JP2010026987A - Network document management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the load of retrieval by user's final visual observation by narrowing down first similar business form images in a system for storing a log while associating the log with a printed image and performing image retrieval to acquire a desired log. <P>SOLUTION: The system for associating an image with a print log to manage them includes: a means for extracting the name of a business form template, data inherent to the business form, and the coordinates of the data; a means for transmitting the same information to a retrieval server; and a means for printing the business form template. Especially, the retrieval server includes a means for collating the data inherent to the business form with data obtained by recognizing the image to specify the image. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to history management, and more particularly to a network document management system for efficiently storing and managing history information, image data, and text data in a document management server in an image processing apparatus.

  With the widespread use of digital multi-function peripherals, anyone can easily print, copy, and transmit originals. However, while these improve convenience for users, new information leakage problems such as printing, copying, and transmission of confidential manuscripts have arisen. As a countermeasure against such a problem, there is a document management system that stores all read image data and text data in a storage device during printing, copying, FAX, e-mail transmission, and the like. This makes it possible to track when, where, and what processing has been performed by the administrator by checking the stored data to determine which printer or digital multi-function peripheral has processed the document with the leaked information. There is an effect.

On the other hand, in the process of investigating the associated “when, where, and what” by picking up printed matter and performing image search using an image reader, the image at a specific position in the image data is obtained by OCR. A technique for identifying an image, that is, when the image was printed “when, where, and how” by collating a result extracted as a character string with a character string associated in advance at the time of image registration is known. In order to improve accuracy, there is a technique for increasing the hit rate by associating a plurality of character strings with an image in consideration of erroneous recognition by OCR. For example, Patent Document 1 is a representative example.
JP 10-31727 A

  In the case of an image in which variable data is superimposed on a specific template depending on the business, such as a form, there are not a few cases in which image data with almost no difference in variable data is accumulated. For example, it is conceivable that a quote for a specific product is printed many times by a specific person regardless of the date, and only the character strings representing the created date and time are different. And the estimate image having the characteristic that the images are the same. When there are many similar images and a desired image is searched using only the characteristics of the image as a clue, many images are presented as candidates, and it is extremely difficult to finally visually check the matching of the images. Become.

  In this situation, the prior art has the effect of increasing the probability of hitting the desired image data, but also has the effect of increasing the probability of hitting other images at the same time. The issue of need remains. Even in order to obtain the effect of narrowing down by increasing the number of places on the printed matter that are read by OCR, it is necessary to register the position information for each form image in advance, which is troublesome.

In order to solve the above-described problems, the present invention is a system in which content data (image data / text data) of a job of a printer or digital multifunction peripheral is stored in a server together with log information, and the job can be traced later.
For all image data finally stored in the server, related to the form generation operation, for example, the information that each image data has uniquely, such as the date when the form was generated, is extracted as text data. Information extraction means;
Unique text position extracting means for extracting the position on the image data of the text obtained by the form generation operation related information extracting means;
Template information printing means for printing information about the form template that is the source of the image data at a fixed position of the image data;
Associating the text obtained by the form generation operation related text information extracting means, the position information obtained by the unique text position extracting means, the original image data, and the information of the form template from which the image data is based Association means to keep
The template information printed by the template information printing means, which is a part of the result of reading the printed matter by the image reading device, and the corresponding form-specific text information (referred to as first text information) from the association means, The position of the first text information on the image data is extracted, and further, the printed material is read by the image reading device, and is the same position as the position of the first text information on the image data. Extracting text information corresponding to the first text information (assumed to be second text information), and specifying image data that matches the first text information and the second text information, image data It has a specific means.

  According to the present invention, it is possible to narrow down a desired image from similar form images, that is, when, where and how the image was printed.

  Next, details of the present invention will be described in accordance with the description of the embodiments.

  FIG. 1 is a diagram showing a system configuration according to an embodiment of the present invention. Here, a digital multi-function peripheral 101, an image processing server 102, a data server 103, a user PC 104, a printer driver 105, a print server 106, a printer 107, and an image search server 108 are connected to the network.

  The digital multifunction peripheral 101 has a scanner / printer function, an application execution environment, and the like. The digital multi-function peripheral 101 generates electronic image data of the job at the same time as the job for which the form generation is executed. The digital multi-function peripheral 101 transfers the image data to the image processing server 102 immediately or once after storing it in its storage device.

  The user PC 104 is a general client PC. When printing is performed on the user PC 104, the printer driver 105 simultaneously generates electronic image data. At this time, the printer driver 105 may simultaneously extract the character string sent as the character drawing command as the text data of the execution job. In this case, the image data and the text data are associated with each other. The user PC 104 immediately or temporarily saves the data in its own storage device or the storage device of the print server 106, and then transfers it to the image processing server 102.

  The image processing server 102 performs image conversion processing on the transferred image data in a format that can be stored in the data server 103 and transfers the image data to the data server 103.

  A data server that stores the form-specific text data, the same coordinates, and the form template name in association with image data, which is characteristic processing of the present invention, is located at 103, and is desired based on the template information extracted by OCR. An image search server for extracting the image data is located at 108.

  Information required for performing a search in the search server (108) when generating a form in the digital multi-function peripheral (101) or the PC (104) using FIG. 2, FIG. 3, FIG. 4, and FIG. The process of extracting, printing, and registering to the data server (103) will be described.

  S201 represents a form. S202 and S203 represent fields on the form template in which the same or different values for both forms are inserted. S204 represents a field that is known in advance to be inserted with a form-specific value related to the form generation operation. S205 is a field for printing information for specifying a form template, and indicates that the same name is printed as a value. S206 and S207 represent the position of S204, that is, the print position of the value specific to the form.

  S301 represents a form. The difference from S201 is only the value of S304 corresponding to S204. S302 corresponds to S202, S303 corresponds to S203, S305 corresponds to S205, S306 corresponds to S206, and S307 corresponds to S207. FIG. 4 shows a management table on the data server to which these pieces of information are sent. The form in FIG. 2 corresponds to “01” in S401, and the form in FIG. 3 corresponds to “002” in S401. S402 in S205 and S305, S403 in S204 and S304, S404 in the specific coordinates of S206 and S207, and S306 and S307 in specific coordinates, S405, the value inserted in S204, and inserted in S304 Each corresponds to a value. S406 is information about the “when, where, and how” the print is actually desired, but since it is not directly related to the feature of the present proposal, a detailed description is omitted.

  FIG. 5 is a flowchart of processing for registration in the data server (103). First, processing is started (S501). Receive necessary information (S502), determine whether it is already registered form template (S503), register the template name if it is the first received template not registered, the name of the field having a form-specific value, Associate coordinates (S504). In the example of FIGS. 2, 3, and 4, there is only one field, but a plurality of fields may generally exist in one form. Finally, the template name and the form-specific value are stored in association with the image data (S505).

  A process of narrowing down to a desired form at the time of image search will be described with reference to FIG.

  Search processing is started (S601). The printed material is applied to the image reading apparatus, and the name of the form template printed at the fixed position is extracted by applying the OCR (S602). It is determined whether or not the form template has already been registered (S603). If there is no registration, the process is terminated. If registered, the name, coordinates, and form-specific value of the field to which the form-specific value is registered are extracted (S604), and OCR is extracted for the extracted coordinate position of the scanned image. Is used to extract the form-specific data and compare it with the value obtained in S604 (S605). In S604, values related to a plurality of images are generally extracted. In S606, an image that matches a form-specific value is identified, a print log that is the original purpose of the system is presented to the searcher, and the process ends.

System Configuration Form example 1 Form example 2 Form image management table on the data server Data server registration flowchart Image search flowchart

Claims (2)

The content data (image data / text data) of the job of the printer or digital multi-function peripheral can be saved on the server together with log information, and the job can be traced later. The job is composed of the form template and the form data. In the resulting network document management system,
For all image data finally stored in the server, a form generation operation related information extracting means for extracting, as text data, information inherent to each image data associated with the form generation operation;
Unique text position extracting means for extracting the position on the image data of the text obtained by the form generation operation related information extracting means;
Template information printing means for printing information about the form template that is the source of the image data at a fixed position of the image data;
Associating the text obtained by the form generation operation related text information extracting means, the position information obtained by the unique text position extracting means, the original image data, and the information of the form template from which the image data is based And a network document management system characterized by comprising association means.   The template information printed by the template information printing means, which is a part of the result of reading the printed matter by the image reading device, and the corresponding form-specific text information (referred to as first text information) from the association means, The position of the first text information on the image data is extracted, and further, the printed material is read by the image reading device, and is the same position as the position of the first text information on the image data. Extracting text information corresponding to the first text information (assumed as the second text information), and specifying image data that matches the first text information and the second text information, image data The network document management system according to claim 1, further comprising a specifying unit.

Images