displayed in the attribute input form display area by an attribute value input means. Storing the attribute value assigned to the document for each attribute assignment process by the assignment attribute storage means;
Designating a document to be verified by a document designating means;
The attribute value assigned by the attribute assigning process for the document specified by the document specifying means and the attribute value input field for inputting the attribute value by the verifier are displayed in the attribute input form display area by the attribute input form display means at the time of verification. And steps to
Displaying the document designated by the document designation means in the document display area by the document display means;
A document characterized by being used by an attribute value input means to cause a computer to execute an attribute value input step based on a user operation in the attribute value input field displayed in the attribute input form display area. Computer program for attribute input.

Images