JP2008293353A - Image input device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem that, with respect to electronic storage of a form or the like by an image, it is necessary to judge whether an original to be stored satisfies a condition to be formed as a form or not at first and whether the original is legal or not, and operation for scanning the original, providing a time stamp and electronic signature to the original and then storing the original is non-efficient in a working cost and a system maintenance cost. <P>SOLUTION: The image input device for scan-inputting an image document and storing an image and an index in a document management system is composed of a form recognition means, an OCR or the like and a user interface or the like, and after evaluating the validity of an original, inputs the image document and the index in the document management system. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image input device that inputs an image from an MFP (digital multifunction peripheral) having a scanner device and a scanner device, and stores the image and its index data in a document management system.

  Conventionally, in an image input apparatus, an image read from a scanner is stored in a document management system as it is. In addition, when an image is OCRed, the main purpose of its use is processing for creating index data (image attributes) for search.

  On the other hand, an electronic file such as an electronic form, in which the form format and data are individually managed, can be processed to detect necessary data loss, etc., but in the case of an image, a technique for making such a determination has been proposed. Not.

As a conventional example, for example, an electronic signature and ID using highly accurate time information are attached to document data and stored in an original storage unit (see Patent Document 1).
JP 2003-298575 A

  Regarding electronic storage using images of forms, first, it is necessary to determine whether the original to be stored satisfies the conditions that hold in the first place and is legitimate. For example, a form that does not contain essential items is formed as a form, so there is no point in saving it. In addition, even if the required entry items are filled in, even if the entry details are incorrect, it does not make sense as a form. For example, in the case of a receipt, a receipt that is not filled in with an issuer, date, amount, etc., does not hold as a receipt, and therefore processing such as scanning and saving is useless. In addition, even in company documents, many cases can be considered, such as when the name column of the submitter is not filled in or when the total amount column of the estimate is not filled in.

  It is inefficient in terms of work cost and system maintenance cost to scan such a document and give it a time stamp or digital signature.

  In order to solve the problems caused by the above conventional example, the present invention comprises the following means.

  An image input unit for inputting an image from a scanner device, and a storage unit for storing the image in a document management system.

  Form registration means for recognizing images in forms.

  Condition area setting means for setting the condition area of a form registered form.

  Condition setting means for setting conditions for each condition area.

  Form recognition means for recognizing forms and identifying forms.

  Zone OCR means for reading out the condition area of the form specified by the form recognition means, cutting out the area image, and OCRing only the cut out partial image.

  Collating means for collating the result of the zone OCR means with the condition.

  Index storage means for temporarily storing the results of the zone OCR means.

  Collation result storage means for temporarily storing the result of determining the validity of the document based on the collation result.

  Storage processing determination means for determining a storage processing method according to the collation result.

  Index storage means for storing an index stored in the index storage means in association with an image in the document management system.

  As long as the present invention has the above-described configuration and each means, the present invention may have a configuration in which components are provided in a computer such as a single PC. Needless to say, it may be provided in the form.

  The input software section and document management system having the characteristic means of the present invention are constructed on a computer system as computer software.

As described above, according to the present invention,
(1) By giving a time stamp and an electronic signature only to a document formed as a form, there is an effect that the cost required for the time stamp can be reduced (cost reduction).

  (2) Since only originals formed as forms can be classified and managed by the document management system, there is an effect of improving operability when searching and browsing for a user's desired use.

  (3) By classifying only manuscripts formed as a form and managing them with a document management system, there is an effect of improving corporate compliance with respect to document storage.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing the basic configuration of the system of the present invention.

  1. The scanner device unit is a device that inputs an actual paper document and inputs image data.

  2. The input software unit inputs an image from the scanner device and performs the characteristic processing of the present invention.

  3. The document management system is a data storage unit that stores an original image and index data input from a scanner device.

  As an embodiment, as shown in FIG. 1, an input software unit and a document management system may be connected on a network so that data can be exchanged with each other, and each component is present in one PC. Even in this case, the present invention can be realized if functional elements and means are satisfied.

  FIG. 2 is a flowchart of overall processing according to the embodiment of the present invention.

  FIG. 3 is a conceptual diagram showing scan / save phase processing according to the embodiment of the present invention. This is for conceptually explaining the document validity determination process of the input software unit, which is a feature of the present invention.

  The reason for conceptually indicating that the OCR process and the input process of final determination by the user interface have so far been made with many technical proposals as algorithms, and therefore the process in the present invention is applied by such a new combination of conventional techniques. For the sake of convenience, FIG. 3 shows the concept of the process in order to show the technology and to explain the target process.

  FIG. 4 is a diagram showing the contents of condition setting for each condition area in the concept of a table.

  FIG. 5 shows an example of condition area setting for a document. A rectangular area is selected for a required entry portion as a manuscript on the image.

  Next, the processing of the embodiment of the present invention will be described in detail according to the flowchart of FIG.

  First, in step S101, initial setting of each function relating to the image input apparatus is performed. In S101, since the input software unit and the document management system of the present invention are provided by computer software, the computer program general initialization processing is performed, and the scanner device is ready to start scanning an image.

  In step S102, the scanner device scans the image and captures the image into the input software unit. This step S102 includes a state of waiting for the scanning process until there is a scan execution command through the user interphase. When there is an actual scan execution command, reading of an image from the scanner device is started. Then, the scanned image is taken into the input software unit.

  In step S103, it is determined whether the process in the input software unit is the “registration phase” or the “scan / save phase”. The instruction that is the basis of this determination is input via the user interface phase of the input software unit. In the “registration phase”, the process proceeds to step S104 in order to perform advance setting of the process for proving the validity of the form in the present invention. In the “scan / save phase”, the process proceeds to step S108.

  In step S104, the image input in step S102 is registered as a form. Forms to be registered are registered in a database for registering a plurality of forms such as a form DB. The form registered in the form DB is used for form recognition at the time of form recognition in the “scan / save phase”. When a form is registered in the form DB, the form is registered in a normalized data format obtained by block analysis, and a form name is assigned to each registered form.

  In step S105, a condition area is set. The condition area is a rectangular area on the image, and is a place where there is information necessary to determine the validity of the form. In the example of FIG. 5, the contents described in the positions A1 to A5 have a meaning as a form, and if there is no information at this position, the place where the manuscript cannot be formed as a form is displayed on the screen through the user interface phase. Set by specifying the rectangle position.

  Next, in step S106, the condition of the condition area set in S107 is set. In the example of FIG. 5, it is set via the user interface phase of the input software section whether the information from A1 to A5 on the image should be determined and how to determine the information. It is recorded in the condition table shown in the example of 4.

  Next, in step S107, if the condition area setting and whether or not to end the condition setting are not determined and the setting is continued, the process returns to step S105 to set another condition area on the form and its condition. To do. As described above, a plurality of condition areas and their conditions can be set on one form by repeating the processes in steps S105 to S107.

  In the example of A1 in FIG. 5, the condition collation method is “data presence / absence”, which is to determine that some information should be entered in the condition area.

  In the example of the condition setting of A2 in FIG. 5, a date range is designated as the collation method, and the minimum value setting and the maximum value setting are recorded in the condition table as the condition values. This is for determining whether or not the value is between the minimum value and the maximum value in the condition set value.

  Further, in the example of the condition setting “40000” in A3 of FIG. 5, a numerical value range is designated as the collation method. As the condition value, the minimum value setting and the maximum value setting are recorded in the condition table. This is for determining whether or not the value is between the minimum value and the maximum value in the condition set value.

  In other words, in the example of FIGS. 5 and 4, the form with the form name “Receipt 1” has five condition areas A1 to A5 as the condition areas in the form DB in the form DB, and each condition area is valid. The condition setting at the time of collation is made.

  Next, when the form registration, condition area setting, and condition setting are completed in step S107, the process ends in the “registration phase”, returns to step S102, and waits for the next scan execution.

  If it is determined in step S103 that the process is the “scan / save phase”, the process proceeds to step S108.

  In step S108, a form recognition process is performed on the image scanned in S102. The form recognition process is a process for specifying the form by normalizing the image input in S102 and referring to the normalized data of the form registered in the form DB.

  Next, in step S109, it is determined whether the form is registered in the form DB as a result of the form recognition in S108. If the image has been registered in the form, the process proceeds to step S110. If the image is not registered in the form, the image legitimacy determination process, which is a feature of the present invention, cannot be performed, and the process advances to step S122.

  In step S110, the condition area setting information is read one by one from the condition table shown in FIG. 4 which is stored in association with the form registered in the form DB. The setting in the condition area column in FIG. 4 is the condition area. In the example of FIG. 4, (sx1, sy1) (ex1, ey1) is first read. (Sx1, sy1) (ex1, ey1) indicates the coordinate position from the origin on the image. Since the condition area is a rectangular area on the image, (sx1, sy1) indicates the coordinates of the upper left corner of the rectangle. (ex1, ey1) indicates the coordinates of the lower right corner of the rectangle. Sx1 and ex1 indicate the position of the x coordinate. Sy1, ey1 indicates the position of the y coordinate.

  In step S111, the condition setting is read from the condition table.

  In the example of FIG. 4, the collation method is “data presence / absence” under the condition that the ID is indicated by F1-1. In this case, there is no condition setting value as long as there is data.

  Next, in step S112, a partial image is cut out from the image input in S102 according to the contents of the condition area setting read in S110.

  In step S113, OCR processing is performed on the partial image cut out in step S112.

  In step S114, the OCR result in S113 is temporarily stored as index data.

  In step S115, the OCR result in S112 / S113 is compared with the condition read in S111. The condition example of ID “F1-1” in FIG. 4 and FIG. 5 will be described. First, in step 112, a rectangle A1 in FIG. 5 is cut out. The cut image is subjected to OCR processing. Since there is no character information in the rectangle, the OCR result is empty. Next, in step S115, a check is performed against the condition. At this time, since the check method is “data presence / absence”, if the result is an empty OCR result, the check result of the ID “F1-1” is “no”.

  Next, in S116, it is determined whether or not all conditions for the form have been collated. When all the collations are completed, the process proceeds to step S117. When all the collations have been completed, the process returns to step S110 to perform processing up to the collation of the next condition. In the area setting example shown in FIG. 5, the processing from S110 to S115 is repeated for A1 to A5 with reference to the condition table of FIG.

  In step S117, the collation result of the image and all the conditions is displayed on the screen. As an example, the state shown in FIG. 3 is a full screen display, that is, an image and a condition area rectangle on the image are displayed. The line color is displayed. The display is displayed on the display of a computer having an input software section.

  In step S118, a final determination screen is displayed. An example is shown in the “Condition Confirmation” screen of FIG. The entire image display in S117 is a result of matching conditions based on the OCR result, and is not exactly accurate. Therefore, a “Condition Confirmation” screen for final determination is displayed in S118. On this condition confirmation screen, the result of acceptance or rejection by the user input is accepted. In the example of the condition confirmation screen in FIG. 3, the final determination result is “OK” if “OK” is checked, and “No” if “NG” is checked.

  In step S119, the pass / fail result by the user input on the condition confirmation screen is temporarily stored.

  Next, in step S120, it is determined whether or not the final determination screen display and the final determination result are stored for all conditions. If it has been performed for all conditions, the process proceeds to S121. If not yet performed for all conditions, the process returns to S118, and the final determination screen display for the next condition, the final determination input from the user, and the final determination result for each condition are repeatedly stored.

  Next, in step S121, the final determination result of all conditions stored in the repetition process of S118 to S120 is evaluated. If the final determination result of all the conditions is “go”, the process proceeds to S123. If there is at least one “No” in the final determination result of all the conditions, it is assumed that it does not hold as a form, and the process proceeds to S122.

  In step S122, the input image is stored in another storage location of the document management system.

  In step S123, a time stamp is given to the image scanned in S102.

  In step S124, an electronic signature is added to the image scanned in S102.

  The present invention is characterized in that S123 and S124 are processed when all the conditions are “go” in S121. However, the process of adding a time stamp to an image and applying an electronic signature is a general technique and is not a feature of the present invention, and thus detailed description thereof is omitted.

  In step S125, the index data is stored in the document management system in which the image input in step S102 is given a time stamp and an electronic signature and the OCR result stored in step S114.

  Finally, in S126, it is determined whether or not the processing of the entire image input apparatus is to be ended. If not completed, the process returns to S102 and waits for image scanning. When the process is to be ended, all processes are ended after performing a different software end process.

It is a system configuration block diagram concerning a 1st embodiment of the present invention. It is a flowchart figure of the whole process which concerns on embodiment of this invention. It is a conceptual diagram of the scan / save phase process according to the embodiment of the present invention. It is a key map of a condition table concerning an embodiment of the invention. It is a figure which shows the condition area setting of the registration phase based on embodiment of this invention.

Claims (4)

An image input apparatus according to the present invention includes a scanner device that performs image input, an input software unit that captures an image from the scanner device, and an image input by a document management system that stores an image read by the scanner together with an index and further searches and browses stored documents. In the device
As a process of the input software section, there is a method for evaluating the validity of the input document. The condition area on the image for the required entry items of the form is read from the image information, and the image information is converted into character information by OCR (optical character recognition). Convert and refer to the character information in the condition DB registered for each form in advance. Only when all the conditions in one form are met, the input image is judged to be valid and registered in the document management system. It has a means of judging the validity of the image,
In order to determine the validity of the input image, first, form registration means for automatically recognizing the document type, condition area setting for each form, and conditions for each condition area are set in association with each other. A series of registration phase means,
In addition, the input software unit includes a scan / save phase means for inputting a picture, judging the validity, and saving a series of processes in the document management system.
In the scan / save phase, first, an image is input from the scanner using the input software section, and the "condition area" setting for the form that matches the form registered in the form registered automatically and the "condition" setting for each condition area are set. A condition reading means for reading,
Next, automatic condition matching means for cutting out the “condition area” image of the matched form, subjecting the partial image to OCR processing, and matching the character string of the OCR result with “condition”,
Furthermore, the result of the verification by the automatic previous verification means is displayed in a rectangle on the image as a condition area when the input image is displayed on the screen. It has a matching result display means for displaying,
Furthermore, a final judgment result input means by a user interface for the operator to finally judge the collation result by the automatic collation means,
Based on the contents entered by the final judgment result input means, if the final judgment results are all “match”, the image is saved in a predetermined storage location in the document management system, and “does not match” is included. If so, an image input device comprising image storage means for storing in another storage location in the document management system. The present invention relates to claim 1, and when the “condition” is set in the registration phase process, the condition verification method is set for the verification process in the verification process in the scan / save phase. Set the conditions such as data match, presence / absence of data, within the set condition range as the condition matching method,
When judging the validity of the image input from the scanner in the scan / save phase, the image is input from the scanner in the input software section, and the "Condition area" of the form that matches the form registered in the form registered automatically is recognized. “Condition” setting for each condition area is read out, then the “condition area” image of the matched form is cut out, the partial image is subjected to OCR processing, the character string of the OCR result is collated with “condition”, and collation In this case, the image input apparatus is characterized in that collation is performed according to the condition for each condition area set in the “condition” setting.   In the document storage in the scan / save phase of claim 1, if all conditions are “matched” by the final result of the validity judgment, the image is given a time stamp and an electronic signature and stored in the document management system. An image input device.   In the document storage in the scan / save phase according to claim 1, the OCR result for each condition area is temporarily stored in the input software unit at the time of validity judgment, and the OCR result is stored in the document management system when the image is stored. An image input apparatus that stores data as index data (attribute data associated with an image) in a document management system.

Images