JP2013196479A - Information processing system, information processing program, and information processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique by which a necessary attribute described in a free-format document and a character string can be efficiently extracted and associated with each other, and the extracted content can be corrected with an easy procedure.SOLUTION: An information processing system according to the embodiment includes a first display part, a first input part, a second input part, and a second display part. The first display part displays an image of a document to which character recognition is performed. The first input part detects a range specified by a user in the image. The second display part displays, in the form of a list, character string, which is a result of the character recognition performed in the document, for each attribute information indicative of what attribute the character string has, and displays the character string, which is a result of the character recognition performed in the range detected by the first input part, in a column in the list specified by the user by means of the second input part.

Description

  Embodiments described herein relate generally to a technique for correcting characters recognized and identified by an optical character reader.

  Optical character recognition (OCR) technology and an optical character reader that realizes character recognition using this OCR technology as a technology for identifying characters on paper and converting them to electronic data Yes, it has been put to practical use.

  In addition to identifying the characters on the business card, the characters were identified based on the size and position of the characters on the business card and keywords such as "Tel:", "Fax:", and "E-mail:". There is also a technique for identifying character strings and attribute information such as company, name, title, contact information, etc., and holding them in association with each other.

  The following documents are disclosed as related technologies.

JP-A-5-151387

  If the format is customarily determined to some extent, such as a business card, and the description content is limited, the above prior art is effective, but a large number of character strings described in a free format document of a large number of pages When it is detected, it becomes difficult to determine the combination of the written character string and the attribute information.

  Further, when there is a keyword as described above in the vicinity of the character string to be extracted, the extraction is easy, but in the case of a free format document, the keyword is not necessarily in the vicinity. When there is omission of extraction, the user specifies and extracts a character string in the document, but the conventional technique does not consider a user interface that is efficiently extracted.

  The problem to be solved by the present invention is a technology that can efficiently extract and associate necessary attributes and character strings described in a free format document, and can correct the extracted contents by an easy procedure. Is to provide.

  The information processing system according to the embodiment includes a first display unit, a first input unit, a second input unit, and a second display unit. The first display unit displays an image of the document on which character recognition has been performed. The first input unit detects a range designated by the user in the image. The second display unit displays a list of character strings, which are the result of character recognition in the document, for each attribute information indicating what attribute the character string is a character string of. A character string that is a result of character recognition in the range detected by the first input unit is displayed in a column designated by the user using the input unit.

  The information processing system of the embodiment includes a digital pen and a computer. The digital pen acquires position information on a sheet on which an image subjected to character recognition is printed. The computer displays a list of character strings written in the image for each attribute information indicating what kind of attribute the character string is, and is specified by the user specified in the sheet image obtained by the digital pen. Matches the specified attribute information in the list based on the position information of the character string and the position information of the entry frame specified by the user and specified in the sheet. The designated character string written in the sheet image is displayed in the blank space of the attribute information column to be displayed.

It is a figure which shows the structural example of 1st Embodiment. It is a flowchart which shows the operation example of the document reading system of embodiment. It is a figure which shows an example of the document image of embodiment, and an example of the area | region after layout analysis. It is a figure which shows the functional block of the correction terminal of 1st Embodiment. It is a figure which shows the example of a screen which the correction terminal of 1st Embodiment displays. It is a figure which shows the example of an operation display with respect to the example of a screen shown in FIG. It is a figure which shows the example in which a character string is displayed on the designation | designated column after operation shown in FIG. 6 is performed. It is a figure which shows the other operation example of 1st Embodiment, and the example of a display update. 5 is a flowchart illustrating an operation example of the correction terminal according to the first embodiment. It is a figure which shows the structural example of the correction system of 2nd Embodiment. It is a figure which shows the example of a display of the document image displayed with the correction system of 2nd Embodiment, and an example of the screen for extraction result / correction. It is a figure which shows the structural example of 3rd Embodiment. It is a block diagram which shows the structural example of the correction system of 3rd Embodiment. It is a figure which shows the example of an entry and display example of 3rd Embodiment.

(First embodiment)
FIG. 1 is a diagram illustrating a hardware configuration example of a document reading system according to the present embodiment. In the following embodiment, for example, a mode when a personnel change notification is received by FAX or mail will be described, but the mode is not limited to this.

  A document reading system 300 according to the first embodiment includes a correction terminal 1 (information processing system), a server system 2, and a network 3. Note that a configuration in which a plurality of correction terminals 1 are provided or a configuration in which a plurality of server systems 2 are provided may be employed.

  The correction terminal 1 is an operation terminal for a user to compare a document and a character recognition result with respect to the result of character recognition, and to correct what needs correction. The correction terminal 1 is a computer having a processor 101 which is an arithmetic processing unit, a memory 102 which is a main storage device, and an auxiliary storage device 103 such as a flash memory or an HDD (hard disk drive). The correction terminal 1 includes a display 105 as an output device, a keyboard 106 and a mouse 107 as input devices, and a communication device 104 that controls data communication with the outside. These units perform command transmission / reception and data transmission / reception with each other via the data bus 108.

  The server system 2 includes a FAX server 2A, an OCR server 2B, and a file server 2C. The servers 2A to 2C are conventional computers each including hardware such as a processor, a memory, and an auxiliary storage device.

  The FAX server 2A includes a communication board that can communicate with the outside via a public line, receives FAX data from the outside, and converts it into digital data. The converted digital data is stored, for example, as an image file in its own storage device or the file server 2C. The file output destination is not limited as long as it is a storable device or medium.

  The OCR server 2B recognizes the area of the character or design drawn in the image file (layout analysis), and identifies the character drawn in the image file for each area based on the conventional OCR technology. (In this example, text data). In the OCR server 2B, a specific keyword is stored in advance in a storage device (hereinafter referred to as a keyword dictionary), and it is obtained by layout analysis whether the identified character string data includes the keyword. Determine for each region. The keyword registered here is attribute information indicating the meaning (category) of the character string to be extracted, for example, “name”, “address”, and “phone number”. . The document reading system 300 manages a character string adjacent to the character string “name” as a category of “name” when, for example, a character string “name” is clearly specified in the document image. Hold.

  The OCR server 2B associates region information (for example, coordinate information with the upper leftmost end point of the image as a reference point) and character string information in the region, and outputs the result as an extraction result file. When the character string registered in the keyword dictionary is hit, the OCR server 2B outputs the keyword as an extraction result file in association with the area information and the character string data. In the present embodiment, regardless of whether or not the keyword is hit, all the area information and character string information are associated with each other, and an extraction result file is output. The output destination may be in its own storage device or the file server 2C. The mode is not limited as long as it is a storable device and medium.

  The file server 2C is a server that permanently stores data that has been confirmed and corrected via the correction terminal 1. The file server 2C may store FAX data digitized by the FAX server 2A, or may store an extraction result file output from the OCR server 2B.

  It is assumed that the FAX server 2A, the OCR server 2B, and the file server 2C in the server system 2 can transmit / receive data to / from each other. The correction terminal 1 transmits / receives data to / from the server system 2 via the communication device 104 and the network 3. In this embodiment, the network 3 is a LAN (Local Area Network), and an IP address (IP: Internet Protocol) is assigned to each device. However, the mode is not limited to this, and the mode is via an Internet network. It may be. The communication means may be wired or wireless.

  Each auxiliary storage device in the servers 2A to 2C is preinstalled with software for controlling the hardware and software for realizing the present embodiment. The server system 2 may have a configuration including an uninterruptible power supply (UPS) as shown in FIG. 1 or may have a configuration including an input keyboard, a mouse, a display for display, and the like. The function that can be realized by the FAX server 2A may be an MFP (Multifunction Peripheral) having a FAX transmission / reception function, a printer function, a scanner, a copy function, and the like. The file server 2C may be a network compatible HDD. In addition to this, various system configurations are conceivable. Depending on the work scale, one computer may be responsible for all the functions of each of the above devices.

  Further, the image file to be processed may be image data obtained by scanning data on a sheet of paper by an MFP or the like, or an electronic document file conforming to a predetermined format, in addition to those received by FAX.

  Next, an operation example of the document reading system 300 is shown in the flowchart of FIG. The operation example of FIG. 2 is an operation example after an image file to be processed is obtained by the FAX server 2A or the like.

  The OCR server 2B acquires an image file (hereinafter referred to as a document image) on which a document to be processed is drawn, and performs layout analysis on the document image (S101). This layout analysis is a process for determining in which area on the image a character or a pattern is drawn and acquiring the coordinates of the area. The layout analysis will be described with reference to FIG. FIG. 3A shows an example of the original document image, and FIG. 3B shows an example of the area after the layout analysis of the document image. A plurality of thick line rectangles shown in FIG. 3B are determined as areas where characters and designs are drawn. By the layout analysis, characters (character strings) and symbols within a certain interval are collected as one area, and each coordinate information is obtained. A known technique may be used for the layout analysis.

  The OCR server 2B identifies the character drawn in the area obtained by the layout analysis and converts it into digital data (S102). The OCR server 2B determines whether the character is a keyword by comparing the identified character string data with the keyword in the keyword dictionary (S103). In the present embodiment, a special character (wild word) can be included in the keyword, and matching can be performed. In this example, it is assumed that the character strings “name”, “previous job”, and “new job” are included as keywords.

  The OCR server 2B acquires a character string that matches the keyword as attribute information (S104), and extracts another character string included in the area having the attribute information (S105). The OCR server 2B also extracts a character string included in another area that is close to the area having the attribute information. The proximity interval can be specified by a parameter. The OCR server 2B associates the area information, the character string in the area, and the attribute information if the attribute information is in the area (or the attribute information if there is attribute information in the vicinity of the area). And output as an extraction result file. The file name of the document image is also output in association with it.

  The correction terminal 1 obtains the extraction result file and supports confirmation and correction work based on the user's operation (S106). The user operation and the operation of the correction terminal 1 will be described later. When the correction work is completed, the correction terminal 1 outputs the data after confirmation / correction to the file server 2C as a CSV file (S107). In the present embodiment, the CSV file of this embodiment is a text file that includes attribute information and character string data (extracted character string data) in the document image associated with the attribute information in one record. To do. Note that the file format is not limited to a CSV file, and may be implemented in a database instead of a file.

  Next, details of the correction terminal 1 of the present embodiment will be described. FIG. 4 is a block diagram illustrating the configuration of the correction terminal 1.

  The correction terminal 1 includes an input unit 11, a designation field acquisition unit 12, a correction screen display unit 13, an area acquisition unit 14, a character information acquisition unit 15, and a document image display unit 16. The auxiliary storage device 103 in the correction terminal 1 is preinstalled with software for controlling hardware and software for realizing the functions of the blocks shown in FIG. Each block shown in FIG. 4 is realized by the processor 101 executing this software, and is realized in cooperation with each hardware shown in FIG. 1, such as the display 105, the keyboard 106, and the mouse 107.

  The operation of each block in FIG. 4 will be described with reference to FIGS. The correction terminal 1 acquires the extraction result file and the document image output from the OCR server 2B, and displays them on the display 105 in the layout shown in FIG. As shown on the left side of FIG. 5, the document image display unit 16 displays a document image and color-codes a character string associated with attribute information according to the correspondence in the extraction result file for each attribute information. Draw with emphasis. In this example, since the information registered in the keyword dictionary is “name”, “previous job”, and “new job”, the document image display unit 16 displays the character strings corresponding to these attribute information in different colors. To do. The document image display unit 16 controls to display a scroll bar and display the next page or the like by a mouse operation when the document images cannot be displayed at once, for example, when there are a plurality of document images.

  As shown on the right side of FIG. 5, the correction screen display unit 13 refers to the extraction result file and displays each attribute information in the uppermost column, and displays a character string corresponding to each attribute information in the lower column. List it for editing. Further, in order to improve the visibility of the user, the correction screen display unit 13 displays the uppermost column in a color-coded manner so as to match the color coding for each attribute information in the document image display unit 16.

  The user compares the character string color-coded for each document image and attribute information displayed on the document image display unit 16 with the extraction result displayed on the correction screen display unit 13 / the display character string on the correction screen. . When there is an erroneous character recognition by the OCR server 2B, the user corrects the recognition error portion using the mouse 107 and the keyboard 106. An extraction result / correction screen is used for correction.

  The processing of the OCR server 2B includes the above-described layout analysis processing and processing for associating attribute information with a character string in addition to character identification. About the character string with which attribute information was matched, the character string is displayed on the display list of the correction screen display part 13 (right side of FIG. 5). However, the character strings of “Jiro Toshiba”, “Account Manager”, “Director”, “Saburo Toshiba”, “Director”, and “Advisor” on the document image on the left side of FIG. 5 are in free format. Therefore, since the attribute information of “name”, “previous job”, and “new job” is not specified in the vicinity, it is recognized that they do not correspond to any attribute, and the extraction is omitted.

  A method for correcting the correspondence between the attribute information and the character string in such a case will be described with reference to FIGS. Here, a method of displaying a character string in a list in association with attribute information of “name” is illustrated.

  First, as shown in FIG. 6, the user designates an insertion location in the column displayed by the correction screen display unit 13. In this example, the designation field C in FIG. 6, which is a blank space in the “name” field, is designated by the input unit 11 (which is the mouse 107, but may be an arrow key or the like on the keyboard 106). The designation field acquisition unit 12 acquires position information (column position information indicated by a row number and column number) and attribute information indicating which field is designated by the input unit 11. Next, the user uses the input unit 11 to specify a document image (left side in FIG. 6) displayed by the document image display unit 16 by a drag operation so as to surround a character string to be associated. The area acquisition unit 14 acquires coordinate information of an area surrounded by the input unit 11.

  The character information acquisition unit 15 determines which of the region information in the extraction result file the specified region obtained by the region acquisition unit 14 matches. Alternatively, area information that is closest to the specified area is extracted from the area information in the extraction result file. The character information acquisition unit 15 acquires a character string corresponding to the matched or approximated region information from the extraction result file. As described above, since all the region information and character strings of the document image are associated with each other in the extraction result file, the character information acquisition unit 15 acquires a character string corresponding to the specified region from the extraction result file. can do.

  The correction screen display unit 13 inserts and displays the character string data acquired by the character string information acquisition unit 15 in the user specification column acquired by the specification column acquisition unit 12. Further, the document image display unit 16 acquires the attribute information of the column specified by the user from the specified column acquisition unit 12, and highlights the character string in the specified region by color coding according to the attribute information. . The display at this time is shown in FIG. By applying such an operation and the operation of the correction terminal 1 to the “previous job” field and the “new job” field, it is possible to eliminate omission of association with attribute information.

  Further, the designation shown in FIG. 8 is also possible. That is, when one column is specified and an area extending over a plurality of lines is specified, a plurality of lines of character strings are inserted into the one column. In this case, the line feed code between lines is removed and inserted.

  In this example, two columns are selected so that the number of columns matches the number of character strings, and two character strings are designated from the document image. When the button is selected, the character string may be inserted into each column according to the line feed code of the character string selected on the document image. For example, one margin in the “Name” field is selected, and multiple lines of text on the document image, in this example, “Toshiba Jiro (+ line feed code)” and “Toshiba Saburo (+ line feed code)” are selected. In this case, the correction screen display unit 13 first displays “Toshiba Jiro” in the first column, and when the line feed code of “Toshiba Jiro” is confirmed, the cursor is advanced to the next column below “Toshiba "Saburo" is displayed. That is, the display is the same as in FIG.

  FIG. 9 is a flowchart illustrating an operation example of the correction terminal 1 when correcting the above-described correspondence deficiency (missing extraction) between the attribute information and the character string. The designation column acquisition unit 12 acquires position information indicating which column in the attribute information list has been selected by the user from the input unit 11 (S201). This position information is, for example, a row number and a column number. The area acquisition unit 14 acquires information on an area in the document image designated by the user using the input unit 11 (S202). The area information is coordinate information in this example. The character information acquisition unit 15 acquires a character string included in the region acquired by the region acquisition unit 14 based on the correspondence relationship between the region information and the character string set in the extraction result file (S203). The correction screen display unit 13 inserts the character string acquired by the character string information acquisition unit 15 into the column acquired by the designation column acquisition unit 12 (S204). Thus, the attribute information and the character string for which the deficiencies are eliminated are output to the confirmation / correction CSV file as normal information.

  In the above example, the main communication item in the document image is about personnel changes, and information on personnel changes is described on one sheet. If it is a free-format document, there are different main communication items on multiple sheets, and if some of them have information on personnel changes, the information on personnel changes is not the beginning of the document. In the middle or end of a document, it may be described in several places. Also, depending on the situation, there may be an enormous amount of information regarding personnel changes over multiple pages. By applying this embodiment, it is possible to improve the operability in such a situation, and to improve the convenience for the user. In other words, according to the present embodiment, the user turns the page in the display area of the document image display unit 16 and confirms a location that is color-coded for each attribute information. In addition, the user can correct the correspondence between the attribute information and the character string in the document with a simple operation without performing manual input (direct input of the character string using the keyboard). Extraction leakage can be reduced. Therefore, more accurate information can be acquired.

(Second Embodiment)
In the second embodiment, a configuration in which operability is further improved as compared with the first embodiment will be described. In the second embodiment, in addition to a normal PC (PC: personal computer), a tablet PC having a touch panel display is used, and the correction work is performed by linking the tablet PC and the normal PC.

  FIG. 10 shows a configuration example of the correction system of the second embodiment. The correction system 100 (information processing system) includes a tablet PC 1B and a terminal 1A (stationary PC or notebook PC). The correction system 100 can transmit and receive data to and from the server system 2 described in the first embodiment via the network 3, and also between the tablet PC 1B and the terminal 1A, a wireless LAN or Bluetooth ( Data transmission / reception is possible using a registered trademark communication means.

  The hardware configuration of the tablet PC 1B is a conventional tablet PC including a processor, a memory, and a storage device that can be stored in a nonvolatile manner. The tablet PC has a touch panel display in which a touch panel type input device that detects pressing of a fingertip (may be a pen tip) and a display device are combined. The input unit 11A and the document image display unit 16A are functional units realized by cooperating with the touch panel display. The terminal 1A has the same hardware configuration as that of the correction terminal 1 of the first embodiment, and the other functional units shown in FIG. 10 have the same functions as those of the first embodiment. Yes.

  FIG. 11 shows a screen image (left side) displayed on the touch panel display of the tablet PC 1B and a screen image (right side) displayed on the display 105 of the terminal 1A. The document image display unit 16A of the tablet PC 1B displays a character string color-coded for each document image and attribute on the touch panel display. Further, an extraction result / correction screen is displayed on the display 105 by the correction screen display unit 13 of the terminal 1A.

  The tablet PC 1B has high visibility of the document (document image), can be page-fed by an operation of sliding the fingertip, and can easily specify a range by a drag operation with the fingertip, but is not suitable for character input. On the other hand, the terminal device 1A has lower operability and visibility for document images than the tablet PC 1B, for example, page feed is performed by a mouse operation, but character input operations are input using the keyboard 106. Therefore, it is easy. In this way, by using the touch panel display and the keyboard and complementing the disadvantages of the respective operability, the operability is improved and the correction work is smoothly performed.

  The user operation procedure is the same as that shown in FIGS. 6 and 7 of the first embodiment except that the input / output devices are different. If there is a defect in the attribute information correspondence and there is omission of extraction, the user first specifies the column in the extraction result / correction screen, and then specifies that the corresponding character string in the document image is surrounded by the fingertip. . As a result, the corresponding character string is displayed in the designation field in the extraction result / correction screen.

  In this embodiment, the document image that is the basis of the correction work is held on the tablet PC 1B side, and the extraction result file is held on both the tablet PC 1B and the terminal 1A, but the mode is limited to this. Rather than being placed in the right place. In addition, when the character information acquisition unit 15 in the tablet PC 1B acquires a character string from the extraction result file, the acquired character string is transmitted to the correction screen display unit 13 in the terminal 1A. Further, the attribute information of the designated field acquired by the designated field acquisition unit 12 in the terminal 1A is transmitted to the document image display unit 16A in the tablet PC 1B. With the transmission from the designation field acquisition unit 12, the document image display unit 16A highlights the character string in the area designated by the fingertip of the user with a color corresponding to the attribute. The operation of each functional unit other than this is the same as that of the first embodiment (see the flowchart in FIG. 9).

  In the present embodiment, the character information acquisition unit 15 is in the tablet PC 1B, but may be in the terminal 1A. The fingertip operation on the touch panel display can be the same as the previous operation method. Display control on the touch panel display (enlarged display, reduced display, display control when turning pages, etc.) and area acquisition on the document image designated by the user are performed based on coordinate information detected by the touch panel. A conventional technique may be employed for this control.

  The area designation screen is displayed on the tablet PC 1B, the screen for character input is displayed on the terminal 1A, and by interlocking these, the page is turned with the fingertip, the range is designated with the finger, and the character is entered with the keyboard. It is possible to operate using the features to be. Thereby, an interface with good work efficiency can be realized.

  In the first embodiment, since it is necessary to display the document image and the extraction result / correction screen on one display 105, the display area of the extraction result / correction screen is reduced accordingly. According to the second embodiment, since the entire display 105 can be used as a correction screen, a document with many attributes can be handled, and an operation with high visibility can be performed.

(Third embodiment)
In the third embodiment, a mounting example using a digital pen or paper dedicated to the digital pen will be described. FIG. 12 is a diagram illustrating a configuration example of the third embodiment. The document reading system 300B has a configuration in which the correction system 100A (information processing system) and the MFP 4 are connected to the server system 2 described in the first embodiment via the network 3. In the MFP 4, a sheet P1 dedicated to the digital pen is stored in the sheet cassette, and a document image is acquired and printed on the sheet P1.

  The correction system 100A includes a sheet P1, a digital pen T1, and a terminal 1C. The user uses the digital pen T1 to specify an area for the document image drawn on the sheet P1. The terminal 1C acquires information indicating which area on the sheet P1 is designated as coordinate information.

  On the sheet P1, dots and grid cells are printed so that the entry position can be identified. The dots and grid cells have different pattern shapes depending on the positions.

  The body of the digital pen T1 incorporates a camera that is arranged so as to capture the direction of the pen tip. The digital pen T1 has a writing pressure sensor, and the camera operates when the writing pressure sensor detects writing pressure. The body of the digital pen T1 includes a processor and a memory. The processor calculates and obtains an entry position based on the captured dot pattern and grid pattern, and the entry position is stored in the memory. Memorize temporarily. In the memory, entry positions are stored in the order according to the handwriting. The digital pen T1 has a built-in device capable of Bluetooth communication and the like, and transmits entry position information in the memory to the terminal 1C.

  As such a sheet P1 and digital pen T1, a conventional digital pen dedicated sheet or a conventional digital pen may be employed.

  Since the terminal 1C is a computer having the same hardware configuration as that of the correction terminal 1 of the first embodiment, description thereof is omitted here (see FIG. 1).

  FIG. 13 is a functional block diagram illustrating a configuration example of the correction system 100A. The correction system 100A includes a paper surface coordinate acquisition unit 21 including a digital pen T1 and a sheet P2. The paper surface coordinate acquisition unit 21 acquires handwriting information (position information on the sheet P1) of the digital pen T1 for the sheet P1 on which the document image is printed. The other functional units shown in FIG. 13 are those that the terminal 1C has, and the functions thereof are the same as those in the first embodiment.

  The aspect of 3rd Embodiment is further demonstrated using FIG. The upper part of FIG. 14 shows an example of a document image printed by the MFP 4. As shown in this example, a check box and attribute information of “name”, “previous job”, and “new job” are specified at the bottom of the sheet P1. Similarly to the first and second embodiments, which part of the sheet P1 has already been extracted as the name, old job, and new job may be displayed in different colors. In this embodiment, the OCR server 2B creates the document image to which the attribute information and the color code are assigned, but the terminal 1C may create the document image.

  The page coordinate acquisition unit 21 acquires position information of a character string surrounded by a frame or underlined by the user. Further, the page coordinate acquisition unit 21 acquires which check box the user has marked with position information. The position information is acquired by the digital pen T1 and stored in the memory of the digital pen T1.

  The region acquisition unit 14 compares the position information acquired by the paper surface coordinate acquisition unit 21 with the coordinate information in the extraction result file, and acquires a region in the document image. The character information acquisition unit 15 acquires a character string corresponding to the region acquired by the region acquisition unit 14. On the other hand, the designation field acquisition unit 12 acquires information indicating which attribute information is specified based on the position information with a check mark acquired by the paper surface coordinate acquisition unit 21. Thereby, the designation column acquisition unit 12 can identify which column on the extraction result / correction screen shown in the lower part of FIG. 14 is designated. The correction screen display unit 13 displays the character string acquired by the character information acquisition unit 5 in an editable state in the column acquired by the designation column acquisition unit 12 (see the lower part of FIG. 14).

  Since the digital pen P1 can store handwriting (position information) in order, write a frame or underline in the character string to be extracted as a name, and then mark the box related to the name If the user follows the order of performing the above, it is possible to determine that the frame or underline described above is a character string serving as name attribute information. Therefore, using the same sheet, specify the previous job (mark the "previous job" check box after specifying the character string), specify the new job (mark the new job check box after specifying the character string) ) Can also be performed. Since the position information is stored sequentially, the area acquisition unit 14, the designation field acquisition unit 12, and the character information acquisition unit 15 that perform subsequent processing can accurately associate the attribute information with the character string. it can.

  In the above description, a sheet dedicated to the digital pen is used. However, in the case of a type in which a sensor device is disposed in the vicinity of the digital pen to detect the writing position, the sheet may not be dedicated to the digital pen. In this case, the configuration of the paper surface coordinate acquisition unit 21 includes a digital pen and a sensor device.

  As in the third embodiment, by using a digital pen, information specified by the user can be easily and reliably extracted.

  The input unit 11 in the first embodiment (see FIG. 4) corresponds to a first input unit and a second input unit. The input unit 11 in the second embodiment (see FIG. 10) corresponds to a second input unit, and the input unit 11A corresponds to a first input unit. The document image display units 16 and 16A correspond to a first display unit, and the correction screen display unit 13 corresponds to a second display unit.

  As described above, according to the aspect described in this embodiment, when there is a character string to be extracted in the free format document and the character string cannot be extracted, it can be corrected by an easy procedure, and data acquisition omission can be reduced. it can.

  In addition, although embodiment of this invention was described, the said embodiment is shown as an example and is not intending limiting the range of invention. The novel embodiment can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

  1 correction terminal, 1A, 1C terminal, 1B tablet PC, 2 server system, 2A FAX server, 2B OCR server, 2C file server, 3 network, 4 MFP, 11 input unit, 11A input unit (touch panel type), 12 specification field acquisition unit, 13 correction screen display unit, 14 area acquisition unit, 15 character information acquisition unit, 16, 16A document image display unit, 21 page coordinate acquisition unit, 100, 100A correction system, 300, 300B document reading system, T1 digital pen, P1 digital pen dedicated sheet.

Claims (6)

A first display for displaying an image of a document that has undergone character recognition;
A first input unit for detecting a range designated by the user in the image;
A second input unit;
A list of character strings that are the result of character recognition in the document is displayed for each attribute information indicating what character string the character string is, and the second input unit in the list is displayed. A second display unit that displays a character string that is a result of character recognition in a range detected by the first input unit in a field designated by the user using;
Information processing system including The information processing system according to claim 1,
The first display unit and the first input unit are information processing systems included in a computer having a touch panel display. The information processing system according to claim 1,
The second display unit displays the attribute information separately for each color, and the first display unit includes a character string in which attribute information is already defined among character strings in the image. An information processing system for displaying the same color as attribute information displayed on the second display unit. On one or more computers,
Display an image of a document that has undergone character recognition,
A range designated by the user in the image is detected;
A character string that is a result of character recognition in the document is displayed in a list for each attribute information that indicates what attribute the character string is a character string, and is displayed in a column specified by the user in the list. An information processing program for displaying a character string that is a result of character recognition in the detected range. One or more computers
Displays an image of a document that has been character-recognized,
Detecting a range specified by the user in the image;
A character string that is a result of character recognition in the document is displayed in a list for each attribute information that indicates what attribute the character string is a character string, and is displayed in a column specified by the user in the list. An information processing method for displaying a character string that is a result of character recognition in the detected range. A digital pen for acquiring position information on a sheet on which an image on which character recognition has been performed is printed;
The character string described in the image is displayed in a list for each attribute information indicating what attribute the character string is a character string of, and the user described in the image of the sheet obtained by the digital pen Based on the position information of the designated character string and the position information of the entry frame which is an entry frame for designating attribute information described on the sheet and specified by the user, the designated of the list A computer for displaying the designated character string written in the image of the sheet in a column of attribute information matching the attribute information;
Information processing system including

Images