JP2015176410A - Information processor, document reading method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology for efficiently extracting description information from a document including items and entry subjects corresponding to the items.SOLUTION: An information processor includes: a storage part for storing item information related to item candidates as the candidates of items included in a document; a reading part for reading the document; a description information extraction part for extracting description information from the document read by the reading part; an item extraction part for extracting the items corresponding to the item candidates from the description information; an entry subject specification part for specifying the entry subjects corresponding to the items from the description information on the basis of the reading sequence of the description information by the reading part; and a customer information generation part for generating customer information in which the items are associated with the entry subjects corresponding to the items.

Description

  The present invention relates to an information processing apparatus, a form reading method, and a program.

  2. Description of the Related Art In recent years, a technique for generating digital data in which input information is converted into a character code by reading a form in which handwritten characters are entered with respect to an item represented by type by a scanner and performing OCR (Optical Character Recognition) processing. Is popular. For example, a credit collection company (servicer) that receives forms from a plurality of companies needs to read different types of forms and generate respective digital data in order to perform processing related to the received forms.

  Japanese Patent Laid-Open No. 2004-133830 describes that in a form processing system that extracts a character line that is included in a form image and exists in a frame that is an area surrounded by a line segment and recognizes the extracted character line, the form processing system Stores frame structure dictionary data in which character strings corresponding to a plurality of extraction target frames from which the character line is to be extracted and the arrangement relationship of the plurality of extraction target frames are associated with each other, from the form image A frame extracting unit that extracts a frame as an extraction frame; a frame matching unit that refers to the frame structure dictionary data and calculates a matching result in which the extracted frame extracted by the frame extracting unit is associated with the extraction target frame; A character line extraction unit that extracts a character line existing in the extraction frame associated with the extraction target frame with respect to the collation result calculated by the frame collation unit; and each collation calculated by the frame collation unit For the results, The extraction target included in each matching result is calculated by calculating a frame score indicating the similarity of the character line to the character string of the extraction target frame corresponding to the extraction frame from which the character line extracted by the character line extraction unit is extracted. A score calculation unit that calculates a matching result score that is a score of each matching result based on the frame score of the frame associated with the frame, and the frame matching based on the matching result score calculated by the score calculating unit A form processing system is disclosed in which one collation result among the collation results calculated by the section is used as an extraction collation result.

JP 2013-105344 A

  In the prior art, when generating digital data from a form, it is necessary to register the form format in advance. For this reason, in a servicer that handles forms of a plurality of different formats, it is necessary to register a format for each form in advance, which is complicated. In addition, it is necessary to sort the forms for each format and then perform reading processing with a scanner, which is not convenient.

  In the technique disclosed in Patent Document 1, frame structure dictionary data in which the position of the reading frame is associated with the character string of the item must be stored in advance as a format.

  The present invention has been made in view of the above points, and it is an object of the present invention to provide a technique capable of efficiently extracting description information from a form including items and entries for the items.

  The present application includes a plurality of means for solving the above-described problems, and examples thereof are as follows.

  In order to solve the above problems, an information processing apparatus according to the present invention is read by a storage unit that stores item information related to item candidates that are candidates for items included in a form, a reading unit that reads a form, and the reading unit. A description information extraction unit for extracting description information from the form, an item extraction unit for extracting the item corresponding to the item candidate from the description information, and a reading order of the description information by the reading unit, An entry specifying unit that specifies an entry for an item from the description information, and a customer information generation unit that generates customer information in which the item is associated with the entry for the item. It is characterized by.

  In order to solve the above-mentioned problem, the item information associates the item candidate with the attribute of the entry for the item, and the entry specifying unit is configured by the reading order and the item extraction unit. The entry for the item may be specified based on the extracted item and the attribute associated with the item information.

  In order to solve the above-described problem, the information processing apparatus according to the present invention includes a candidate form specifying unit that determines whether a form candidate corresponding to the form read by the reading unit is stored in the storage unit. The storage unit stores, for each item, form information that associates the description order of the items with the description order of the items to be entered in the storage unit as the form candidate, The description information extraction unit refers to the form information using the item extracted by the item extraction unit and the reading order of the item, and the reading order and the description order of the items. If it is determined that the form information corresponding to is stored in the storage unit, the entry information read in the reading order corresponding to the entry order of the entries in the form information is the entry. Was identified as the customer information generation unit may be characterized by associating with said items and the entry item by using the form information.

  In order to solve the above problem, the information processing apparatus according to the present invention relates to a form for which the candidate form specifying unit does not determine that the form information corresponding to the reading order and the description order is included in the form information. A temporary form generation unit that generates a temporary form using the items extracted by the item extraction unit and the description order of the items, and the storage unit stores the temporary form as the form candidate The candidate form specifying unit determines whether or not the temporary form corresponding to the reading order and the description order of the form is stored in the storage unit, and the entry specifying unit When it is determined that the temporary form is stored by the form specifying unit, the entry items may be specified using the description order included in the temporary form.

  In order to solve the above problem, the information processing apparatus according to the present invention according to the present invention includes a case in which the entry for the item is not specified by the entry specifying unit in the written information, An error determination unit that performs an error determination when there is the description information that is not specified as an entry, and a display unit that displays a form correction screen when the error determination is performed by the error determination unit; The correction screen may receive the designation of the item or the entry for the description information.

  In order to solve the above-described problem, a form reading method according to the present invention is a form reading method using an information processing apparatus including a control unit, and the control unit is an item candidate for an item included in the form. Included in the storage unit that stores item information related to candidates, a reading step of reading a form, a description information extracting step of extracting description information from the form read in the reading step, and the form stored in the storage unit An item extraction step for extracting the item corresponding to the item candidate that is an item candidate from the description information, and an entry for the item is specified from the description information based on the reading order of the description information in the reading step. A customer information generation step for generating customer information in which an entry specifying step, the item, and the entry for the item are associated with each other. Characterized in that it comprises a and.

  In order to solve the above problem, a program according to the present invention is a program that causes a computer to execute a step of performing a form reading process, causing the computer to function as a control unit, and A storage unit that stores item information regarding item candidates that are candidates for items included in the form, a reading step of reading the form, a description information extracting step of extracting description information from the form read in the reading step, and the storage An item extraction step for extracting the item corresponding to an item candidate that is a candidate for an item included in the form stored in a section from the description information, and the item based on the reading order of the description information in the reading step. An entry specifying step for specifying an entry for the item from the written information, the item, and the item Characterized in that to execute, and customer information generating step of generating a customer information associated with said entry item for.

  ADVANTAGE OF THE INVENTION According to this invention, the technique which can extract description information efficiently from the form containing the item and the entry item with respect to this item can be provided.

  Problems, configurations, and effects other than those described above will be clarified by the following description of embodiments.

3 is a functional block diagram of the form reading device 10. FIG. It is a figure which shows an example of item DB. It is a figure which shows an example of form DB. It is a figure which shows an example of customer master DB. It is a figure which shows the hardware structural example of information processing apparatus. It is a flowchart (the 1) which shows the flow of a form reading process. It is a flowchart (the 2) which shows the flow of a form reading process. It is an example of a form. It is a figure which shows an example of the description information extracted by the description information extraction part. It is a figure which shows an example of the description information in which the item identifier and the attribute were linked | related. It is a figure which shows an example of the choice as description information. It is a figure which shows an example of a form correction screen.

  Hereinafter, examples of embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram of the information processing apparatus 10.

  The information processing apparatus 10 is a terminal device such as a PC (Personal Computer), a PDA (Personal Digital Assistant), or a smartphone. Further, the information processing apparatus 10 may be a multi-function machine or the like having a general scanner function.

  The information processing apparatus 10 includes a control unit 110 and a storage unit 120. The control unit 110 extracts the description information from each form read by the scanner connected to the input I / F included in the information processing apparatus 10 and controls processing for generating a customer master. The storage unit 120 stores data necessary for processing performed by the control unit 110.

  The control unit 110 includes a reading unit 111, a description information extracting unit 112, an item extracting unit 113, an entry specifying unit 114, a customer information generating unit 115, a candidate form specifying unit 116, and a temporary form generating unit 117. The error determination unit 118 and the display unit 119 are provided.

  The reading unit 111 acquires image data (mainly raster data) of a form read by the scanner. The form includes an item indicating the content requested to be entered by the user or an entry entered by the user within a frame represented by an orthogonal straight line. In the following description, it is assumed that the general term of items and entries is written information.

  Further, the reading unit 111 stores the read image data in an area (not shown) of the storage unit 120.

  The description information extraction unit 112 performs OCR processing on the acquired image data, and acquires character data of description information included in the form. Since well-known means are used for the OCR process, a detailed description thereof is omitted.

  Further, the description information extraction unit 112 identifies a frame described in the form. A known method is used to specify the frame. For example, the method described in JP2013-105344 is used. The description information extraction unit 112 acquires character data of description information described in a frame for each frame included in the form. The written information extraction unit 112 assigns a number corresponding to the reading order by the reading unit 111 to the acquired character data.

  The item extraction unit 113 extracts items from the description information by using the item DB 121 in which item candidates are recorded.

  The entry specifying unit 114 specifies the entry entered by the user using the number assigned to the description information. The specified entry may be handwritten or printed type. The method for identifying the entry will be described in detail later.

  The customer information generation unit 115 generates a customer master DB as customer information in which items extracted from the form are associated with the entry items.

  The candidate form specifying unit 116 uses the form DB 122 that is form information stored in the storage unit 120 in advance to determine whether the form read by the reading unit 111 is a form having a format stored in the form DB 122. To do. In addition, the candidate form specifying unit 116 determines whether the form read by the reading unit 111 has a format common to the temporary form stored in the temporary registration storage area.

  When the form read by the reading unit 111 is not a form having the format stored in the form DB 122, the temporary form generation unit 117 generates a temporary form that associates items with the reading order of the items.

  The error determination unit 118 determines whether there is an error in the form read by the reading unit 111. If the error determination unit 118 determines that there is an error, the error determination unit 118 classifies the contents of the error and causes the storage unit 120 to store them. The error is, for example, “1” when a character that is difficult to read is included, “2” when the entry is not specified for the item, and “3” when the item is not specified for the entry. Associated with each description information.

  The display unit 119 displays a screen on the display device. For example, the display unit 119 displays a form correction screen, which will be described later, for a form including an entry determined to have an error.

  The storage unit 120 stores an item DB 121, a form DB 122, a customer master DB 123, and a temporary registration storage area 124.

  The item DB 121 is information regarding item candidates. In the item DB, an item attribute including a character string that is the item content is associated with an attribute of an entry to be entered corresponding to the item. Based on the item DB 121, the entry corresponding to the item is specified.

  The form DB 122 stores information on the form format. The form DB 122 includes a form name, an identifier of an item included in the form, a reading identifier indicating a reading order of entries corresponding to the item, and the like for each form format.

  The customer master DB 123 stores, for each read form, an item identifier that identifies an item, an input character string that is an entry entered for the item, and the like.

  The temporary registration storage area 124 stores the temporary form generated by the temporary form generation unit 117.

  In the present embodiment, customer information is generated from the read form. At this time, if a format corresponding to the read form exists in the form DB 122, a customer master DB is generated based on the form DB 122. If the corresponding form does not exist in the form DB 122, it is determined whether or not a temporary form based on the read form and the corresponding form exists in the temporary registration storage area 124, and the temporary form corresponding to the temporary registration storage area 124 is determined. Is present, customer information is generated according to the format of the temporary form. If the temporary form corresponding to the read form does not exist in the temporary registration storage area 124, a temporary form is generated based on the read form and stored in the temporary form storage area 124. As a result, even when a form whose form is not registered in the form DB 122 in advance is read, customer information can be generated efficiently. Therefore, processing from reading a form to generating customer information is efficient. Can be

  Next, each piece of information stored in the storage unit 120 will be described.

  FIG. 2 is a diagram illustrating an example of the item DB 121. The item DB 121 stores an item identifier 121a, item content 121b, the number of digits 121c, an attribute 121d, and a type 121e for each item.

  The item identifier 121a is identification information uniquely assigned to each item content 121b. The item content 121b is a character string indicating the content of the item. The number of digits 121c is a value indicating the upper limit of the number of single-byte digits of the entry entered for the item.

  The attribute 121d is an attribute of an entry entered for the item. In the attribute 121d shown in FIG. 2, “C”, “N”, and “S” are displayed. “C” means a full-width character, “N” means a half-width character, and “S” means an option. Yes.

  The type 121e is information that identifies the type of item. In the type 121e shown in FIG. 2, “aa” is a type related to “customer name”, “ab” is a type related to “home address”, and “ac” is a type related to “home phone number”.

  For example, the uppermost record of the item DB 121 shown in FIG. 2 has an item identifier 121a of “00001” and an item content 121b of “reading”. Indicates that the number of digits 121c is “48” digits, and the entry attribute 121d is “C”, that is, a full-width character. The item type 121e is “aa”, that is, “customer name”.

  FIG. 3 is a diagram illustrating an example of the form DB 122. The form DB 122 stores a form name 122b, an item identifier 122c, an input field reading identifier 122d, and an input mandatory flag 122e in association with each other for each form type identifier 122a.

  The form type identifier 122a is identification information that identifies the form format. The form name 122b is a character string indicating the name of the form. The item identifier 122c is identification information that identifies an item included in the form, and corresponds to the item identifier 121a of the item DB 121. When the form specified by the form type identifier 122a has a plurality of items, a plurality of item identifiers 122c are associated with the form type identifier 122a. Each item identifier 122c is associated with the form type identifier 122a in accordance with the item description order. The item identifiers 122c shown in FIG. 3 are recorded so that the description order follows from the top to the bottom.

  When the reading unit 111 reads a form, for example, the form is scanned in the horizontal direction from left to right, and then the reading head is moved in the lower left direction and further scanned in the horizontal direction from left to right. Therefore, each item identifier 122c is defined as having a description order later from the upper left to the lower right in the state described in the form. Hereinafter, the description order and the reading order will be described using the same definition. In other words, the present embodiment is not limited to the reading order and the description order, as long as the description order or the reading order of the description information is defined by a common rule.

  The input column reading identifier 122d is identification information that specifies the reading order of the entries input for the item specified by the item identifier 122c. As described above, the description information read by the reading unit 111 is numbered according to the reading order. The input field reading identifier 122d is a number assigned to the entry.

  The input mandatory flag 122e is a flag indicating whether or not the entry for the item specified by the item identifier 122c is essential. For example, in the input mandatory flag 122e shown in FIG. 3, “0” indicates that the input is not essential, and “1” indicates that the input is essential.

For example, in the form DB 122 shown in FIG. 3, the form name 122b is “transfer application” and the item identifier 122c is “00001” for the form having the form type identifier 122a specified by “AA-AAA”. “00003”, “00001”, “00007” and the like are included.
The form includes these item identifiers 122c in the order of “00001”, “00003”, “00001”, and “00007”. It should be noted that the item whose item identifier is “00001” has the item content 121b “reading” according to the item DB 121 of FIG. For the item whose item identifier is “00003”, the item content 121b is “name”.

  In the uppermost record of the form DB 122, the entry entered for the item having the item identifier 122c of “00001”, that is, the item content of “reading”, is specified by the input field reading identifier 122d of “002”. This indicates that the description information is read in the reading order. The entry indicates that the input mandatory flag 122e is “1”, that is, the input is essential.

  FIG. 4 is a diagram illustrating an example of the customer master DB 123. The customer master DB 123 stores a form type identifier 123b, an item identifier 123c, an item type 123d, an input character string 123e, and error information 123f in association with each other for each form identifier 123a.

  The form identifier 123a is identification information uniquely assigned to the read form. The form type identifier 123b is identification information for specifying a format when the form specified by the form identifier 123a corresponds to the format stored in the form DB 122, and corresponds to the form type identifier 122a of the form DB 122. If the corresponding format does not exist in the form DB 122, nothing is stored in the form type identifier 123b.

  The item identifier 123c is identification information that identifies an item included in the form, and corresponds to the item identifier 121a of the item DB 121. The item type 123d is information for specifying the type of item. The item type 123d corresponds to the type 121e of the item DB 121. The item type 123d may be a symbol specifying the type, or a character string indicating the type.

  The input character string 123e is a character string indicating an entry entered for the item specified by the item identifier 123c. The error information 123f is information for specifying an error classification when an error determination is made on the item specified by the item identifier 123c or the input character string 123e.

  As described above, when the error information 123f is “1”, it is a case where characters that are difficult to read are included. Therefore, in the customer master DB 123 shown in FIG. 4, the item identifier 123c is “00007”, that is, according to the item DB 121, the input character string 123e is “Osaki Shinagawa-ku, Tokyo” for the item whose item content 121b is “current address”. .. ”, But the error information 123f associated with the item is“ 1 ”, which indicates that the item or the description includes characters that are difficult to read.

  Further, when the error information 123f is “2”, it indicates that no entry is specified for the item. In the customer master DB 123 of FIG. 4, the input character string 123e is empty for the item whose item identifier 123c is “01025”, and “2” is associated with the error information 123f. This indicates that the entry cannot be specified even though the entry is indispensable for this item.

  Further, when the error information 123f is “3”, it indicates that no item is specified for the entry. In the customer master DB 123 of FIG. 4, “3” is associated with the error information 123 f for the input character string 123 e of “3/1 12:00”. This indicates that no item has been identified for this entry.

  FIG. 5 is a diagram illustrating a hardware configuration example of the information processing apparatus 10. The information processing apparatus 10 includes a CPU (Central Processing Unit) 130, a memory 131, a network I / F (Interface) 132, an auxiliary storage device 133, an input I / F 134, and an output I / F 135. The components are connected by a bus.

  The CPU 130 is a central processing unit, and executes processing according to a program recorded in the memory 131 or the auxiliary storage device 133. In the information processing apparatus 10, processing is performed by the CPU 130 that operates according to a program read out on the memory 131 or the auxiliary storage device 133. Each processing unit constituting the control unit 110 realizes each function by the CPU 130 executing a program.

  The memory 131 is a storage device such as a RAM (Random Access Memory) or a flash memory, and functions as a storage area from which programs and data are temporarily read. The network I / F 132 is an interface for connecting the information processing apparatus 10 to the network. The auxiliary storage device 133 is, for example, an HDD (Hard Disk Drive), a CD-R (Compact Disc-Recordable), a DVD-RAM (Digital Versatile Disk-Random Access Memory), etc. Device. The function of the storage unit 120 is realized by the memory 131 or the auxiliary storage device 133. The function of the storage unit 120 may be realized by a storage device connected via the network I / F 132.

  The input I / F 134 is an interface for accepting an input operation from a system administrator, and accepts input from, for example, a touch panel, a keyboard, a mouse, a microphone, and the like. The output I / F 135 is an interface for outputting information to a display device such as an LCD (Liquid Crystal Display).

  An input I / F 134 of the information processing apparatus 10 is connected to the scanner 20, and image data read by the scanner 20 is acquired by the control unit 110 via the input I / F 134. The scanner 20 may be connected to the output I / F 135. Further, the information processing apparatus 10 may have a scanner inside.

  Note that the processing of each component of the information processing apparatus 10 may be executed by one hardware or a plurality of hardware. Further, the processing of each component of the information processing apparatus 10 may be realized by a single program or may be realized by a plurality of programs.

  FIG. 6 is a flowchart (part 1) showing the flow of the form reading process. For example, when a process start instruction is received via the input device, the process of this flowchart is started. By this processing, one form is read. Regardless of the format, for example, when 100 forms are read, this process may be repeated for the number of sheets.

  First, the reading unit 111 reads a form (step S11). The reading unit 111 acquires image data of a form read by the scanner 20. The reading unit 111 stores image data in an area (not shown) in the storage unit 120.

  Next, the written information extraction unit 112 performs OCR processing (step S12). The description information extraction unit 112 acquires character data of description information described in a frame for each frame included in the form.

  FIG. 8 is an example of a form. In the form, “Furigana” is displayed in the upper left frame, “Sweet Witch” is displayed to the right of the frame, and “Gender” is displayed in the right frame. The description information extraction unit 112 extracts the description information described in the frame according to the reading order from the upper left to the lower right. When the “customer information registration application” which is the title of the form is surrounded by a frame, the title is also extracted as description information.

  FIG. 9 is a diagram illustrating an example of description information extracted by the description information extraction unit 112. An example of the description information shown in FIG. 9 is extracted from the form shown in FIG. The description information extraction unit 112 attaches a reading identifier according to the reading order of the extracted description information.

  Returning to FIG. Next, the written information extraction unit 112 specifies a title (step S13). The description information extraction unit 112 identifies character data existing in a predesignated area as a form title. For example, the description information extraction unit 112 identifies a character described in the upper center area of the read form as a title. Or the description information extraction part 112 specifies the description information described in the character larger than the other description information among the description information acquired by step S12 as a title. In step S12, if a reading identifier is attached to the title, the reading identifier is deleted, and a reading identifier is newly attached to the description information other than the title.

  Next, the item extraction unit 113 extracts items from the description information (step S14). Specifically, the item extraction unit 113 refers to the item DB 121 using the description information extracted in step S12, and when the item content 121b corresponding to each description information exists, the item extraction unit 113 is associated with the item content 121b. An item is extracted by specifying the item identifier 121a. The item extraction unit 113 associates the identified item identifier with the extracted item.

  For example, in the description information shown in FIG. 9, “reading” of the description information associated with the reading identifier “001” corresponds to the item identifier 121 a “00001” in the item DB 121. Therefore, the item extraction unit 113 associates “00001” as the item identifier with this description information. Similarly, “00013” is associated as the item identifier for the description information “sex” whose reading identifier is “003”, and the item for the description information “name” whose reading identifier is “004”. Associate "00002" as an identifier.

  Next, the item extraction unit 113 associates attributes with the description information that is not extracted as items in step S14 (step S15). Specifically, the item extraction unit 113 sets the attribute to “N” when the description information that is not extracted as an item includes a predetermined number or more of alphanumeric characters, and sets the attribute when determining that the description information is an option. When the description information is other than “S”, the attribute is “C” and is associated with each description information. For example, the item extraction unit 113 identifies the attribute as “N” when the description information includes 6 or more alphanumeric characters. The options will be described later.

  FIG. 10 is a diagram illustrating an example of description information in which item identifiers and attributes are associated. An item identifier is not associated with “Satoichiro”, which is the description information identified by the reading identifier “002”. The item extraction unit 113 associates the attribute “C” with the description information. The item extraction unit 113 associates the attribute “N” because “July 1, 1962”, which is the description information identified by the reading identifier “012”, contains six numbers.

  FIG. 11 is a diagram illustrating an example of options as description information. In FIG. 11A, “alphabet +“. ”” Is added to each option. In FIG. 11B, a check box is displayed for each option. In an area (not shown) of the storage unit 120, a plurality of patterns indicating options are stored in advance, and the item extraction unit 113 associates the attribute “S” when the description information matches the option pattern.

  In addition to the above-mentioned “continuous alphabet +“. ”” And “multiple check boxes”, the selection pattern includes “continuous multiple numbers +“. ””, “Continuous multiple kana characters“. ””. Etc. In FIG. 10, for the read information “1. male 2. female” (“1” is circled) specified by the read identifier “006”, the item extraction unit 113 sets “ It is determined that the number is “+” and the attribute “S” is assigned.

  When the item extraction unit 113 determines that the attribute of the description information is an option, the item extraction unit 113 may determine which option is selected and overwrite the description information with the selected limb. Thereby, customer data DB can be generated using the selected limb.

  Returning to FIG. Next, the candidate form specifying unit 116 refers to the form DB using the title included in the read form, the item identifier, and the order of each item identifier, and the order of the title, the item identifier, and each item identifier. It is determined whether or not there is data in the form DB 122 (step S16). Specifically, the candidate form specifying unit 116 refers to the item DB 121 using the title specified in step S13 and the item identifier of the item extracted in step S14. The candidate form specifying unit 116 determines whether or not the item identifier 122c associated with the form name 122b corresponding to the specified title corresponds to the extracted item identifier.

  As described above, the item identifier 122c of the form DB 122 is stored according to the description order for the form. The candidate form specifying unit 116 further specifies the reading order using the reading identifier attached to the item identifier of the item extracted in step S14, and the reading order and the description order of each item identifier 122c in the form DB 122 are determined. It is determined whether or not corresponding data exists in the form DB 122.

  When the candidate form specifying unit 116 determines that there is data in the form DB 122 in which the title included in the read form, the item identifier, and the order of each item identifier match (in the case of “YES” in step S16) The entry specifying unit 114 specifies an input mandatory flag and a description for each item (step S17). Specifically, the entry specifying unit 114 refers to the data in the form DB 122 that has been determined that the title, the item identifier, and the order of each item identifier included in the form read in step S16 match. The entry specifying unit 114 specifies each item identifier of the item extracted in step S14 and the corresponding item identifier 122c in the form DB 122. The entry specifying unit 114 specifies the input field reading identifier 122d associated with the specified item identifier 122c and the input mandatory flag 122e.

  The entry specifying unit 114 specifies the description information corresponding to each of the specified input field reading identifiers 122d and the reading identifier assigned to the description information in step S12 as an entry for each item. The entry specifying unit 114 associates each extracted item identifier, the entry corresponding thereto, and the input mandatory flag.

  Next, the error determination unit 118 has at least one of (1) no entry related to the item identifier, (2) no item identifier related to the entry, and (3) a predetermined error in the written information. If this is the case, error information is associated with the written information (step S18). Specifically, the error determination unit 118 associates the item identifier with the item identifier when the input mandatory flag identified in step S17 is not associated with the item indicating that the input is mandatory in the same step. Error information indicating that there is no entry to be associated is associated with the item identifier.

  In addition, the error determination unit 118 confirms that there is no item identifier associated with the entry for the description information that is not extracted as an item in step S14 and is not associated with the item identifier in step S17. Associate the error information shown. In other words, it can be said that the description information that is not extracted as an item and that is not associated with the item identifier is description information that is not specified as an item or an entry.

  In addition, when there is a predetermined error in the entry, the error determination unit 118 associates error information indicating that with the description information. The predetermined error is, for example, a case where written information includes characters that are difficult to read. For example, the error determination unit 118 refers to the item DB 121 using the item identifier assigned to the item and the entry associated with the item, and the digit associated with the corresponding item identifier 121a. When the number of entries exceeds the number 121c, or when the attribute 121d is different from the attribute of the entry, error information indicating that may be associated with the description information.

  Next, the customer information generation unit 115 creates the customer master DB 123 by associating each item identifier with the input information (step S19). Specifically, the customer information generation unit 115 attaches a form identifier 123a to the form, and associates the form type identifier 122a of the form DB 122 determined to have a format corresponding in step S16 as the form type identifier 123b. The customer information generation unit 115 associates the item identifier of the item extracted in step S14 with the form identifier 123a as the item identifier 123c according to the reading order.

  Further, the customer information generation unit 115 refers to the item identifier 121a corresponding to each item identifier 123c and the item DB 121, and identifies the type 121e associated with the item identifier 121a. The customer information generation unit 115 associates the identified type 121e with the item identifier 123c as the item type 123d. Note that the item type 123d may be a character string indicating the type of the item as described above.

  Further, the customer information generation unit 115 associates the entry associated in step S17 with each item identifier 123c as the input character string 123e.

  Also, the description information that is not extracted as an item and is not associated with the item identifier is stored in the input character string 123d. When the error information is associated with the description information specified by each item identifier 123c or the input character string 123e in step S18, the associated error information is stored in the error information 123f, and the customer master DB 123 Is generated. A plurality of pieces of error information may be associated with one piece of description information. The customer information generation unit 115 thereafter ends the process of this flowchart.

  In step S16, the candidate form specifying unit 116 does not determine that there is data in the form DB 122 in which the title included in the read form, the item identifier, and the order of each item identifier match (NO in step S16). ”), The candidate form identification unit 116 proceeds with the process to step S20.

  FIG. 7 is a flowchart (part 2) showing the flow of the form reading process.

  Next, the candidate form specifying unit 116 determines whether or not the provisional registration data area 124 includes data on a temporary form in which the title included in the read form, each item identifier, and the order of each item identifier match. Determination is made (step S20).

  In the temporary form storage area, data relating to forms that have been read before the start of the process and whose form title, items included in the form, or the order in which items are described differs from the data in the form DB 122 is stored. Yes. In the temporary form storage area, for each form, description information including items included in the form and the description order of the items is stored in association with the title of the form.

  The candidate form specifying unit 116 refers to the temporary form storage area using the title specified in step S13 and the item identifier of the item extracted in step S14. The candidate form specifying unit 116 specifies the data in the temporary form storage area where the specified title matches the extracted item. The candidate form specifying unit 116 specifies the reading order using the reading identifier attached to the item identifier of the item extracted in step S14, and refers to the data specified in the temporary form storage area. The candidate form specifying unit 116 determines whether there is data in which the reading order of items matches the description order among the specified data.

  When the candidate form specifying unit 116 determines that there is data in the temporary registration storage area 124 in which the title included in the form, each item identifier, and the order of each item identifier match (“YES” in step S20) ), The entry specifying unit 114 specifies an entry for each item (step S21).

  Identification of the entry by the entry identification unit 114 will be described using the description information of FIG. First, the entry specifying unit 114 specifies the description information with the youngest reading identifier among the description information associated with the item identifier. The entry specifying unit 114 refers to the item DB 121 using the item identifier associated with the description information, and specifies the attribute 121d associated with the item identifier 121a corresponding to the item identifier. Next, the entry specifying unit 114 specifies description information that has a description order later than the specified description information and that is not associated with an item identifier, in the order from the specified description information. When the specified description information matches the attribute 121d, the entry specification unit 114 specifies the description information as an entry for the item identifier.

  The entry specifying unit 114 may further specify the corresponding entry using the number of digits 121c associated with the item identifier 121a in the item DB. Specifically, the entry specifying unit 114 determines that the description information does not correspond to the item when the description information determined to match the attribute exceeds the number of digits 121c associated with the item.

  The entry specifying unit 114 repeats the above processing from the lowest reading order with respect to all the description information associated with the item identifier.

  In FIG. 10, the description information associated with the item identifier and having the smallest reading identifier is the description information relating to the reading identifier “001”. The described information is associated with an item having an item identifier “00001”. The entry specifying unit 114 refers to the item DB 121 using “00001” as the item identifier, and specifies “C” as the attribute 121d associated with the item identifier 121a. The attribute “C” indicates a full-width character.

  Next, the entry specifying unit 114 reads the information closest to the reading identifier “001” among the description information after the description information related to the reading identifier “001” and not associated with the item identifier. Identify the order information. Therefore, the entry specifying unit 114 specifies the description information related to the reading identifier “002”. The reading identifier “002” is associated with the description information “Satoichiro”. The entry specifying unit 114 determines whether or not the written information “Satoichiro” matches the attribute “C”. Since “Sweet Ichiro” is written in double-byte characters, it is determined that it matches the attribute “C”. The entry specifying unit 114 specifies the description information related to the read identifier “002” as an entry for the item related to the read identifier “001”.

  Next, the entry specifying unit 114 specifies the description information related to the read identifier “003” as the description information associated with the item identifier, and the description information having the youngest reading order after the read identifier “001”. . The description information related to the reading identifier “003” is associated with the item identifier “00013”, and the attribute associated with the item in the item DB 121 is “S”.

  The entry specifying unit 114 is the description information whose reading order is later than the description information related to the reading identifier “003”, and is the description information with the youngest reading order among the description information not associated with the item identifier. The description information related to the reading identifier “005” is specified. The reading identifier “005” is associated with the description information “Ichiro Sato”, but the attribute of the description information “Ichiro Sato” is “C” which is a full-width character and the attribute “S” associated with the item identifier. is not.

  Next, the entry specifying unit 114 specifies the description information associated with the reading identifier “006”, which is the description information with the next lowest reading order. Since the description information is an option, it matches the attribute “S”. Therefore, the entry specifying unit 114 specifies the description information related to the reading identifier “006” as an entry for the item related to the reading identifier “003”.

  As a result of specifying the entry items for each of the items included in the description information according to FIG. 10 according to the above-described processing, the correspondence relationship indicated by the arrow shown on the right side of FIG.

  Next, the candidate form specifying unit 116 determines whether each item and each entry are in a one-to-one relationship (step S22). Specifically, the candidate form specifying unit 116 determines that each item and each entry item when there is no description information that is not associated with the item identifier and is not associated with the item identifier in step S21. Are determined to be one-to-one.

  When the candidate form specifying unit 116 determines that each item and each entry are in a one-to-one relationship (in the case of “YES” in step S22), the error determination unit 118 has a predetermined error in the written information The error information indicating that is associated with the description information (step S23). As described above, the predetermined error is, for example, a case where the description information includes a character that the reading unit 111 cannot read.

  Next, the customer information generation unit 115 creates a customer master DB 123 by associating each item identifier with the input information for the form read in this process (step S24). The form to be processed in this step does not conform to the format defined in advance in the form DB 122, but the form read in common before the start of this process is common, and each item and entry are excessive. It is a form that corresponds without a shortage. In this step, the customer master DB 123 is generated for the form read in this process.

  Specifically, the customer information generation unit 115 attaches a form identifier 123a to the form, and associates the item identifier of the item extracted in step S14 with the form identifier 123a as the item identifier 123c according to the description order. Further, the customer information generation unit 115 refers to each item identifier 123c and the corresponding item identifier 121a in the item DB 121, identifies the type 121e associated with the item identifier 121a, and associates it with the item identifier 123c as the item type. Further, the customer information generation unit 115 associates the entry associated in step S21 with each item identifier 123c as the input character string 123e. When error information is associated with the description information in step S23, the error information 123f is associated with the description information.

  Next, the customer information generating unit 115 creates a customer master DB by associating each item identifier with the input information when each item of the temporary form is one-to-one, and creates the temporary form. Is deleted from the temporary registration storage area 124 (step S25). Specifically, the customer information generation unit 115 determines in step S20 the data in the temporary registration storage area 124 that has been determined that the title included in the read form, each item identifier, and the order of each item identifier match. A customer master DB is generated using (provisional form). In the temporary registration storage area 124, for example, the description information shown in FIG. 10 is stored in association with the title of the form. The customer information generation unit 115 attaches a form identifier 123a to the temporary form in the temporary registration storage area 124, and associates the item identifier 123c, the item type 123d, and the input character string 123e. If error information is associated with the description information in step S23, the error information 123f is associated with the description information.

  Further, the customer information generation unit 115 deletes the temporary form data from the temporary registration storage area 124. Thereafter, the customer information generation unit 115 ends the process of this flowchart.

  In step S22, when the candidate form specifying unit 116 does not determine that each item and each entry is 1: 1 (“NO” in step S22), the error determining unit 118: (1) item identifier Associate error information when there is at least one of the following: (2) There is no item identifier associated with the entry, (3) There is a specified error in the written information ( Step S26). Specifically, the error determination unit 118 sets error information indicating that there is no entry related to the item identifier in the item identifier when no entry is specified in step S21 for the item extracted in step S14. Associate with each other.

  Further, the error determination unit 118 confirms that there is no item identifier associated with the entry for the description information that is not extracted as an item in step S14 and is not associated with the item identifier in step S21. Associate the error information shown. In addition, when there is a predetermined error in the entry, the error determination unit 118 associates error information indicating that with the description information.

  Next, the temporary form generation unit 117 stores the temporary form in which each item identifier is associated with the input information and the image data of the form in the temporary registration storage area 124 (step S27). Specifically, the provisional form generation unit 117 associates the reading identifier, the item identifier extracted in step S14, and the reading identifier of the entry specified in step S21 with respect to the description information acquired in step S12. Then, a temporary form is generated in correspondence with the title specified in step S13 and stored in the temporary registration storage area 124. The temporary form is the description information shown in FIG. 10, for example.

  Further, the temporary form generation unit 117 reads out the image data read by the reading unit 111 in step S11 from the storage unit 120 and associates it with the temporary form. Thereafter, the temporary form generation unit 117 ends the process of this flowchart.

  In step S20, the candidate form specifying unit 116 does not determine that there is data in the temporary registration storage area 124 in which the title included in the form, each item identifier, and the order of each item identifier match (in step S20). In the case of “NO”, the entry specifying unit 114 specifies an entry for each item (step S28). Since the process performed in step S28 is the same as the process performed in step S21, description thereof is omitted.

  Next, the error determination unit 118 (1) there is no entry related to the item identifier, (2) there is no item identifier related to the entry, (3) the description information acquired in step S12. When at least one of the predetermined errors is found, error information is associated (step S29). The processing performed from step S29 to step S30 is the same as the processing performed from step S26 to step S27, and thus description thereof is omitted.

  According to the present embodiment, when the read form corresponds to the format stored in advance in the form DB 122, the customer master DB 123 is generated according to the format. In the case of not corresponding to the form DB 122, the temporary registration storage area 124 in which information related to the form format read before this processing is stored is referred to and the form corresponding to the temporary form format in the storage area is referred to. When a pair of an item and an entry for the item is specified, the customer master DB 123 is generated using these. Thereby, it is not necessary to register in advance the format of the handled form, and the description information can be extracted efficiently.

  Further, by repeating this process for the number of forms, a customer master DB relating to different types of forms can be suitably created. In particular, in a servicer that handles forms of different types, it is not necessary to sort the forms for each type, and work efficiency is improved.

  In addition, when it is considered that there is a problem in accuracy if the customer master DB 123 is generated as it is, such as when an entry for an item is unknown, correction using the form correction screen 140 described later is performed by attaching error information. And the customer master DB 123 can be generated appropriately.

  In this embodiment, in step S21, the item DB 121 is referred to using each item identifier, and the entry is specified using the attribute 121d (and the number of digits 121c) associated with the item identifier 121a. However, the attribute 121d and the number of digits 121c are not necessarily used as long as the entry for the item can be specified based on the description order. By omitting the process of referring to the attribute 121d and the number of digits 121c, the load on the information processing apparatus 10 can be reduced.

  FIG. 12 is a diagram illustrating an example of the form correction screen 140. The contents of the form stored in the temporary registration storage area 124 can be corrected on this screen. The display unit 119 displays the screen on the display device by receiving selection of a form stored in the temporary registration storage area 124 based on, for example, an input via the input device. The display unit 119 may automatically display this screen when a form including predetermined error information is stored in the temporary registration storage area 124. This screen is displayed for each form.

  The form correction screen 140 includes a description information selection area 141, a character string change button 142, a correspondence change button 143, and a customer information generation button 144. In addition, on the form correction screen 140, the description items included in the form are either (1) no entry related to the item identifier or (2) no item identifier related to the entry. In the case where the information is associated with the error information indicating “”, an area for displaying that the item does not correspond to the entry may be included.

  In the description information selection area 141, description information included in the form is displayed so as to be selectable. The character string change button 142 is a button for accepting a change of the character string included in the description information. When the description information displayed in the description information selection area 141 is selected and the input of the character string change button 142 is accepted, the display screen transitions to a screen that accepts an input relating to the change of the character string in the description information. For example, this operation is performed when the character string related to the entry extracted by the written information extraction unit 112 is changed.

  The correspondence change button 143 is a button for accepting addition or change of description information. When the descriptive information displayed in the descriptive information selection area 141 is selected and the input of the correspondence change button 143 is accepted, the display screen transitions to a screen for accepting input regarding the addition of the descriptive information and the correspondence between items and entries. For example, this operation is performed when adding description information that has not been extracted by the description information extraction unit 112 or when associating an entry with an item extracted by the item extraction unit 113.

  The customer information generation button 144 is a button for receiving a generation instruction of the customer master DB 123 based on this form. When the customer information generation button 144 is input, the customer master DB 123 is generated based on this form.

  With this display screen, it is possible to correct the error for the description information for which the error determination unit 118 has determined the error. Even for the forms stored in the temporary registration storage area 124, the customer master DB 123 can be generated after receiving an appropriate correction by performing an operation based on this screen. Thereby, generation | occurrence | production of the appropriate customer master DB123 based on the read form is ensured.

  Note that when performing an operation on the main display screen, the image data stored in the storage unit 120 in association with the temporary form may be displayed together with the main display screen. Thereby, the read image can be collated with the main display screen, and the efficiency is improved.

  As mentioned above, although each embodiment which concerns on this invention has been demonstrated, this invention is not limited to an example of above-described embodiment, Various modifications are included. For example, the above-described exemplary embodiment has been described in detail for easy understanding of the present invention, and the present invention is not limited to the one having all the configurations described here. A part of the configuration of an example of an embodiment can be replaced with the configuration of another example. Moreover, it is also possible to add the structure of another example to the structure of an example of a certain embodiment. In addition, for a part of the configuration of an example of each embodiment, another configuration can be added, deleted, or replaced. Each of the above-described configurations, functions, processing units, processing means, and the like may be realized by hardware by designing a part or all of them with, for example, an integrated circuit. In addition, the control lines and information lines in the figure indicate what is considered necessary for the description, and do not necessarily indicate all of them. It can be considered that almost all configurations are connected to each other.

  In addition, the functional configuration of the information processing apparatus 10 is classified according to main processing contents in order to facilitate understanding. The present invention is not limited by the way of classification and names of the constituent elements. The configuration of the information processing apparatus 10 can be classified into more components depending on the processing content. Moreover, it can also classify | categorize so that one component may perform more processes.

  10: Information processing device, 20: Scanner, 110: Control unit, 111: Reading unit, 112: Description information extraction unit, 113: Item extraction unit, 114: Item specification unit, 115: Customer information generation unit, 116: Candidate Form specifying unit, 117: provisional form generation unit, 118: error determination unit, 119: display unit, 120: storage unit, 121: item DB, 122: form DB, 123: customer master DB, 124: temporary registration storage area, 130: CPU, 131: Memory, 132: Network I / F, 133: Auxiliary storage device, 134: Input I / F, 135: Output I / F, 140: Form correction screen, 141: Description information selection area, 142: Character string change button, 143: correspondence change button, 144: customer information generation button

Claims (7)

A storage unit that stores item information related to item candidates that are candidates for items included in the form;
A reading unit for reading a form;
A description information extraction unit that extracts description information from the form read by the reading unit;
An item extraction unit for extracting the item corresponding to the item candidate from the description information;
Based on the reading order of the description information by the reading unit, an entry specifying unit for specifying an entry for the item from the description information;
A customer information generating unit that generates customer information in which the item and the entry for the item are associated;
An information processing apparatus comprising:
The information processing apparatus according to claim 1,
The item information associates the item candidate with the attribute of the entry for the item,
The entry specifying unit specifies the entry for the item based on the reading order, the item extracted by the item extracting unit, and the attribute associated with the item information. Information processing apparatus.
An information processing apparatus according to claim 2,
A candidate form specifying unit that determines whether a form candidate corresponding to the form read by the reading unit is stored in the storage unit;
The storage unit stores, for each of the forms, form information that associates the description order of the items with the description order of the entries in the items for each item as the form candidate,
The described information extraction unit
The candidate form specifying unit refers to the form information using the items extracted by the item extracting unit and the reading order of the items, and the form information corresponding to the reading order and the description order of the items Is determined to be stored in the storage unit, the description information read in the reading order corresponding to the description order of the entries in the form information is specified as the entry,
The information processing apparatus, wherein the customer information generation unit associates the item with the entry using the form information.
The information processing apparatus according to claim 3,
The items extracted by the item extraction unit and the description order of the items for the forms that are not determined to be included in the form information by the candidate form specifying unit, the forms corresponding to the reading order and the description order. And a temporary form generation unit that generates a temporary form using
The storage unit stores the temporary form as the form candidate,
The candidate form specifying unit determines whether or not the temporary form corresponding to the reading order and the description order of the form is stored in the storage unit,
The entry specifying unit specifies the entry using the description order included in the temporary form when the candidate form specifying unit determines that the temporary form is stored. Information processing apparatus.
An information processing apparatus according to any one of claims 1 to 4,
Among the description information, when the entry for the item is not specified by the entry specification unit, or when there is the description information that is neither specified by the item nor the entry, an error determination unit that performs error determination; ,
A display unit that displays a form correction screen when an error is determined by the error determination unit;
The information processing apparatus, wherein the form correction screen accepts designation of the item or the entry for the description information.
A form reading method using an information processing apparatus including a control unit,
The controller is
A storage unit that stores item information related to item candidates that are candidates for items included in the form;
A reading step for reading a form;
A description information extraction step of extracting description information from the form read in the reading step;
An item extraction step of extracting the item corresponding to an item candidate that is a candidate for an item included in the form stored in the storage unit from the description information;
Based on the reading order of the description information in the reading step, an entry specification step for specifying an entry for the item from the description information;
A customer information generating step for generating customer information in which the item and the entry for the item are associated;
A form reading method comprising:
A program for causing a computer to execute a step of performing a form reading process,
Causing the computer to function as control means;
For the control means,
A storage unit that stores item information related to item candidates that are candidates for items included in the form;
A reading step for reading a form;
A description information extraction step of extracting description information from the form read in the reading step;
An item extraction step of extracting the item corresponding to an item candidate that is a candidate for an item included in the form stored in the storage unit from the description information;
Based on the reading order of the description information in the reading step, an entry specification step for specifying an entry for the item from the description information;
A customer information generating step for generating customer information in which the item and the entry for the item are associated;
A program characterized by having executed.

Images