JP2008282091A - Document processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To discriminate between a document and an attached file, to integrate the document and the attached file together, to automatically perform filing more accurately, and to efficiently carry out correction when the document and its attached file are input together. <P>SOLUTION: In this document processing device, a document type of an input image is identified, then, it is determined whether the following input image is an attached file or not based on attached file presence/absence information of the identified document type, and the image for which the document type is identified and the image to be determined as the attached file are filed as a pair of data. The number of sheets of the attached file is counted, and based on the counting result, a warning on a pair data, in which a read error may occur, is given. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a form processing apparatus and form processing method for processing a form.

  Previously, it was possible to register a single form or multiple forms and file a document containing only the form.

On the other hand, Patent Document 1 discloses a filing method for filing a fixed document with a fixed format and an atypical attachment with a fixed format as one data.
JP 2001-5831 A

  However, Patent Document 1 has a problem in that a form is identified and all images that cannot be identified are determined as attachments, and even an unknown form that cannot be actually identified becomes attachments. In addition, if it is erroneously identified as an attached material, there is a problem that correction takes time.

  An object of the present invention is to identify a form and an attached material even in a form with an attached material, integrate the form and the attached material, and automatically and more accurately file and efficiently correct the form.

  In order to solve the above problems, a form processing apparatus of the present invention is a form processing apparatus for filing a form and an attached material of the form as a set of data, and an input means for continuously inputting a plurality of images; Form identification means for identifying the form type of the input image, and attachment determination for determining whether the next input image is an attachment based on the attachment presence / absence information of the identified form type And filing means for filing an image whose form type is identified by the form identifying means and an image determined as an attached material by the attached material determining means as a set of data. .

  Further, there is a possibility that a reading error has occurred based on the counting means for counting the number of the attached materials and the result counted by the counting means when the attached material judging means judges the attached material. And warning means for warning a set of data.

  According to the present invention, even a form with an attached material can be automatically and more accurately filed and corrected efficiently.

Example 1
Next, an embodiment of a form recognition apparatus according to the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing an embodiment of a form processing apparatus according to the present invention, FIG. 2 is a view showing an example of input when filing a plurality of images, and FIG. FIG.

  In FIG. 1, the form processing apparatus has an image input means 11 such as a scanner, a camera, or a file reading apparatus, and a one-page or multiple-page form or attached material is input from the image input means 11 as image data. In this embodiment, it is assumed that a document scanner including an ADF (Auto Document Feeder) is used as an image input unit.

  The image input means 11 is connected to the bus 17, and is connected to the processor 12 that executes a form processing process via the bus 17.

  A keyboard 13, a disk 14, a memory 15, and output means 16 are further connected to the bus 17.

  The memory 15 stores temporary data when the processor 12 creates processing control information, or stores image data read by the image input means 11. Also, registered form data (data for identifying the form type) including the attached document presence / absence information of the registered form is read from the disk 14 and held. In addition, a control program 15c for causing a computer to execute processing corresponding to a flowchart described later is stored. Note that these computer-readable control programs may be supplied by a computer-readable storage medium, or may be supplied via a network or the like.

  The processor 12 functions as a form registration unit 12a, a form identification unit 12b, and a filing unit 12c by executing a control program 15c stored in the memory 15.

  The keyboard 13 performs instruction operations such as registration and identification of forms, and the disk 14 functions as an auxiliary storage device.

  The output means 16 is a display, a printer, or the like, and outputs various information such as identification results.

  The form image read by the image input means 11 is stored in the memory 15 as a form image 15b. The form image 15b is binarized by an image processing means (not shown) provided in the processor 12 and sent to the form registration means 12a and the form identification means 12b.

  Then, the area included in the form image is identified by taking a histogram of black dots, etc., and the area included in the form image 15b is attributed to table, text, picture, etc., and the resolution, position information of each area, The information is extracted, input form data is generated, and stored in the memory 11 and the disk 14.

  FIG. 2 is a diagram illustrating an example of input when filing an image composed of a form and an attached material. The first form A (201) has three attached materials (202 to 204), the next form B (205) has no attached material, and the next form A (206) has two attached materials (207, 208). The next form C (209) is composed of three attached materials (210 to 212). In this state, it is assumed that continuous input is performed from the auto document feeder.

  FIG. 3 is a diagram showing attachment presence / absence information of the form A, the form B, and the form C. Form A, form B, and form C are registered as form types, and attached material presence / absence information indicating whether or not there is a possibility of attaching attached materials is registered in association with each form type. . In the example 3, the form A is registered as “present”, the form B is registered as “not present”, and the form C is registered as “present”. Even when the presence / absence information of the attached material is “present”, the number of attached materials is not specified, and the number of attached materials is one of 0 to plural. The attached material presence / absence information is designated in advance by the user. The type of form may be designated by the user or may be automatically assigned. Also, feature data used when identifying the type of each form is registered in association with each form type. The form identification process is compared using these feature data.

  Assume that images are input in the order shown in FIG. First, if the first image (201) is identified as a form and identified as a form type A, it is determined that there is a possibility that an attached material is attached to the form 201 from the attached material presence / absence information. Next, when the second image (202) is read and the forms are identified and it is determined that there is no similar form type (the form type is unknown), the attachment of the form 201 that can be read and identified immediately before is attached. Since the material presence / absence information is “present”, it is determined that the second image 202 is an attached material of the form 201 identified immediately before. Similarly, it is assumed that the form type of the third image (203) and the fourth image (204) is unknown, and the fifth image (205) is identified as the form B. Therefore, since the attached document presence / absence information of the form 201 is “present” and the three images 202 to 204 read next are continuous as unknown forms, the images 202 to 204 are attached to the form 201. Judging and filing the first four sheets (201 to 204) together into one set of data. The fifth image is identified as form B, and it is determined that there is no attached material from the attached material presence / absence information. Therefore, filing is performed using only the fifth image. When the next image (206) is read and identified as a form, it is identified as form A. Similarly, the next two unknown forms (207, 208) are determined as attachments, and three (206-208). ) Is filed as a set of data, the next sheet (209) is identified as form C, and the next three sheets (210-212) are determined as attachments, and the set of data (209-212). Filed.

  Thus, even a form in which the number of attached materials fluctuates can be automatically filed for each form.

  FIG. 4 is a flowchart showing an example of a form registration process when registering a registered form used for form identification. In a form identification process (FIG. 5) described later, the form type of the input image to be identified is identified using the registered form data registered in FIG.

  Step S41: The form is read from the image input means 11 and transferred to the memory as image data.

  Step S42: The operator inputs the form type from the keyboard 13.

  Step S43: The operator inputs from the keyboard 13 whether there is a possibility that an attached material is attached to the form.

  Step S44: A feature amount is extracted from the image read in step S41, and processing such as creation of format data is executed. The form type input in step S42 and the attached material input in step S43 may be attached. The information is registered as shown in FIG.

  Step S45: It is determined whether the registered image is over. If there are still images to be registered, the process returns to step S41.

  FIG. 5 is a flowchart illustrating an example of the form identification process in the present embodiment.

  Step S501: The first image is read from the input image and transferred to the memory.

  Step S502: A feature amount is extracted from an image, form feature data is created, and form identification is performed by comparing with feature data of a registered form registered in FIG.

  Step S503: The form is identified, and it is determined whether it is similar to the registered form. If it is similar to the registered form, it is determined that the image is a form, and the process proceeds to step S504. If there is no similar registered form, the process proceeds to step S506.

  Step S504: If it is determined in step S503 that there is a similar registered form, the filing information for the image is set as “form”. Then, the process proceeds to step S505.

  Step S505: Attached material presence / absence information associated with the registration form determined to be similar in step S503 is obtained and stored in association with the image. In step S509, the file is filed as a new form file.

  Step S506: If it is determined in step S503 that there is no registered form similar to the input image to be processed, the process proceeds to step S506, and whether the attached material presence / absence information of the form file identified immediately before is “present” or “not present”. To determine whether it is an attachment. If the attached material presence / absence information is “present”, it is determined that the image is an attached material of the immediately preceding form file, and information “attached material” is set in step S507. If the attached material presence / absence information is “none”, it is determined that it is unknown, and information “unknown” is set in step S508.

  Step S509: Filing is performed based on whether the filing information set in the image is “form”, “attached material”, or “unknown”. If it is “form”, a new form file is created (or stored in a new form storage folder). In the case of “attached material”, it is combined with the form file and stored as an attached material of the form file created immediately before (or stored as an attached material in the form storage folder created immediately before). In the case of “unknown”, an unknown image is stored in an unknown image storage folder or the like. The user can check the unknown image storage folder later and manually sort it or register it as a new registration form.

  Step S510: It is determined whether or not the continuously input images are finished. If there is an unprocessed image, the process returns to step S501. If all the images have been processed, the process ends.

  As described above, according to this embodiment, by specifying and registering attachment presence / absence information when registering a form, by identifying and checking the information when identifying the form, the attached material Forms with can be integrated into a set of data and filed, and unknown forms can be easily modified later.

(Example 2)
FIG. 6 is a block diagram illustrating an embodiment of a form processing apparatus according to the second embodiment.

  In FIG. 6, the form processing apparatus is substantially the same as the configuration of the first embodiment, and thus description of common parts is omitted. The processor 12 according to the second embodiment executes a control program 15c stored in the memory 15, thereby forming a form registration unit 12a, a form identification unit 12b, an attached material determination unit 12c, an attached material number counting unit 12d, and a filing unit 12e. Function.

  FIG. 7 is a diagram showing an example of input when filing an image composed of a form and an attached material. The first form A (701) has three attachments (702 to 704), the next form B (705) has no attachments, and the next form A (706) has two attachments (707, 708). The next form C (709) is composed of three attached materials (710 to 712), and the next form A (713) is composed of three attached materials (714 to 716). In this state, it is assumed that an input is made from an auto document feeder or the like.

  The attached material presence / absence information of each form type (form A, form B, form C) is assumed to be held in a table as shown in FIG.

  Then, the document is placed on the auto document feeder in the order as shown in FIG. 7 and the image is read. Here, it is assumed that the ninth image: form C (709) is double-fed and a reading error occurs.

  First, if the first image (701) is identified as a form and is identified as a form A, it is determined that there is a possibility that an attached material is attached from the attached material presence / absence information. Next, if the second image (702) is read, the form is identified, and it is determined that it is not similar to the registered form, there is no attached material information on the immediately preceding form A. Count the number of documents. Similarly, the third image (703) and the fourth image (704) are also determined as attachments. The fifth image (705) is identified as form B. Therefore, the first four sheets (701 to 704) can be collectively filed into a set of data. As a result, the form data of the first set is the form A and the number of attached materials is 3.

  Next, the fifth image is identified as a form B, and it is determined that there is no attached material from the attached material presence / absence information. Therefore, only the fifth image is filed as the second set of form data, and the number of attached materials is zero.

  Then, when the sixth image (706) is read and the form is identified, it is identified as form A, and the seventh and eighth images (707, 708) are similarly determined as attachments. Here, it is assumed that the original 9th image (709: form C) has been missed (doubled with the original 708). In this case, the attached material (710) of the next form C is the ninth. Then, when the form is identified, this image is not similar to the registered form, and is thus determined to be attached to the form A (706). Similarly, the tenth and eleventh (711, 712) are also determined as attachments. The twelfth image (713) is identified as a form A. As a result, the third set of form data (706 to 708, 710 to 712) has the form type of form A and five attached materials.

  The thirteenth, fourteenth and fifteenth (714 to 716) are similarly attached to the form A. Therefore, the fourth set of data (713 to 716) is a form A with three attached materials.

  In this example, the number of attachments for each data as a result of filing is displayed as shown in FIG. In this way, the form type and the number of attached materials identified for each set of data are identified and displayed.

  Here, for example, when the mode value of the number of attached materials of the form A is obtained, it becomes three. The first group data and the fourth group data have a difference of 0, but the third group data has two differences. Therefore, a warning is displayed on the third set of data. For example, the color of the row of the third set data is changed and displayed.

  The user can verify that there is a mistake by verifying the document of the third set of data, and can re-read the image data of that portion and file it correctly.

  In this way, when the number of attached materials is counted, by warning the difference between the optimum value, it is possible to easily find and correct a portion having a reading error.

  The form registration process in the second embodiment is performed in the same manner as the flowchart of FIG. 4 described in the first embodiment, but the processes in steps S43 and S44 are partially different.

  Step S41: The form is read from the image input means 11 and transferred to the memory as image data.

  Step S42: The operator inputs the form type from the keyboard 13.

Step S43: The operator inputs from the keyboard 13 whether there is a possibility that an attached material is attached to the form. Here, in the second embodiment, the average number of attached materials is input. (In another embodiment, the minimum and maximum number of attached materials may be input instead of the average number.)
Step S44: The feature amount is extracted from the image read in step S41 and processing such as format data (feature data) creation is executed, and the form type input in step S42 and the attached material input in step S43 are attached. Attached material presence / absence information indicating whether or not there is a possibility and an average number of sheets are registered in association with each other. (If the minimum and maximum number of attached materials are entered instead of the average number, they are registered in association with each other.)
Step S45: It is determined whether the registered image is over. If there are still images to be registered, the process returns to step S41.

  FIG. 9 is a flowchart illustrating an example of form identification processing according to the second embodiment.

  Step S601: The first image is read from the input image and transferred to the memory.

  Step S602: Feature identification is performed by extracting feature values from the image, creating form feature data, and comparing it with feature data of the registered form.

  Step S603: Identify the form and determine whether it is similar to the registered form. If it is similar to the registered form, it is determined that the image is a form, and the process proceeds to step S604. If there is no similar registered form, the process proceeds to step S606.

  Step S604: If it is determined in step S603 that there is a similar registered form, the filing information for the image is set as “form”. Then, the process proceeds to step S605.

  Step S605: Attached material presence / absence information associated with the registered form determined to be similar in step S603 is obtained and stored in association with the image. In step S609, filing is performed as a new form file.

  Step S606: If it is determined in step S603 that there is no registered form similar to the input image to be processed, the process proceeds to step S606, and whether the attached material presence / absence information of the form file identified immediately before is “present” or “not present”. To determine whether it is an attachment. If the attached material presence / absence information is “present”, it is determined that the image is an attached material of the immediately preceding form file, and information “attached material” is set in the filing information in step S607. If the attached material presence / absence information is “none”, it is determined that the information is unknown, and information “unknown” is set in the filing information in step S608.

  Step S609: Filing is performed based on whether the filing information set in the image is “form”, “attached material”, or “unknown”. If it is “form”, a new form file is created (or stored in a new form storage folder). In the case of “attached material”, it is combined with the form file and stored as an attached material of the form file created immediately before (or stored as an attached material in the form storage folder created immediately before). In the case of “unknown”, an unknown image is stored in an unknown image storage folder or the like. The unknown image storage folder can be examined later, and can be manually sorted or registered as a new registration form.

  Step S610: When the filing information is “attached material”, the number of attached materials is counted for each data (for each form file).

  Step S611: It is determined whether or not the continuously input images are over. If there is an unprocessed image, the process returns to step S601, and if all the images have been processed, the process proceeds to step S612.

  Step S612: When all the images have been processed, the form type and the number of attached materials are displayed for each filed data.

  Step S613: The optimum number of attached materials is acquired for each form type. The optimum number of acquired attachments is compared with the number of attachments of data, and if there is a difference, the data is displayed as a warning. For example, the number of attached materials is displayed for each data, and data determined to have a large difference is displayed with a different color. This optimum number of sheets may be determined based on the mode value of the number of attached materials of each form type identified when continuously entered, or the average value entered when registering the form Or a range may be used. Which of these is used is determined in advance. For example, when an average value registered in advance is used, the average value is compared with the number of identified attached materials, and data having a difference of a certain value or more is displayed as a warning. In addition, when using the minimum number and the maximum number registered in advance, a warning is displayed for data outside these ranges.

  As described above, according to the present embodiment, when filing a form with attached materials into one set of data, the number of attached materials is counted and stored, and attached for each data. By displaying the number of documents and warning the difference between the optimum number for each form type, it is possible to easily find the place where the error occurred even if a reading error occurs.

  Note that the above-described form processing apparatus extracts feature amounts from the image, generates the registered form data, and performs form registration and form identification.However, the form is identified by reading characters in a certain area, etc. It may be realized by another form identification method.

It is a block diagram which shows one Embodiment of the form processing apparatus which concerns on this invention. It is a figure which shows the example of an input when filing a some image. It is a figure which shows the example which registered attachment presence / absence information on the form. It is a flowchart which shows one Embodiment of a form registration process. 6 is a flowchart illustrating an embodiment of a filing process. FIG. 10 is a block diagram of a form processing apparatus according to a second embodiment. It is a figure which shows the example of an input when filing a some image. It is a figure which shows the result of filing an image and counting the number of attached materials. 10 is a flowchart illustrating an embodiment of a filing process according to the second embodiment.

Claims (10)

A form processing device for filing a form and its attached material as a set of data,
Input means for continuously inputting a plurality of images;
Form identification means for identifying the form type of the input image;
Attachment determination means for determining whether or not the next input image is an attachment based on the attachment presence / absence information of the identified form type;
Filing means for filing an image whose form type is identified by the form identifying means and an image judged as an attached material by the attached material judging means as a set of data;
A form processing apparatus characterized by comprising: A counting means for counting the number of attachments when the attachment determination means determines that the attachment is an attachment;
Warning means for warning a set of data that may have caused a reading error based on the result counted by the counting means;
The form processing apparatus according to claim 1, further comprising:   The warning means warns a set of data in which a reading error may have occurred, based on a mode value of the number of attached materials and a result counted by the counting means. The form processing apparatus according to 2.   The warning means warns a set of data in which a reading error may occur based on an average value of the number of attached materials registered in advance and a result counted by the counting means. Item 3. The form processing apparatus according to Item 2.   The warning means warns a set of data in which a reading error may have occurred, based on a maximum value and a minimum value of the number of attached materials registered in advance and a result counted by the counting means. The form processing apparatus according to claim 2.   The attachment determination unit determines whether the next input image is an attachment based on the attachment presence / absence information when the form identification unit does not determine the form type of the next input image. The form processing apparatus according to claim 1, wherein it is determined whether or not it exists.   2. The apparatus according to claim 1, further comprising an unknown image determination unit that determines, as an unknown image, an image whose form type is not determined by the form identification unit and is not determined as an attachment by the attachment determination unit. The form processing device described in 1. A form processing method for a form processing apparatus for filing a form and an attached material of the form as a set of data,
An input process for continuously inputting a plurality of images;
Form identification process for identifying the form type of the input image,
An attachment determination step for determining whether the next input image is an attachment based on the attachment presence / absence information of the identified form type;
A filing step of filing the image in which the form type is identified in the form identification step and the image determined to be an attachment in the attachment determination step as a set of data;
A form processing method for a form processing apparatus, comprising: A computer program for causing a form processing apparatus to file a form and an attached material of the form as a set of data,
In the computer of the form processing apparatus,
Form identification procedure for identifying the form type of the input image,
An attachment determination procedure for determining whether or not the next input image is an attachment based on the attachment presence / absence information of the identified form type;
A filing procedure for filing an image whose form type is identified in the form identifying procedure and an image determined as an attached material in the attached material determining procedure as a set of data;
A computer program for running.   A computer-readable storage medium storing the program according to claim 9.

Images