JP2009070088A - Information processor and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an information processor which efficiently detects a place corrected by handwriting on a business form and creates information for confirming the description content of the corrected place, at a glance. <P>SOLUTION: The information processor 1 is configured to detect the place corrected by handwriting, with respect to description on a business form based on the image of the business form read by a scanner 2. A cancellation place detection part 12 detects the place where a straight line in a fixed direction, with a length which is equal to or larger than a prescribed value is described or the place where there are description disturbing fixed regularity on the business form 20. A correction region determining part 13 determines that the region to which the place detected by the cancelation place detection part 12 belongs, among the respective regions on the business form 20 designated, in advance, is the region to which the correction by handwriting has been performed. An information creation part 15 generates a business form image in which the region determined by the correction region determining part 13 is made conspicuous, when it is displayed. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an information processing apparatus and a program for detecting a place where a description on a form is corrected by handwriting based on an image of the form read by a scanner.

  In an insurance company, in general, contents written in a form such as a contract application form are registered and managed as electronic information in an information processing apparatus such as a server apparatus. For example, when a contractor renews insurance or makes a new insurance contract, an application form in which the registered address and name of the contractor are printed is printed. If the contractor has a change in his / her address, the contractor draws a strike-through line etc. on the old address written in the address field of the application form, and writes the current address in the blank area by handwriting.

  The insurance company visually checks such handwritten corrections and manually corrects the registered data by an operator through a manual operation to make the correction part conspicuous using a marker pen or the like. However, the task of searching for corrections is burdensome and may be overlooked. In addition, forgetting to fill in with a marker pen or incorrect writing may occur. In addition, the form may be soiled with a marker pen or the like.

On the other hand, a technique for detecting a correction location on a form by using an optical character reader or the like is known. For example, in Patent Document 1, when the apparatus determines that a character has been canceled by an erasing symbol or erasing line, the pattern is deleted, and the character pattern written in the upper margin of the canceling character is validated as a corrected character. Is disclosed.
JP-A-10-116315 (paragraph 0004 etc.)

  However, with the technology of Patent Document 1, even though the corrected character reading result can be obtained, the canceled character information and the corrected portion information on the character form do not remain. For this reason, the operator cannot confirm at a glance what character is corrected to what character in which part of the form. In addition, it is impossible to perform correction processing of registered data after performing such confirmation work.

  The present invention has been made in view of the above circumstances, and is an information processing apparatus capable of efficiently detecting a handwritten correction portion on a form and generating information that allows the description of the correction portion to be confirmed at a glance. And to provide a program.

  An information processing apparatus according to the present invention is an information processing apparatus that detects, based on an image of a form read by a scanner, a place where a description on the form has been corrected by handwriting, A detecting means for detecting a place where a straight line in a certain direction having a length equal to or greater than a value is described or a place having a description that disturbs a certain regularity, and individual areas on the form designated in advance, A determination unit that determines that the region to which the part detected by the detection unit belongs is a region that has been corrected by handwriting, and generates a form image that makes the region determined by the determination unit stand out when displayed And generating means.

  The program according to the present invention is a program for detecting a location where a description on the form is corrected by handwriting based on an image of the form read by the scanner, and has a length longer than a predetermined value on the form. A function for detecting a portion where a straight line having a certain direction is described or a portion having a description disturbing a certain regularity, and the detected portion of the individual areas on the form designated in advance A computer that realizes a function of determining that the area to which the image belongs is an area that has been corrected by handwriting, and a function of generating a form image that highlights the determined area when displayed. To do.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to generate | occur | produce the information which can confirm the description content of the correction location at a glance while detecting the correction location by the handwriting on a form efficiently.

  Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing an example of the configuration of an information processing system according to an embodiment of the present invention.

  This information processing system is, for example, a system used for insurance contract work in an insurance company, and includes an information processing device 1, a scanner 2, a display device 3, an input device 4, and a server device 5. The scanner 2, the display device 3, and the input device 4 may be included in the information processing device 1 or may be connected to the information processing device 1 as external devices. In addition, some of the various functions provided in the information processing apparatus 1 may take a form provided in another information processing apparatus connected to the information processing apparatus 1 via a network. The server device 5 is connected to the information processing device 1 via a network, for example.

  The information processing apparatus 1 is realized as a personal computer (PC), for example. Based on the image of the form 20 read by the scanner 2, the information processing apparatus 1 efficiently detects a handwritten correction part on the form 20 and generates information that can confirm the description of the correction part at a glance. Then, the contract information reflecting the correction content can be transmitted to the server device 5 in accordance with the instruction from the operator input through the input device 4 or displayed on the display device 3.

  An example of the form 20 is an insurance application form as shown in FIG. On the form 20, an entry frame (ruled line or the like) for entering an address, a name and the like is printed. The entry frame is printed in dropout color. In the figure, the name, gender, date of birth, telephone number, etc. other than the address are omitted, but it is assumed that there are actually printed characters printed. Also, it is assumed that a type mark (not shown) that allows the scanner 2 or the information processing apparatus 1 to identify the type of the form 20 by using an existing technology is also printed. When the applicant's own address is changed, the applicant draws a strike-through line 21 for example in the old address described in the address column in the application form, and writes the current address in the blank portion 22 by handwriting. The information processing apparatus 1 can detect such a correction location.

  The scanner 2 is a device that optically scans the surface of the form 20 and performs photoelectric conversion to generate information described on the form 20 as a color or monochrome image. For example, the scanner 2 emits light on the form 20. It comprises a light source for irradiation, a self-scanning CCD image sensor that receives the reflected light and converts it into an electrical signal, and the like. The image signal generated by the scanner 2 is supplied to the information processing apparatus 1.

  The display device 3 is a device that displays various types of information generated from the information processing device 1.

  The input device 4 is a device for issuing various instructions to the information processing device 1.

  The server device 5 is a device that manages a database including contract information (application number, application date, name, gender, address, telephone number, insurance item, insurance period, securities number, etc.) related to various insurance contracts. In response to a request from the information processing device 1, the server device 5 transmits the registered contract information to the information processing device 1, or receives the contract information reflecting the correction content transmitted from the information processing device 1. Or register.

  The information processing apparatus 1 includes an image information storage unit D1, a form information storage unit D2, a character dictionary storage unit D3, a character recognition result storage unit D4, and a contract information storage unit D5 as databases for storing various types of information. These are realized using a storage medium such as a hard disk or a memory. In addition, the information processing apparatus 1 includes, as various functions, an image capture unit 11, a cancellation location detection unit 12, a correction area determination unit 13, a character recognition unit 14, an information generation unit 15, a user interface 16, and a correction processing unit 17. Have All or part of these functions are realized as a computer program that can be executed by a processor such as a CPU (Central Processing Unit).

  The image information storage unit D1 stores the image information of the form 20 captured by the image capturing unit 11, or the correction region determined by the correction region determination unit 13 (the region determined to have handwritten correction). The coordinate information to be specified is stored together with the image information.

  The form information storage unit D2 stores information related to forms of various forms including the form 20, and for example, individual check target areas (names, addresses, etc., entry frames such as names and addresses, or the like) Coordinate information of a substantial character entry area in the entry frame is stored. This information includes coordinate information of a region where the presence / absence of a strikethrough or the like is to be checked, coordinate information of a region where character recognition is to be performed, and the like.

  The character dictionary storage unit D3 stores a character dictionary used for character recognition processing in the character recognition unit 14.

  The character recognition result storage unit D4 stores character information obtained as a result of character recognition by the character recognition unit 14.

  The contract information storage unit D5 stores contract information acquired from the server device 5, and stores contract information after the correction processing by the correction processing unit 17 is performed.

  The image capturing unit 11 captures image information generated by the scanner 2 and stores the captured image information in the image information storage unit D1.

  The cancel location detection unit 12 is based on the image information stored in the image information storage unit D1, and a location (for example, single or double) on the form 20 where a horizontal straight line having a length equal to or greater than a predetermined value is described. That detect heavy strikethroughs) or parts that disturb certain regularity (for example, places where a cancellation mark such as x or seal is written for a certain character) The coordinate information of the location is output as a detection result.

  Note that the cancel location detection unit 12 does not set all areas on the form 20 as detection targets, but can set only necessary areas as detection targets. In that case, the cancel location detection unit 12 narrows down the detection target by referring to the coordinate information of the check target region stored in the form information storage unit D2. Thereby, shortening of detection time and prevention of false detection can be realized.

  In order to detect a strikethrough, for example, a straight line in the horizontal direction (the same direction as the character string) having a length of 1 cm or more on the form 20 is searched. Since the width of one character is usually less than 1 cm, a straight line with a width of 1 cm or more can be regarded as a strikethrough. Note that the cancellation part detection unit 12 does not detect the line of the entry frame printed in the dropout color on the form 20. For this reason, the line of the entry frame is not erroneously detected as a cancellation line.

  On the other hand, in order to detect a cancellation mark, for example, a search is made in a region where a description with a certain regularity is required on the form 20, for example, where the proportion of disorder of regularity exceeds a predetermined value. If the form 20 is an area where numbers are required to be written consecutively (area where postal codes, telephone numbers, etc. are written), the place where the continuity of numbers is lost in that area Search etc. The number recognition process can be realized by using a function of a character recognition unit described later, for example. If a part of a plurality of consecutive numbers is not detected as a number, or if the recognition rate as a number is less than the specified value, a mark other than a number (such as X ) Can be considered written. This detection method can be applied not only to numbers but also to characters. Furthermore, by searching for a red portion (red seal impression) on the form 20, it is possible to detect the location of the correction mark by the seal.

  Based on the coordinate information of the location detected by the cancellation location detection unit 12 and the coordinate information of the individual check target regions on the form 20 stored in the form information storage unit D2, the correction region determination unit 13 uses handwriting. The area where correction has been performed (correction area) is determined. That is, the correction area determination unit 13 determines that, among the individual check target areas on the form 20, the area to which the part detected by the cancellation part detection unit 12 belongs is an area that has been corrected by handwriting. . In addition, as a range of the correction area to be determined, an entry frame itself such as an address or a name may be the correction area, or a substantial character entry area in the entry frame may be the correction area.

  Further, the correction area determination unit 13 includes, among the individual check target areas on the form 20, an area including a description that disturbs certain regularity in addition to the area to which the location detected by the cancellation location detection unit 12 belongs. In such a case, the area is also provided with a function for determining that the area has been corrected by handwriting. By using this function, it is possible to detect areas that contain information indicating the intention of canceling characters in a form different from the strikethrough or cancel mark, and further improve the accuracy when determining the correction area. Is possible.

  In addition, after determining the correction area, the correction area determination unit 13 stores the coordinate information specifying the correction area in the image information storage unit D1 together with the image information of the form 20.

  The character recognition unit 14 performs character recognition processing described on the form 20 using the character dictionary stored in the character dictionary storage unit D3, and stores the character recognition result in the character recognition result storage unit D4. It is something to be made. In particular, the character recognition unit 14 performs a recognition process for handwritten characters included in the correction area determined by the correction area determination unit 13. In that case, the character recognition part 14 memorize | stores the coordinate information which shows the position of the said handwritten character with the recognition result of a handwritten character in the character recognition result memory | storage part D4.

  In response to a request from the user interface 16, the information generation unit 15 generates various types of information by selectively using information stored in the various storage units D1, D4, and D5.

  For example, when the information generation unit 15 receives a request for information for forming a form image viewing screen from the user interface 16, the information generation unit 15 stores the image information of the form 20 stored in the image information storage unit D1 (the coordinate information of the correction area). In addition, a form image (for example, a form image for enlarging or highlighting the correction area when displaying the correction area) is generated when the display device 3 displays the correction area. An example of the screen when the form image generated in this way is displayed on the display device 3 through the user interface 16 is shown in FIG.

  In the browsing screen 30 shown in FIG. 3, there is a window 31 for displaying a form image. In the figure, the name, gender, date of birth, telephone number, etc. other than the address are omitted, but it is assumed that there are actually printed characters printed. The form image includes a correction area 32 that is highlighted by a thick outline or a decoration such as shading or coloring. Further, an image 33 in which the correction area 32 is enlarged may be displayed separately. In the example of FIG. 3, the substantial character entry area in the address entry frame is used as the correction area as the correction area range, but the address entry frame itself may be used as the correction area.

  Thereby, the operator can confirm at a glance what character has been corrected to what character at which location on the form.

  For example, when the information generation unit 15 receives a request for information for forming a contract information editing screen from the user interface 16, the information generation unit 15 stores the contract information stored in the contract information storage unit D5 in advance and the character recognition result storage. Information including character information in the correction area stored in the part D4 and image information of the form 20 stored in the image information storage part D1 (with coordinate information of the correction area) is generated. An example of the screen when the edit screen information based on the information generated in this way is displayed on the display device 3 through the user interface 16 is shown in FIG.

  The editing screen 40 shown in FIG. 4 includes a contract information editing window 41 in addition to the window 31 for displaying the above-described form image. In the figure, the name, gender, date of birth, telephone number, etc. other than the address are omitted, but it is assumed that there are actually printed characters printed. In the address input field 42 in the window 41, character information “3-22 Katamachi, Fuchu-shi, Tokyo” is automatically input as a result of character recognition for handwritten characters on the form. In the vicinity thereof, the copy 43 of the correction area 32 is displayed. When the operator depresses an OK button (not shown), the contract information including the contents corrected this time is stored.

  This makes it easy for the operator to compare handwritten characters with printed characters, making it easy to check whether corrections have been entered correctly and ensuring that correct corrections are reflected in the contract information. It becomes possible.

  The user interface 16 performs processing requested by the operator via the display device 3 and the input device 4. For example, when there is a request for viewing a form image through the input device 4, the user interface 16 acquires information that forms a view screen for the form image from the information generation unit 15, and displays the view screen based on the acquired information on the display device 3. To display. When there is a request for editing contract information through the input device 4, the user interface 16 acquires information forming an edit screen for contract information from the information generation unit 15 and displays an edit screen based on the acquired information on the display device 3. To display. Further, when there is a request for saving contract information including correction content through the input device 4, the user interface 16 requests the correction processing unit 17 to reflect the correction content in the contract information and save it.

  In response to a request from the user interface 16, the correction processing unit 17 performs processing for reflecting the correction content in the contract information and storing it.

  The information communication unit 18 transmits the contract information reflecting the correction contents in the contract information storage unit D5 to the server device 5, acquires the contract information from the server device 5, and stores the contract information in the contract information storage unit D5. Is.

  Here, the operation of the present system will be described with reference to the flowchart of FIG.

  The image capture unit 11 captures the image information generated by the scanner 2 and stores the captured image information in the image information storage unit D1 (step S1).

  Next, the cancel location detection unit 12 is based on the image information stored in the image information storage unit D1 on the form 20 where a horizontal straight line having a length equal to or greater than a predetermined value is described (for example, Where there is a single or double strike-through line) or where there is a description that disturbs certain regularity (for example, a place where a cancellation mark such as x or seal is written for a certain character), etc. It detects and outputs the coordinate information of the location as a detection result (step S2).

  Next, the correction area determination unit 13 performs correction by handwriting on the area to which the part detected by the cancellation part detection unit 12 among the individual check target areas on the form 20 belongs. (Step S3). In addition, the correction area determination unit 13 includes, among the individual check target areas on the form 20, an area including a description that disturbs certain regularity in addition to the area to which the location detected by the cancellation location detection unit 12 belongs. In this case, it is determined that the region is also a region where correction by handwriting has been performed. After determining the correction area, the coordinate information for specifying the correction area is stored in the image information storage unit D1 together with the image information of the form 20.

  Next, the character recognizing unit 14 uses the character dictionary stored in the character dictionary storage unit D3 to include characters (particularly included in the correction area determined by the correction area determination unit 13) described on the form 20. (Handwritten character) is recognized, and the character recognition result is stored in the character recognition result storage unit D4 (step S14).

  Next, when the information generation unit 15 is notified through the user interface 16 that a request for viewing a form image or a request for editing contract information is made from the input device 4, the information generation unit 15 stores the information in various storage units D1, D4, and D5. Information that is selectively used to make the correction area stand out when displayed on the display device 3 (for example, a form image that is enlarged or highlighted when the correction area is displayed) or contract information The information for forming the edit screen is generated (step S15).

  Next, the user interface 16 causes the display device 3 to display a form browsing screen and a contract information editing screen based on the information acquired from the information generation unit 15 (step S16).

  Next, when there is a request for saving contract information including correction content through the input device 4, the user interface 16 requests the correction processing unit 17 to reflect the correction content in the contract information and save it. As a result, the correction processing unit 17 performs a process of reflecting the correction content in the contract information and storing it (step S17).

  Next, the information communication unit 18 transmits the contract information reflecting the correction contents to the server device 5. Thereby, the server apparatus 5 performs registration of the contract information in which the correction content is reflected (step S18).

  As described above, according to the present embodiment, in addition to being able to efficiently detect the handwritten correction portion on the form, the operator can determine what character is in what place on the form through the view screen of the form image. You can check at a glance whether the characters have been corrected. In addition, the operator can easily compare handwritten characters and printed characters through the contract information editing screen, making it easy to check whether corrections have been entered correctly and ensuring correct corrections. Can be reflected in the contract information.

  The various processing procedures according to the present invention described in the above-described embodiments are stored in a storage medium (for example, a magnetic disk, an optical disk, or a semiconductor memory) that can be read by a computer (information processing apparatus) as a computer program. If necessary, it may be read and executed by the processor. Such a computer program can also be distributed by transmitting from one computer to another computer via a communication medium.

  The present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

1 is a block diagram showing an example of the configuration of an information processing system according to an embodiment of the present invention. The figure which shows an example of the form handled by this system. The figure which shows an example of the browsing screen of a form image. The figure which shows an example of the edit screen of contract information. The flowchart which shows operation | movement of this system.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Information processing apparatus, 2 ... Scanner, 3 ... Display apparatus, 4 ... Input device, 11 ... Image capture part, 12 ... Cancel location detection part, 13 ... Correction area determination part, 14 ... Character recognition part, 15 ... Information Generation unit, 16 ... user interface, 17 ... correction processing unit, 18 ... information communication unit, 20 ... form, D1 ... image information storage unit, D2 ... form information storage unit, D3 ... character dictionary storage unit, D4 ... character recognition result Storage unit, D5 ... Contract information storage unit.

Claims (5)

An information processing apparatus for detecting a location where a description on the form is corrected by handwriting based on an image of the form read by a scanner,
Detecting means for detecting a place where a straight line in a certain direction having a length of a predetermined value or more is written on the form or a place where there is a description that disturbs a certain regularity;
Of the individual regions on the form designated in advance, a determination unit that determines that the region to which the portion detected by the detection unit belongs is a region that has been corrected by handwriting,
An information processing apparatus comprising: generating means for generating a form image that makes the area determined by the determining means stand out when displayed.   The information processing apparatus according to claim 1, wherein the entry frame on the form is printed in a dropout color that is not detected by the detection unit.   In the case where there is an area including a description that disturbs a certain regularity in addition to the area to which the portion detected by the detection means belongs, among the individual areas on the form designated in advance, The information processing apparatus according to claim 1, wherein the information processing apparatus determines that the area is a corrected area by handwriting.   The information processing apparatus according to any one of claims 1 to 3, wherein the generation unit generates a form image to be displayed in an enlarged or highlighted manner when the area determined by the determination unit is displayed. . Based on the image of the form read by the scanner, a program for detecting a place where the description on the form is corrected by handwriting,
A function for detecting a place where a straight line in a certain direction having a length of a predetermined value or more is written on the form or a place where there is a description that disturbs a certain regularity
A function that determines that the area to which the detected location belongs among the individual areas on the form designated in advance is an area that has been corrected by handwriting;
A program for causing a computer to realize a function of generating a form image that makes the determined area stand out when displayed.

Images