JP2014041442A - Receipt definition data preparation device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable an efficient preparation of definition data required upon collecting sales management data for managing sales of each tenant from various receipts by character recognition.SOLUTION: An image data analysis unit 12 determines a character string area recognized by a character from image data on a receipt as a character item and a money amount column, a relative position to the money amount column is calculated in association with the money amount column having a specific position relation with the character item, and reading information is prepared. A reading item calculation unit 13 extracts a character item coinciding with a name corresponding to a sales management item from the reading information as a reading item while referring to an item dictionary storage unit 103. A definition input screen display unit 14 displays a definition input screen including the sales management item, information on the character item extracted as the reading item of the sales management item and the image data on the receipt. An input data processing unit 15 receives input of a calculation expression of the reading item of the sales management item from the definition input screen. A definition data preparation unit 16 prepares, after input processing of the input data processing unit 15, receipt definition data from information on the definition input screen.

Description

  The present invention relates to a receipt definition data creation apparatus and program, and in particular, necessary for accurately extracting necessary data from receipt data read by character recognition in order to manage sales at a plurality of stores such as a shopping mall. The present invention relates to a technology that supports the creation of definition data.

  The tenant management department that manages stores of tenants (such as companies that operate stores) that open stores in shopping centers, shopping malls, etc. collects daily sales for each store, and breakdowns of sales such as cash sales, coupon sales, etc. ( Sales management item). When creating the sales breakdown, payment receipts provided from each store are collected and the numerical value of the sales management item is calculated. However, since the format of the payment receipt provided from each store depends on the POS system installed in each store, it is currently created in various formats. Therefore, the payment receipt processed by the tenant management department has a wide variety of items, item names, print sizes, print fonts, and the like.

  For example, regarding the names of the items on the payment receipt, even though the total sales, total sales, total sales, total sales, etc. indicate the same contents, different names are printed depending on the POS system. In addition, since the tenant's store that opens a store is frequently replaced, a new payment receipt format must also be supported. For this reason, the tenant management department requires a great deal of labor for the sales totaling work for totaling the payment receipts as unified sales management data, and the human and time burden for that is very expensive. It is.

  Note that Patent Document 1 discloses a receipt processing apparatus. In this receipt processing apparatus, text data is generated by recognizing characters from the receipt image, a predetermined character string is selected from the text data, and a line position including the selected predetermined character string is detected. Then, line data in units of lines is sequentially selected from a predetermined line range on the text data determined based on the detected line position, and data relating to the item is extracted from the line data.

  Patent Document 2 discloses a technique for automatically creating a form definition for any form including an indefinite form without a ruled line without manual intervention. In this technique, a receipt is read as an image, a stamped item is cut out as a table area, x-coordinate and y-coordinate are calculated for each table area, and a frequency distribution table of x-coordinate is created. Then, the form definition is created by determining the alignment and determining the product field, the number field, and the price field.

JP 2004-164218 A JP 2006-178802 A

  In the tenant management department of a company (organization) that operates a shopping center or the like, the clerk has traditionally used to collect the payment receipts (hereinafter referred to as receipts) provided by each store such as a shopping center. The receipt was read visually, and the read data was manually input to the data processing device. The work amount of the sales totaling work is very heavy because the number of stores may reach 100 stores or more and must be performed every day.

  As a proposal for reducing this burden, it is conceivable to automatically collect sales data (receipt data) of each store from a terminal of each store via a network. However, in general, each store has a different business entity, so it is necessary to obtain permission to open the network individually, and the type of network to be connected and the difficulty of adding / modifying applications differ depending on the installed system. There are difficult problems.

  Another way to reduce the above-mentioned work load by changing the existing system, especially information devices such as terminals in each store as much as possible, is to recognize character images instead of reading receipt data by humans. The reading method can be considered. That is, the receipt of each store is read by, for example, a reader or scanner having an OCR (Optical Character Recognition) function, and the necessary receipt items and their numerical data are extracted as sales management items, and the sales management data is extracted. This is a method to re-enter.

  In this way, when recognizing a character image of a receipt and automatically creating sales management data, correspondence information on how sales management items required as sales management data correspond to read items of receipts, In addition, it is necessary to prepare in advance the receipt definition data as the character string (item name) of the read item of the receipt and the position information on the image of the amount column corresponding to the read item.

  Once this receipt definition data is created and registered in the receipt recognition device, the receipt image is read by a reading device or the like having an OCR function, and the receipt items necessary for creating sales management data are extracted from the read receipt image. , It is thought that daily sales management data can be collected to some extent automatically.

  The problem is how to create this receipt definition data. First, it is conceivable to manually input the receipt definition data one by one. However, in a large-scale shopping mall or the like, the number of stores may be more than 100, and the receipt definition data is manually input. This is a huge amount of work. Furthermore, in a shopping mall or the like, stores that are opened frequently are frequently replaced, and it is troublesome to manually input new receipt definition data every time a new store joins.

  Therefore, a computer-aided technology capable of creating receipt definition data as easily as possible is desired. However, the technology for automatically creating a form definition from the receipt image of Patent Document 2 is a case that is covered by the present invention. Is insufficient and cannot be used for the following reasons.

  (1) In Patent Document 2, sales management data including a specific group of sales management items common to sales management of each store required by the tenant department is not considered at all.

  (2) Also, with the technology of Patent Document 2, it is possible to create item definition information for each receipt, but all items printed on the receipt are targeted, and only the necessary items must be selected. I can't.

  (3) In addition, the sales management items particularly required by the tenant management department do not always match the item names of the receipt items printed on the receipts of each store, and the sales management items and the receipt items are not necessarily It does not necessarily correspond one-on-one.

  FIG. 32 is a diagram illustrating an example of a correspondence relationship between sales management items and receipt items.

  The sales management items required by the tenant management department are, for example, “cash sales”, “payable sales”, “credit sales”,..., “Net sales” as shown in FIG. It is a unified item common to stores. In the receipt shown in FIG. 32B, the “cash sales amount” on the 14th line corresponds to the first sales management item “cash sales”, and the second sales management item “overtime sales”. There is no receipt item corresponding to, and the “other voucher amount” on the 18th line corresponds to the third sales management item “credit sales”.

  On the other hand, although not shown in the drawings, in the receipt of another store, for example, the sum of the receipt item “cash amount on daily report” and the receipt item “thank you card” is the first sales management item “cash sales”. The receipt item “Daily Overtime Sales” corresponds to the second sales management item “Overhead Sales”, and so on, the correspondence between each sales management item and the receipt item is Every store is completely different.

  In view of such a background, an object of the present invention is to provide means for easily creating receipt definition data for collecting sales management data by recognition of a receipt. That is, the present invention reads receipts of different formats output from a plurality of stores such as shopping malls, and easily and efficiently creates receipt definition data for acquiring only necessary sales management items accurately and efficiently. It aims to provide a means.

The receipt definition data creation device disclosed as one aspect of the present invention is as follows. In the following, a case where a sales management item is used as a specific basic management item common to each store collected from a receipt will be described as an example.
(1) For each basic management item (also referred to as a checkout item), an item dictionary is prepared in which each item is associated with the name of a reading item (also referred to as a receipt item) corresponding to the image data of the receipt.
(2) Read the receipt issued by the store where the receipt definition data for collecting sales management data is issued as an image, and calculate the amount based on the character string area obtained by character recognition of the receipt image data. Read information including the position of the character item other than the area to be shown, the correspondence with the amount column having a certain positional relationship, and the positional relationship between the character item and the corresponding amount column is created. Further, read information including information indicating whether the receipt is unique (unique) may be created.
(3) Necessary basic management items determined in advance are displayed on the screen, and receipt definition data is created by screen operation.

  At this time, with reference to the read information, for each basic management item, the corresponding character item on the receipt image data is set as the read item of the basic management item and displayed on the screen.

The receipt definition data created as described above includes the following information.
-Correspondence information between basic management items and read items in receipts. That is, information indicating which reading item in the receipt the amount to be recorded as a basic management item.
-Information indicating the positional relationship between the character string of the read item in the receipt and the corresponding money amount column.
• Information indicating the calculation formula when the amount of the basic management item is calculated by the four arithmetic operations of the amounts of the reading items.
・ In addition, when there are multiple character items with the same name (same character string) as the reading item in the receipt, the position of the character information and the reading character such as the marker item for identifying the reading item associated with the basic management item Information indicating the relationship.

  More specifically, the disclosed receipt definition data creation device is configured to execute each store executed by a receipt recognition system that collects basic management data including specific basic management items from images of receipts issued by a plurality of stores. A processing device for generating receipt definition data used for image recognition processing of a receipt, and a basic data storage means for storing information on basic management items, and names of items read from the receipt corresponding to the basic management items An item dictionary storage means for storing an item dictionary indicating image data, an image data storage means for storing image data of a receipt, a character recognition process for the image data, and a position of the recognized character string on the image data Information is acquired, and the recognized character string is an amount field for displaying an amount or other than the amount field It is determined whether it is a character item, the character item is associated with an amount field having a preset positional relationship, the character item, the associated amount field, and the associated amount field An image data analyzing means for creating read information including a positional relationship with the image data, and reading for extracting information on a character item that matches the name of the item corresponding to the basic management item from the read information with reference to the item dictionary As the item calculation means and receipt definition data information, the basic management item information, a reading item to be read from the image data for the basic management item, and a definition input screen for displaying the image data are generated and extracted from the reading information. And a definition input screen display means for displaying the definition input screen in which the information of the read character item is set as the information of the read item.

According to the disclosed receipt definition data creation device, the character information read from the receipt corresponding to the basic management item is displayed as the read item from the read information created from the character recognition result of the image data of the receipt. It is possible to work while confirming the displayed reading items. In addition, the operator can save the trouble of searching for a character item as a reading item from the image data of the receipt.
(4) Check the existence of a character item with the same name in the receipt. If there are multiple character items with the same name, the unique character item in the receipt is used as the marker item for identifying the read item. Display as a candidate.

  More specifically, in the receipt definition data creation device, when the reading item calculation means extracts information on a plurality of character items that match the names of items corresponding to the basic management items from the reading information. The unique character item in the image data is extracted from the read information, and the definition input screen display means sets the information on the unique character item as the information on the candidate item of the marker item that identifies the read item. The definition input screen is displayed.

  As a result, the operator can save the trouble of searching for a character string as a marker item from the receipt.

  (5) When reading item information is confirmed by a pointing operation or dragging operation using an input device such as a mouse on the screen, information on the character item extracted as the reading item from the reading information (character item to be the reading item) Receipt definition data is created based on the amount field corresponding to the character item and the positional relationship between the character item and the amount field. In addition, if a marker item is required, information on the marker item (position of the character item that became the marker item, target read item, marker, etc.) based on the information on the character item selected from the candidate items of the marker item (Positional relationship between item and amount column) is added to the receipt definition data.

  More specifically, the disclosed receipt definition data creation device extracts a character item set as a reading item of the basic management item from the receipt definition data, and uses the name of the extracted character item as the basic item of the item dictionary. Definition data analysis means for registering the name of the item corresponding to the management item is provided.

  Since the name of the read item associated with the sales management item can be collected from the created receipt definition data, it can be registered in the read character string related to the corresponding sales management item in the item dictionary used for the read information creation processing. The creation and update of item dictionaries can be automated. As a result, the operator can omit the work of creating the item dictionary.

The effects of the present invention are as follows.
(1) When reading items necessary for basic management items are extracted from receipts, receipt definition data indicating which basic management items correspond to which reading items can be easily created using actual receipts. . In particular, since the reading items necessary for basic management items are displayed on the screen in advance, the operator can omit the task of searching for the reading items from the receipt data. Reading items can be set easily and efficiently.
(2) When there are a plurality of character items with the same name for a reading item necessary for the basic management item from the receipt, a character item serving as a plurality of keywords for identifying the reading item and a marker serving as a reference for the position of the reading item Marker definition information such as items can be set in the receipt definition data. In particular, since character items that are candidates for marker items can be displayed on the screen, the operator can save time and effort to search for unique character items that become marker items from the receipt data, and there is little work experience. Even an operator can easily and efficiently set marker items.
(3) Since the item dictionary is automatically generated based on the created receipt definition data, the operator can omit the work of creating the item dictionary.

  As described above, by using the receipt definition data created according to the present invention, basic data required by the tenant's management department is uniformly collected from receipt data of various stores with different receipt reading items. , It can be stored in a basic data management database such as a sales management database.

It is a block diagram which shows the structural example of a receipt definition data creation apparatus. It is a flowchart which shows the outline | summary of a process of a receipt definition data creation apparatus. It is a figure which shows the example of the receipt read by the reader. It is a figure which shows the example of the image data stored in an image data storage part. It is a figure which shows the example of the reading information stored in a reading information storage part. It is a figure which shows the example of the item dictionary stored in an item dictionary memory | storage part. It is a figure which shows the example of the basic table stored in a basic data storage part. It is a figure which shows the example of a store master. It is a figure which shows the example of normal receipt definition data. It is a figure which shows the example of receipt definition data in case there exist two or more reading items of the same name in a receipt. It is a figure which shows the example of the receipt definition data which defines a computing equation. It is a figure which shows the example of the history data stored in a history data storage part. It is a figure which shows the example of the dictionary part of log | history data. It is a flowchart of the whole receipt definition data creation process. It is a flowchart of the whole receipt definition data creation process. It is a flowchart of the whole receipt definition data creation process. It is a figure which shows the structural example of a definition input screen. It is a figure which shows the example of a sales management item list. It is a figure which shows the example of a display of an OCR reading definition list display area. It is a figure which shows the example of a change of a reading item. It is a flowchart which shows the detail of an initial screen display process. It is a figure which shows the example of a display of an OCR reading definition list display area and a marker definition list display area when two or more character items of the same name exist. It is a flowchart which shows the detail of a calculation definition process. It is a figure which shows the example of a calculation definition. It is a figure which shows the example in which the correspondence of the receipt and sales management item of a certain store is defined by the calculation formula. It is a flowchart which shows the detail of a marker definition process. It is a figure which shows the example of a definition of a marker item. It is a figure which shows the example of a setting of a several keyword. It is a figure which shows the example of a setting of an alternative item. It is a flowchart which shows the detail of the process which judges the presence or absence of similar data. It is a flowchart which shows the detail of a history data creation process. It is a figure which shows the example of the correspondence of a sales management item and a receipt item.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following, an example of an apparatus for totalizing store sales will be described. However, the present invention is not limited to sales totalization, and totals actual data printed on purchase total data or other slips. The present invention can be applied to a device for the same.

  FIG. 1 shows an apparatus configuration example for explaining the outline of the present invention. In FIG. 1, reference numeral 1 is a CPU and a storage device such as a hard disk or memory, etc., and a receipt definition data creation device that executes processing for creating receipt definition data based on image data read from a receipt, and 2 is a receipt. A receipt for which definition data is to be created, 3 is a display device, and 4 is an input device such as a keyboard or a mouse.

  The receipt 2 is read by the reader 10 while being sandwiched between transparent carrier sheets and the like. The reader 10 is a device that reads an image on the receipt 2 by scanning an image using an optical method or the like. The image data reading unit 11 has a known OCR processing function and converts an image read by the reader 10 into predetermined image data. When the reader 10 has an OCR processing function, the image data reading unit 11 may not have an OCR processing function. The image data includes, for example, an image file in which image data is stored and character data recognized by OCR. The image data reading unit 11 stores the image data of the OCR processing result in the image data storage unit 101.

  The image data analysis unit 12 specifies the position of the rectangular area of the character string extracted by character recognition from the image data of the receipt stored in the image data storage unit 101, and the amount of money that has a predetermined positional relationship with the character item A reading information storage unit that creates a reading information that associates an item (amount field) with the item, a positional relationship between the character field and the amount field associated with the item, and information indicating whether the character item is unique in the receipt data 102. Details of the created read information will be described later.

  The item dictionary stored in the item dictionary storage unit 103 is dictionary information for registering, for each sales management item, the name (character string) of the character item read from the receipt corresponding to the item.

  The read item calculation unit 13 refers to the item dictionary stored in the item dictionary storage unit 103 and reads basic information stored in the basic data storage unit 104 from read information stored in the read information storage unit 102. For the item (sales management item), the information of the character item (read item) read from the image data of the receipt includes the character item to be read item, the associated amount field, and the positional relationship between the read item and the amount field. Extract information.

  The definition input screen display unit 14 displays a definition input screen on the display device 3. The definition input screen is a user interface for creating receipt definition data of a designated store, and information on basic sales management items stored in the basic data storage unit 104 and read information corresponding to the sales management items As a character item to be read from the receipt, position information of the money amount column corresponding to the character item to be read, information of the marker item, image data of the receipt stored in the image data storage unit 101, and the like. The definition input screen display unit 14 sets the character item information extracted by the read item calculation unit 13 as a read item, and displays it in the OCR reading definition list on the definition input screen. Further, the definition input screen display unit 14 displays a rectangle at the position of the money field corresponding to the character item on the image data of the receipt based on the positional information of the text item as the read item and the corresponding money field. Thus, the operator can check which character item is set as the read item and which amount column corresponds.

  The input data processing unit 15 inputs information on the read item of the displayed receipt and the correspondence between the basic management item and the character item (read item) from the displayed definition input screen. For this reason, the input data processing unit 15 reads the read item input unit 151 that acquires the read item input by the operation of the operator when the read item is changed, etc., into a single sales management item. An operation definition unit 152 that defines an operation expression to be associated is provided.

  Further, the input data processing unit 15 inputs and sets a plurality of keywords that serve as a reference for specifying the read item from the definition input screen when there are a plurality of character items with the same name extracted as read items in the receipt. A keyword setting unit 153 and a marker item setting unit 154 that determines other unique character items serving as a reference for specifying a reading item are provided. Further, the input data processing unit 15 causes the operator to designate another receipt item to be substituted for the read item as a substitute item when the input read item has a recognition error or the character recognition accuracy is poor. The alternative item processing unit 155 is input.

  The definition data creation unit 16 creates receipt definition data for each store based on the character item information extracted by the read item calculation unit 13 and the information input by the input data processing unit 15, and the definition data storage unit 106. To store.

  The receipt definition data includes information such as the item name in the read item of the receipt and the relative position of the corresponding monetary item (the position of the relative monetary column) and the correspondence information with the sales management item.

  The history data creation unit 18 creates history data of the receipt definition data created by the definition data creation unit 16 and stores it in the history data storage unit 107. This history data can be used by referring to the receipt definition data of other similar receipts.

  FIG. 2 is a flowchart showing an outline of processing of the receipt definition data creation device 1 shown in FIG.

  In step S <b> 11, the image data reading unit 11 reads the payment receipt by the reader 10. The payment receipt is total data of sales information for one day at each store, and the receipt item varies depending on the POS system used by each store. The image data reading unit 11 stores in the image data storage unit 101 the item details obtained by character recognition and the data of the entire image of the receipt.

  In step S12, the image data analysis unit 12 specifies the position of the rectangular area of the character data (character string) extracted by the character recognition process from the image data of the payment receipt stored in the image data storage unit 101, and The character string is divided into an amount field indicating the amount of money and other character items. The amount column is a character string composed of only a specific symbol (for example, “¥”, “,”, etc.) and a number, and the character item is a character string that can be a read item other than the amount item. Further, the image data analysis unit 12 associates the character item with the amount field having a predetermined positional relationship, the positional relationship between the character item and the amount field, information indicating whether the character item is unique in the receipt, and the like. The read information that is included is created and stored in the read information storage unit 102.

  In step S <b> 13, the read item calculation unit 13 refers to the item dictionary stored in the item dictionary storage unit 103 and stores the read information stored in the read information storage unit 102 in the basic data storage unit 104. The information of the character item read from the receipt corresponding to the sales management item is extracted. Then, the definition input screen display unit 14 displays the image data storage unit 101, the basic data storage unit 104, the store master 105 data, the character item information extracted as the reading item, and the receipt indicating the position of the reading item and the amount column. A definition input screen including the entire image is created and displayed on the display device 3.

  In step S14, the input data processing unit 15 sets an arithmetic expression or the like for the sales management item of the payment receipt and determines the correspondence between the read item and the sales management item, etc., by inputting from the definition input screen.

  In step S 15, the definition data creation unit 16 creates receipt definition data based on the information input by the input data processing unit 15 and the information set by the reading item calculation unit 13, and stores it in the definition data storage unit 106. To do.

  In step S <b> 16, the history data creation unit 18 creates history data for the created receipt definition data and stores it in the history data storage unit 107.

  In step S <b> 17, the definition data analysis unit 17 sets the name of the read item included in the created receipt definition data to the read character of the setting item (basic management item) corresponding to the item dictionary stored in the item dictionary storage unit 103. Add to column.

  The rough processing executed by the receipt definition data creation device 1 has been described above. Specific processing contents and data structure will be described in detail below.

  FIG. 3 is a diagram illustrating an example of the receipt 2 read by the reader 10. When the reader 10 reads the receipt 2 as shown in FIG. 3, character recognition is performed by analyzing the image by the OCR processing of the reader 10 or the image data reading unit 11. In this character recognition, each recognized character candidate constituting the character string and its position for each series of character strings, such as character strings # 1, # 2,..., # 14,. Information is extracted as image data and stored in the image data storage unit 101.

  FIG. 4 is a diagram illustrating an example of image data stored in the image data storage unit 101. As shown in FIG. 4A, the image data storage unit 101 stores data relating to the image data of the receipt image, and data of the character recognition result of the receipt (hereinafter, recognized character data) as shown in FIG. Are stored.

  As the image data of the receipt image, for example, the image image of the rectangular area of the entire receipt or the entire portion of the receipt printed with characters is stored as a bitmap image file, and the store that identifies the store that issued this receipt 2 A code is given. This store code may be input in advance from the input device 4, or a two-dimensional identification for each store given to a receipt sandwiching sheet (carrier sheet) when the receipt 2 is read by the reader 10. The code or the like may be analyzed and input.

  In the recognized character data shown in FIG. 4B, the result of character recognition for a series of character strings #n (n = 1, 2,...) Is stored, the contents of which are character string # 14 shown in FIG. An example will be described. First, the character string # 14 is the 14th character string in the receipt 2, and the identification number indicates it. The number of characters in the character string # 14 is 3 (the first candidate is “total sales”), and the coordinates of the character string # 14 are (x, y) − (x ′, y ′). This coordinate is a coordinate whose origin is the predetermined reference position of the image of the receipt 2 (for example, the upper left corner of the receipt image or the upper left corner of the rectangular area surrounding the character string group).

  Further, the recognized character data has coordinate information for each of the three characters # 141, # 142, and # 143 constituting the character string # 14. In addition, some characters of the recognition result are stored together with the degree of difference in ascending order of the degree of difference between the recognition result character and the recognition dictionary. For example, the first candidate of the recognition result of the character # 141 is “total” (difference = 1311), the second candidate is “brittle” (difference = 1581), and the third candidate is “聡 ”(difference = 1640).

  Here, the image data reading unit 11 does not store all of these recognition result candidates in the image data storage unit 101, but each character combination of the recognition candidates constituting each character string #n and all stores. This is a recognition result word dictionary prepared in advance in common, and by narrowing down the recognition result character string by collating with words in the dictionary where the character string that may be printed on the receipt is stored in advance, Only the narrowed result may be stored in the image data storage unit 101.

  FIG. 5 is a diagram illustrating an example of read information stored in the read information storage unit 102. As shown in FIG. 5A, the reading information has a character item, a reading position, a relative position in the amount column, and a unique data item. The “character item” is a character string other than the money amount field that has been character-recognized from the payment receipt. The image data analysis unit 12 determines a character string composed of only predetermined symbols and numbers such as “¥”, “,”, and the like from the character string recognized from the payment receipt as an amount column, A character string other than a field is a character item. All character strings determined as character items are recorded in the “character item” of the read information.

  “Reading position” is the position (coordinate information) on the image data of the rectangular area of the character item. As shown in FIG. 5 (B), the image data analysis unit 12 specifies a rectangular area of each character string recognized from the image data of the payment receipt stored in the image data storage unit 101, and on the receipt, The position (coordinates) at which the character string is read is calculated. The coordinate information of each character item is recorded in the “read position” of the read information.

  “Relative position of money field” is information (relative coordinate information) indicating the position of the rectangular area of the money field corresponding to the character item as a relative position from the rectangular area of the character item. As shown in FIG. 5C, the image data analysis unit 12 associates the amount of money field having the Y coordinate value that approximates the Y coordinate value of the character item from the coordinate information of the character item and the amount of money field, The relative position from the rectangular area of the character item to the rectangular area of the associated money amount column is calculated. The relative position of the amount field associated with the character item is recorded in the “relative position of the amount field” of the read information. If the amount field is not associated with the character item, the position information is not set in the “relative position of the amount field”.

  “Unique” is information indicating whether or not the character item is unique in the checkout receipt. The image data analysis unit 12 checks whether there is a character item with the same name in the image data. If the character item is unique, a unique value (circle) is set for the reading information “unique”. If there are multiple character items, it indicates that there is more than one in the reading information “unique”. A value (x mark) is set.

  FIG. 6 is a diagram illustrating an example of an item dictionary stored in the item dictionary storage unit 103. The item dictionary is a dictionary in which “setting items (sales management items)” indicating items set as basic management items and “reading character strings” indicating the names of character items associated with the setting items are associated with each other. It is data. The item dictionary is stored in the item dictionary storage unit 103 in advance.

  The definition data analysis unit 17 extracts the name of the read item associated with the sales management item aggregated by the tenant management department from the receipt definition data, and stores it in the read character string of the corresponding basic management item in the item dictionary storage unit 103. sign up. Note that the name of the setting item, which is a basic management item, and the read character string may be registered by an input operation by the administrator.

  FIG. 7 is a diagram illustrating an example of a basic table stored in the basic data storage unit 104. This basic table stores information on sales management items that are aggregated by the tenant management department. Specifically, in the case of managing sales management data in a shopping center, information on a shopping center code for identifying the shopping center, the name and location of the shopping center, and the number of stores in the shopping center is stored. In addition, the item name of each sales management item necessary for tenant management is stored.

  In this example, there are eight items as sales management items, and the names of these items are “cash sales”, “payment sales”, “credit sales”, “gift certificate”, “total sales”, “consumption tax” ”,“ Other deductions ”and“ Net sales ”. This system enables receipt definition data for automatically extracting information on these sales management items from various receipt items that differ for each store to be created with a simple operation. It is an object.

  In this basic table, a lower limit value and an upper limit value that can be taken by the calculation management item are determined and stored for each calculation management item. The lower limit value and the upper limit value are used for checking to reduce erroneous recognition and erroneous input when the amount of the sales management item is automatically extracted from the receipt by character recognition of the receipt image.

  FIG. 8 is a diagram illustrating an example of the store master 105. The store master 105 is a table that stores information on individual stores that issue receipts, and is prepared in advance. The store code which is the first setting item is a number for identifying the store. The second store name is the name of the store. The third business type code is a code representing the business type of the store. The fourth industry name is the name of the industry. Subsequently, as the POS system information 1, 2 and 3, the identification code of the installed POS system (example: F001), the manufacturer of the POS system (example: AAAAA), the date of introduction of the POS system (example: 2000) Such as January 1st) is stored. Further, as the payment receipt information 1, 2, 3 and 4, the length of the payment receipt, the reading format, the number of print items in the payment receipt, and other necessary information are sequentially stored. In the column of the reading format, “0” is read when the reader 10 reads the receipt 2 as it is, and “1” is read when the receipt 2 is cut into a plurality of pieces because the receipt 2 is too long. Value is set.

  This system creates receipt definition data based on the data stored in the read information storage unit 102, basic data storage unit 104, and store master 105 described above. The store master 105 is not essential when store information is known or input in advance from the input device 4.

  FIG. 9 shows an example of the most basic data as receipt definition data. The store code is a store identification code indicating which store has the receipt definition data. Subsequently, receipt item information read from the receipt is stored for each item. The reading item 1 identification code is an identification code of the first reading item 1, and is a code for uniquely identifying the reading item in the receipt definition data.

  In the item attribute, the attribute of the read item is set. If there is no reading item with the same name (the same reading character) in the receipt, the attribute value = 1. When there is a reading item with the same name in the receipt, the attribute value = 2 when the reading item is specified based on the marker item. When there is a reading item with the same name in the receipt, when specifying the reading item based on a plurality of keywords, attribute value = 3. When it is difficult to recognize the read character and an alternative item is set, attribute value = 4.

  The receipt definition data in FIG. 9A is an example when there are not a plurality of read items with the same name, and the attribute value of the item attribute is 1. The read character 1 of the first read item is “total sales” in this example. The relative reading position 1 in the amount column indicates in which position the character string of the amount corresponding to “total sales” is printed in a relative position from the position of the character string “total sales”. Here, the relative position is, for example, the coordinates of the position of the amount column when the upper left corner of the character string area of “total sales” is the origin, and the character item “total sales” stored in the read information storage unit 102. Are the coordinates of the “relative position of the amount column” corresponding to. The position of the amount column is indicated by the relative position from the position of the read character 1 because the number of lines actually printed on the receipt differs depending on the day even if the receipt is the same, and the position of the read character 1 is variable. It is.

  Similarly, for the reading character 2 of the second reading item “other amount of cash voucher” and the reading character 3 of the third reading item 3 “cash sales”, the identification code, the item attribute, the reading character, the data of the relative reading position of the amount column Is stored as receipt definition data.

  The receipt definition data in FIG. 9B is an example when it is difficult to recognize the characters of the read item, and the attribute value of the item attribute is 4. In this case, an item that replaces recognition of a reading item (for example, consumption tax) is defined as an alternative item (for example, a subtotal), and a relative distance from the alternative item is set as a relative reading position in the amount column.

  FIG. 10 shows an example of receipt definition data when a plurality of identical names exist as receipt items in a receipt. 10A shows an example when the attribute value of the item attribute is 2, and FIG. 10B shows an example when the attribute value of the item attribute is 3.

  When there are multiple items with the same read item name in the receipt (for example, the item name of the read item is "consumption tax" and the term "consumption tax" is printed on multiple lines) Is required to specify whether to extract the amount of money as a sales management item.

  If the attribute value of the item attribute is 2, define one of the items with the same name as the actual reading item by defining another unique item name that serves as a reference for specifying the position of the reading item as a marker item To do. In the example of FIG. 10A, “total sales” is the read character 1, but there are a plurality of item names of “total sales” in the receipt. The item of “Holding” is defined as a marker item. In addition to the reading position of the reading character 1 and the relative reading position 1 of the amount column corresponding to the item, information on the distance (including direction) from the marker item is defined. The distance information from the marker item indicates the upper limit value of the distance (x coordinate, y coordinate) from the character position of the marker item “loan amount” to the character position of the read character 1 “total sales”. As for the coordinates, when the read character 1 is present in the right direction, a positive value is assumed. The y-coordinate takes a positive value when the read character 1 exists in the upward direction, and takes a negative value when the read character 1 exists in the downward direction. In the example of FIG. 10A, this value is “(−50, +100)”, −50 (left direction) in the x coordinate from the character position of the marker item “loan amount”, and +100 in the y coordinate. This indicates that the read character 1 “total sales” to be collected exists within the range (upward).

  When the attribute value of an item attribute is 3, by defining multiple item names that are consecutively arranged in multiple lines as keywords and defining the multiple keywords to be unique in the receipt, the items with the same name One is specified as an actual reading item. In the example of FIG. 10B, “total sales” is the read character 1 necessary as a sales management item, but since there are a plurality of item names of “total sales” in the receipt, it is unique in the receipt. A series of items covering multiple lines, such as “total sales”, “net sales”, “cash amount”, and “loan amount”, are set as multiple keywords. The position of is specified. The reading position 1 of the reading character 1 is the upper left and lower right coordinates of the minimum rectangular area on which a plurality of keywords “total sales”, “net sales”, “cash amount”, and “loan amount” are printed. The relative reading position 1 in the money amount column is a coordinate defined with the upper left coordinate of the reading position 1 of the reading character 1 as the origin.

  Further, as the receipt definition data, the definition information of the arithmetic expression indicating the correspondence between the read item defined by the receipt definition data shown in FIGS. 9 and 10 and the sales management item in the basic table shown in FIG. 7 is set. Is done. FIG. 11 shows an example of receipt definition data that defines the arithmetic expression.

  As shown in FIG. 11, the definition information of the arithmetic expression has correspondence information between each sales management item and one or a plurality of read item identification codes in addition to the store code for identifying the store. When a plurality of reading items are associated with one sales management item, arithmetic expressions of four arithmetic operations for a plurality of reading item identification codes are used. In the example of FIG. 11, the read item identification code corresponding to the sales management item 1 “cash sales” is “A + B”. This is a value obtained by adding the numerical value printed in the amount column corresponding to the reading item with the reading item identification code = A and the numerical value printed in the amount column corresponding to the reading item with the reading item identification code = B, It shows that it is recorded as the amount of sales management item 1 “cash sales”. For the second sales management item 2 “paid sales”, the read item of the read item identification code “C” is used as the amount of “paid sales” as it is. Similarly for other sales management items, the correspondence information between each sales management item and the read item from the receipt is defined by the operation formula of the read item identification code (including the case of one-to-one correspondence).

  Next, an example of history data created by the history data creation unit 18 and stored in the history data storage unit 107 will be described with reference to FIG.

  The history data is collected so that the contents of the receipt definition data created by the definition data creation unit 16 can be used for the creation of receipt definition data at other stores in the same industry. As the industry information, the industry code, industry name, POS system information 1 to 3, payment receipt information 1 to 4 and the like are stored. In addition, as the item information, an identification code, an item attribute, a reading character, and the like of each reading item are stored. In addition, as the calculation information, an arithmetic expression using each sales management item and the corresponding read item identification code is stored. These specific contents are the same as the contents of the receipt definition data described with reference to FIGS.

  Further, the history data creation unit 18 creates dictionary information as history data and stores it in the history data storage unit 107. FIG. 13 shows an example of the dictionary portion of the history data.

  In the dictionary section of the history data, as shown in FIG. 13, character string information (words etc.) actually used for receipt definition data such as industry codes, POS system information, sales management items, read characters,. Is done. The character string information stored in the dictionary part is used to infer the reading items corresponding to the sales management items when the receipt definition data of a new store is created by using the created receipt definition data. For example, it can be used to extract a character string of a correct recognition result in the image data of the character recognition result described with reference to FIG. It can also be used to prevent erroneous recognition and erroneous input automatically in the actual sales management data collection operation.

  Next, the process executed by the receipt definition data creation device 1 will be described in more detail. 14 to 16 are flowcharts of the entire receipt definition data creation process according to the present embodiment.

  First, in step S201, an operator input from the input device 4 is waited. If the operator designates whether receipt definition data is newly created or created by referring to the created receipt definition data by using a menu or the like, the process proceeds to step S202.

  In step S202, it is determined whether the receipt definition data is newly created or reference created. If reference creation is to be performed, the process proceeds to step 220 in FIG. 15, and if newly created, the process proceeds to step S203.

  In step S203, an initial screen of a definition input screen for creating receipt definition data is displayed on the display device 3.

  FIG. 17 shows a configuration example of the definition input screen. The definition input screen 30 includes a sales management item list display area 31, an OCR reading definition list display area 32, a receipt image display area 33, a marker below the display area of the facility for creating receipt definition data and tenant (store) information. A definition list display area 34 and a calculation formula input area 35 are provided.

  FIG. 18 is a diagram showing a display example of the sales management item list display area 31. In the sales management item list display area 31, a sales management item list is displayed. The sales management item list is calculated when an item name of the sales management item stored in the basic table described in FIG. 7 and a calculation formula (calculation formula) that associates the sales management item with the read item are determined. Has a field that displays an expression.

  FIG. 19 is a diagram showing a display example of the OCR reading definition list display area 32. In the OCR reading definition list display area 32, the reading item extracted by the reading item calculation unit 13, information (item area) indicating that the position (area information) of the reading item is set, and an amount column corresponding to the reading item Information (amount area) indicating that the position (area information) is set is displayed.

  In the receipt image display area 33, an image (image data) of a receipt read by the reader 10 is displayed. In the display of this area, functions such as scrolling and enlargement / reduction are provided.

  The marker definition list display area 34 displays information for defining a unique marker item for specifying a reading item when there are a plurality of reading items with the same name in the receipt. In the calculation formula input area 35, a calculation formula for associating the input sales management item and the read item is displayed.

  In step S203, the initial screen of the definition input screen 30 as described above is displayed, and in step S204, an operator input is waited for. In step S205, when all the items have been input, the operator presses the “define definition” key. Therefore, it is determined whether or not the input has been made, and if the “define definition” instruction is input, step S208 is performed. Proceed to

  If there is no “define definition” instruction, the process advances to step S206 to change the read item.

  On the initial screen of the definition input screen 30, the OCR reading definition list display area 32 displays the reading item ID (for example, a) automatically given by the reading item calculation unit 13 and the reading information, as shown in FIG. The reading item, item area, and monetary area extracted from are displayed.

  Assume that a receipt image as shown in FIG. 20A is displayed in the receipt image display area 33.

  Here, it is assumed that the operator wants to change “total sales” of the reading items displayed in the OCR reading definition list to “net sales”. In this case, the operator points and drags or clicks on the character string portion of “net sales” with the input device 4 such as a mouse on the image in the receipt image display area 33. The read item acquisition unit 151 of the input data processing unit 15 detects this input, and displays the character string of “net sales” in the receipt image surrounded by a frame line. At this time, for example, as shown by the arrow in the left diagram of FIG. 20B, even if the character string portion of “net sales” is largely dragged, the right side of FIG. A frame line is displayed with a minimum necessary size as shown by a dotted line frame in the figure. Note that the automatic adjustment of the size of the area surrounded by the frame line can be performed by referring to the reading information shown in FIG. The position information of the area of the character item selected in this way is extracted from the read information and stored in the internal memory.

  The definition input screen display unit 14 reads the information related to the character item selected by the operator's input operation from the reading information storage unit 102 via the reading item calculation unit 13 as the changed reading item. Set to list.

  Subsequent to the reading item definition processing described above, in step S207, calculation definition processing is performed. The calculation definition process is a process for defining a calculation formula indicating the correspondence between the sales management item and one or more reading items defined in step S206. Details will be described later. Then, it returns to step S204 and repeats a process similarly.

  If there is an instruction for “definition of definition”, the process proceeds to step S208. In step S208, the reading item is rechecked. Here, character recognition is performed again for the item area of the read item. Alternatively, the recognition accuracy is determined based on the difference information for each recognized character of the read item in the recognized character data stored in the image data storage unit 101 described with reference to FIG. It is checked whether or not the recognition accuracy is worse than a predetermined threshold.

  If it is determined in step S209 that there is no recognition error, or if it is determined that the recognition accuracy is not bad, the process proceeds to step S230 in FIG. If not, the process proceeds to step S210 to prompt the operator to input a substitute item. Because the substitution item is difficult to recognize or has a high possibility of misrecognition of the original reading item, other receipt items are to be recognized as substitution items, and the original item is determined based on the relative position information from the substitution item. This is an item for the purpose of acquiring the amount of money for the read item. If there is an input operation of the substitute item from the operator, the substitute item definition process is performed in step S211. Thereafter, the process returns to step S204. Specific examples of alternative items will be described later with reference to the drawings.

  If reference creation of receipt definition data is specified in step S202, the process proceeds to step S220 in FIG. 15, and an instruction is input to select the receipt definition data to be referred to manually or automatically. When selecting manually, it progresses to step S221. In step S221, information specifying a store code of a store for which receipt definition data has already been created is input, and receipt definition data corresponding to the store code is acquired from the definition data storage unit 106, and the process proceeds to step S226.

  When the receipt definition data to be referred to is automatically selected, the process proceeds to step S222, and the history data stored in the history data storage unit 107 is referred to. Next, in step S223, the presence / absence of similar data is determined. The process for determining the presence or absence of similar data will be described later with reference to the drawings.

  If it is determined in step S224 that there is no data similar to the receipt definition data of the store to be created in the history data, reference creation is not possible, so the process returns to step S203, and processing similar to that for new creation is performed. Execute.

  If it is determined that there is similar data in the history data, the process proceeds to step S225, and receipt definition data of the similar store is acquired.

  In step S226, the read item of the acquired store is inferred from the receipt definition data acquired in step S221 or step S225, and the definition input screen 30 is edited. That is, as an initial screen of the definition input screen 30, a screen with an empty OCR reading definition list is not displayed, but reading item data defined in the receipt definition data to be referred to and sales management in the sales management item list are displayed. Create and display a screen with the item formula embedded. Thereafter, the process proceeds to step S204. The processing after step S204 is the same as the processing described above. As a result, in the case of reference creation, the operator can easily create receipt definition data for a new store by simply modifying a portion different from the receipt definition data for the store to be referenced as necessary.

  If it is determined in step S209 described above that there is no reading item recognition error or the recognition accuracy is not bad, the process proceeds to step S230 in FIG. In step S230, the receipt item defined in step S206 is checked for an item of the same character string in the receipt. If it is determined in step 231 that there is an identical character string, the process proceeds to step S233. If it is determined that there is no identical character string, the process proceeds to step S232.

  In step S232, after the history data is created, the process ends. Details of the history data creation processing will be described later.

  If there is an identical character string, a marker definition process for specifying a character item to be read as a sales management item reading item is performed in step S233. Here, a unique character item is set as a marker item in the payment receipt, which will be described in detail later. After the marker definition process, the process returns to step S204 and continues in the same manner.

  FIG. 21 is a flowchart showing details of the initial screen display process (step S203 in FIG. 14). The following processing is executed by the read item calculation unit 13 and the definition input screen display unit 14.

  In step S301, the read item calculation unit 13 refers to the item dictionary stored in the item dictionary storage unit 103, and reads the read character of the sales management item in the item dictionary from the read information stored in the read information storage unit 102. Extract the character items that correspond to the column. Further, as the information of each extracted character item, the position of the character item and the positional relationship (relative position) to the corresponding money amount column are acquired.

  In step S302, the definition input screen display unit 14 sets the character item extracted as a reading item as a reading item in the OCR reading definition list. Further, information indicating that the position (area information) is set in the payment receipt is displayed in the item area and the money area of each character item.

  In step S303, it is determined whether or not there are a plurality of items with the same name in the character items extracted from the read information. If there are a plurality of items with the same name, the process proceeds to step S304.

  In step S304, an alarm indicating that marker definition processing is necessary is displayed in the reading items of a plurality of character items in the OCR reading definition list.

  In step S305, when it is determined from the read information that there are a plurality of character items having the same name, a character item that is unique in the checkout receipt is extracted. The extracted character item information is set as a candidate item in the marker item definition list, and a selection radio button for each candidate item is displayed.

  FIG. 22 is a diagram illustrating a display example of the OCR reading definition list display area 32 and the marker definition list display area 34 when there are a plurality of character items having the same name.

  In the OCR reading definition list display area 32 shown in FIG. 22A, there are displayed a plurality of character items “consumption tax” extracted as reading items and an indication that there are duplicate items (“duplicate”). The

  In the marker definition list display area 34 shown in FIG. 22 (B), the read item calculation unit 13 sets character items that are candidates for marker items extracted from the read information and item areas on the checkout receipt. Information (item area) and a radio button for selection. The display example shown in FIG. 22 indicates that the character item “return details” has been selected as the marker item.

  FIG. 23 is a flowchart showing details of the calculation definition process (step S207 in FIG. 14). The following processing is executed by the calculation definition unit 152 in the input data processing unit 15.

  In step S351, an input for the calculation definition is awaited. If there is an input, the process proceeds to step S352, and based on the input, a sales management item that defines the calculation and a reading item to be used for the calculation are selected. A reading item can be selected by specifying a reading item ID. One or a plurality of reading items may be selected.

  In step S353, an arithmetic operation operator is input, and the operator is selected. Arithmetic symbols such as% and parenthesis symbols that specify the order of operations can also be used. The number of operators to enter depends on the number of selected reading items. In the case of one reading item, it is not necessary to input an operator.

  Until the selection of all necessary reading items and operators is completed by the determination process in step S354, the process returns to step S351 and the same process is repeated.

If the selection of the reading item and the operator is completed, the input operator is checked in step S355. This is the following check.
1) Is an undefined reading item ID used?
2) Is an operator that cannot be used specified?
3) Whether the result is within the allowable range when the amount is calculated using the set formula.

  Whether or not the result is within the allowable range is calculated from the lower limit value of the corresponding sales management item stored in the basic table when calculated using the numerical value of the amount actually obtained from the image data of the receipt. Judgment is made based on whether it is within the range up to the upper limit.

  If an error is detected as a result of the check in step S356, the process returns to step S351 to reselect the reading item or re-enter the operator. If there is no error, the input calculation formula is displayed in the calculation formula column of the corresponding sales management item in the sales management item list, and the calculation definition for one sales management item is terminated. Calculation definition is basically performed for all sales management items, but may be undefined for sales management items for which no corresponding read item exists.

  FIG. 24 shows an example of calculation definition. As shown in FIG. 24A, in the OCR reading definition list display area 32, net sales (ID = a), consumption tax (ID = b), and cash voucher (ID = c) of reading items are defined. Suppose. Here, if the sales management items “total sales” and “gift certificate” have a corresponding relationship with the read items as shown in FIG. 24B, as shown in FIG. Define a calculation formula of “a + b” for “total sales” and a calculation formula of “c” for “gift certificate”. By setting this calculation formula as the operation definition in the receipt definition data, the amount of the sales management item can be calculated from the actual receipt issued by the store.

  FIG. 25 shows an example in which the correspondence between the receipt of a certain store and the sales management item is defined by a calculation formula. Receipt reading items generally differ from store to store, so the correspondence between sales management items required by the tenant management department and receipt reading items at each store differs from store to store. However, as shown in FIG. 25, since a calculation formula can be defined for each store, it is possible to flexibly associate sales management items with receipt items.

  FIG. 26 is a flowchart showing details of the marker definition process (step S233 in FIG. 16). The following processing is executed by the multiple keyword setting unit 153 and the marker item setting unit 154 in the input data processing unit 15.

  In step S401, an alarm indicating that there are a plurality of reading items defined in the OCR reading definition list display area 32 having the same name is displayed, and a marker definition setting input from the operator is waited.

  In step S402, the type of the marker item is determined in response to the operator input. That is, it is determined whether a reading item in a plurality of character items with the same name is specified by one unique character item (marker item) or specified by a plurality of character items (multiple keywords). If a marker item is selected, the process proceeds to step S403, and if a plurality of keywords are selected, the process proceeds to step S406.

  In step S403, a marker item designation input is awaited. The character item selected by the radio button for selecting each candidate item displayed in the marker definition list is designated as the marker item. In step S404, the position of the character item that is the designated candidate item is acquired. Subsequently, in step S405, based on the position of the reading item and the position of the designated candidate item, among the reading items of the same character string, the marker item in which the reading item used in the definition of the sales management item is specified. Calculate the direction indicating whether it is up or down and the distance from the marker item. If the reading item is above the marker item, the distance is a positive value. If the reading item is down, the distance is negative. Define with the value of. A distance corresponding to a predetermined number of lines may be added to the distance value as a margin in consideration of fluctuations in the line position of the reading item in a receipt that is actually issued every day.

  FIG. 27 shows a definition example of marker items. In the receipt image displayed in the receipt image display area 33 shown in FIG. 27A, it is assumed that the reading item actually required as a sales management item is “consumption tax”. Here, there are a plurality of items with the same name “consumption tax”, and it is assumed that “consumption tax” above is an item to be defined as a read item. Therefore, it is assumed that another character item unique in the receipt (in this example, “return item”) is designated as the marker item.

  That is, as shown in FIG. 27B, it is assumed that the marker item is “return details” and the distance to “consumption tax” is “110” in the right direction and “100” in the upward direction. At this time, (110, 100) is defined as the distance from the marker item “details of return” to the read item “consumption tax”.

  The number of marker items is not limited to one, and a plurality of marker items may be defined. In the example of FIG. 27C, in addition to defining “returned goods details” as a marker item, marker items are also defined for “sales details”. In this case, only the read items that satisfy the conditions defined by the definition of all marker items are extracted as the corresponding read items. If the read item does not satisfy any of the marker item definition conditions, Will be certified.

  If it is determined in step S402 in FIG. 26 that a reading item is specified by a plurality of keywords, the process proceeds to step S406, and input for selecting a plurality of keywords is waited. If there is an input for selecting a plurality of keywords, in step S407, the character items to be used as a plurality of keywords and the coordinates of the amount of money are acquired and defined as a plurality of keywords.

  FIG. 28 shows an example of setting a plurality of keywords. In the example of FIG. 28, there are three character items having the same name as the read item “total sales”, Str # 12, Str # 22, and Str # 32. In addition, there are three character strings corresponding to each of Dat # 12 “¥ 168,735”, Dat # 22 “¥ 48,195”, and Dat # 32 “¥ 32,025”.

  If the read item required here is “total sales” of Str # 12, the read item of Str # 12 cannot be specified only by the character string of “total sales”. Therefore, a plurality of character items including Str # 12 are included. Is defined as a plurality of keywords Key # 1. In this example, “total sales”, “net sales”, “cash balance”, and “lending balance” are defined as a plurality of keywords Key # 1. In this receipt definition data, “3” indicating a plurality of keywords is defined as item attribute # 1, and “total sales”, “net sales”, “cash balance”, and “loan amount” are defined as read character # 1. To do. The relative reading position # 1 of the amount column of the character string Dat # 12 necessary for collecting the sales management item is also relative to the relative reading position # 1 of the amount column of the character string Dat # 12 acquired by the reading item calculation unit 13. Is defined in the receipt definition data. As a result, even if there are a plurality of character strings of “total sales” in the receipt, if multiple keywords are unique, the reading item of “total sales” included in the character string matching the plurality of keywords is specified. It will be.

  FIG. 29 is a diagram illustrating an example of setting alternative items. Processing performed by the alternative item processing unit 155 in steps S210 and S211 of FIG. 14 described above will be described.

  Now, as shown in FIG. 29A, it is assumed that the consumption tax is defined in the OCR reading definition list display area 32 and the coordinate information is also acquired for the item area. Here, for the original image of the receipt that is the display source of the receipt image display area 33 shown in FIG. 29A, this portion is again recognized by OCR processing. If it is actually recognized as “consumption tax” entered by the operator, the substitute item is not processed because it is correct. If there is a recognition error due to recognition of a different character, as shown in FIG. 29 (B), a warning prompting the user to enter an alternative item “This is a hard-to-read character string. Set an alternative item.” Output a message. Instead of actually performing the OCR process, the recognition accuracy may be estimated from the recognized character data shown in FIG.

  When the warning message as shown in FIG. 29B is output, the operator designates an easy-to-read receipt item such as “subtotal” in the receipt image display area 33 in FIG. 29C as an alternative item. To do. The substitute item may be either above or below the original reading item (consumption tax), but it is desirable that the line is close. As in this example, when “subtotal” is specified as an alternative item, the information on the relative reading position of the amount “¥ 7,234” corresponding to the consumption tax is included in the receipt definition data from “consumption tax”. Relative position from “subtotal” that is easy to read is defined, not relative position.

  Further, in order to show that it is defined by the substitute item, in the OCR reading definition list display area 32, as shown in FIG. 29D, the substitute item “subtotal” and the original reading item “consumption tax” are displayed. They are displayed concatenated with a predetermined separator “!”. Also in the receipt definition data, as shown in FIG. 9B, the read character is defined as “subtotal! Consumption tax”, which indicates that it is defined by an alternative item.

  FIG. 30 is a flowchart showing details of the process for determining the presence or absence of similar data (step S223 in FIG. 15).

  When creating receipt definition data for a new store, whether or not there is similar data in order to automatically extract similar receipt definition data from history data when referring to already created receipt definition data Judging. First, in step S501, the store master 105 is referred to. In step S502, the history data stored in the history data storage unit 107 is referred to.

  As a result of referring to steps S501 and S502, the business type code of the new store that newly creates receipt definition data is the same as the business type code in the history data, and the POS system information in the new store and the POS system information in the history data Are determined in steps S503 and S504. As a result of the determination, if both are equal, the process proceeds to step S506. If neither is equal, the process proceeds to step S505, and the determination result is “no similarity”.

  On the other hand, in step S506, it is determined that “similarity exists”, and in step S507, history data (dictionary part) is referred to. In step S508, the reading item corresponding to the sales management item is inferred from the result of referring to the history data (dictionary part).

  FIG. 31 is a flowchart showing details of the history data creation process (step S232 in FIG. 16).

  In step S601, the store master 105 is referred to from the store code of the store where the receipt definition data is newly created. In step S602, the corresponding business type code, POS system information, and payment receipt information are acquired from the store master 105. In step S603, item information and calculation information are acquired from the created receipt definition data. In step S604, history data and a history data dictionary as shown in FIGS. 12 and 13 are created and stored in the history data storage unit 107.

  The processing for creating the receipt definition data described above can be realized by a computer and a software program, and the program can be recorded on a computer-readable recording medium or provided through a network.

DESCRIPTION OF SYMBOLS 1 Receipt definition data creation apparatus 10 Reader 11 Image data reading part 12 Image data analysis part 13 Reading item calculation part 14 Definition input screen display part 15 Input data processing part 151 Reading item acquisition part 152 Calculation definition part 153 Multiple keyword setting part 154 Marker item setting unit 155 Alternative item processing unit 16 Definition data creation unit 17 Definition data analysis unit 18 History data creation unit 101 Image data storage unit 102 Reading information storage unit 103 Item dictionary storage unit 104 Basic data storage unit 105 Store master 106 Definition data Storage unit 107 History data storage unit 2 Receipt 3 Display device 4 Input device

Claims (6)

Receipt definition for creating receipt definition data used for receipt image recognition processing for each store executed in a receipt recognition system that collects basic management data consisting of specific basic management items from receipt images issued by each of a plurality of stores A data creation device,
Basic data storage means for storing information on basic management items;
Item dictionary storage means for storing an item dictionary indicating names of items read from the receipt corresponding to the basic management items;
Image data storage means for storing receipt image data;
Character recognition processing is performed on the image data, position information of the recognized character string on the image data is acquired, and the recognized character string is a money amount column for displaying a monetary amount or a character other than the monetary amount column Determining whether the item is an item, associating the character item with an amount field having a preset positional relationship, the character item, the associated amount field, and the character item and the associated amount field; Image data analysis means for creating read information including the positional relationship of
Reading item calculation means for referring to the item dictionary and extracting character item information that matches the name of the item corresponding to the basic management item from the reading information;
As the receipt definition data information, the basic management item information, the reading item read from the image data for the basic management item, and the definition input screen for displaying the image data are generated, and the character item information extracted from the reading information And a definition input screen display means for displaying the definition input screen in which the information of the read item is set. In the receipt definition data creation device according to claim 1,
When the reading item calculation unit extracts, from the reading information, information on a plurality of character items that matches the name of the item corresponding to the basic management item, a unique character item in the image data is extracted from the reading information. And
The definition input screen display means displays the definition input screen in which the information of the unique character item is set to the information of the candidate item of the marker item that identifies the reading item. . In the receipt definition data creation device according to claim 1 or 2,
Definition data for extracting the character item set in the reading item of the basic management item from the receipt definition data and registering the name of the extracted character item in the item name corresponding to the basic management item of the item dictionary A receipt definition data creation device comprising an analysis means. To create receipt definition data used for receipt image recognition processing for each store executed in a receipt recognition system that collects basic management data consisting of specific basic management items from receipt images issued by a plurality of stores , A computer with storage means,
Character recognition processing is performed on the receipt image data stored in the storage means, and position information of the recognized character string on the image data is obtained,
It is determined whether the recognized character string is an amount field for displaying an amount or a character item other than the amount field, the character item is associated with an amount field having a preset positional relationship, and the character item , Creating the read information including the positional relationship between the associated amount field, and the character item and the associated amount field,
Refers to an item dictionary stored in the storage means that indicates the name of an item read from the receipt corresponding to the basic management item, and matches the name of the item corresponding to the basic management item from the read information. Extract information about character items,
As the receipt definition data information, the basic management item information constituting the basic management data, the reading item to be read from the image data for the basic management item, and the definition input screen for displaying the image data are generated. Display the definition input screen in which the information of the character item extracted from the information of the reading item is set;
Receipt definition data creation program for executing processing. In the receipt definition data creation program according to claim 4,
When extracting information on a plurality of character items that match the name of the item corresponding to the basic management item in the process of extracting information on the character item that matches the name of the item corresponding to the basic management item from the read information In addition, a unique character item in the image data is extracted from the read information,
In the process of displaying the definition input screen, the definition input screen is displayed in which the information of the unique character item is set to the information of the candidate item of the marker item that identifies the read item. Creation program. In the receipt definition data creation program according to claim 4 or 5,
Extracting a character item set in the reading item of the basic management item from the receipt definition data and registering the name of the extracted character item in the item name corresponding to the basic management item of the item dictionary;
A receipt definition data creation program to be executed by the computer.

Images