JP2010033605A - Information processor and information processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform a digitization business of a paper document by which increase of demand is expected from now on with high accuracy and at high speed by using a plurality of different kinds of character recognition servers. <P>SOLUTION: When an image to be an object for document recognition processing is received from a client, a document recognition server suitable for performing document recognition processing of the received image is selected among the plurality of document recognition servers to request the document recognition processing, data which are made into electronic information after the document recognition processing from the document recognition server to which the document recognition processing is requested, are received to be provided to the client. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a document input proxy method that performs a document recognition task between a document recognition server that performs document recognition processing based on an image and a client who requests the document recognition server to perform document recognition processing via a network, and a document The present invention relates to an input proxy computer program and a document input proxy device.

  In recent years, systems for inputting documents via a network have been developed. For example, Patent Document 1 proposes an information processing apparatus that can be freely used in an office, a business trip destination, or the like.

  However, there is no prior art regarding a method for inputting a document using a plurality of document recognition servers via a network.

  As an example similar to the method using a plurality of document recognition servers, Patent Document 2 describes a method using a plurality of character recognition engines. The method described in Patent Document 2 is to increase the processing speed by using different recognition calculation means for each recognition target area after area identification.

  However, the above-described methods using a plurality of character recognition engines are mainly aimed at speeding up the processing, and since each method prepares and processes a plurality of the same character recognition engines, the recognition performance is improved. I can't hope. In addition, no proposal has been made as a document input proxy system.

  The present invention has been made in view of the above, and has as its first object to carry out electronic work of paper documents, for which demand is expected to increase in the future, with high accuracy and high speed.

  A second object of the present invention is to provide an optimal document input environment for each object to be digitized.

  In order to solve the above-described problems and achieve the object, the invention according to claim 1 is a first step of receiving an image to be subjected to document recognition processing from a requester, and the reception among a plurality of document recognition servers. A second step of requesting a document recognition process by selecting an appropriate document recognition server for document recognition processing of an image, and receiving electronically converted data after the document recognition process from the requested document recognition server And a third step of providing the client with the requester.

  The invention according to claim 2 is the document input proxy method according to claim 1, wherein the second step includes a process of obtaining a selection criterion of the document recognition server from a requester.

  According to a third aspect of the present invention, in the second step, after a given image is presented to a plurality of the document recognition servers and processing conditions are collected from each of the document recognition servers, the most appropriate server is selected. The document input proxy method according to claim 1, further comprising a selection process.

  According to a fourth aspect of the present invention, the process for selecting the most appropriate server is determined based on the requester's selection by prompting the client to select the process conditions collected from the document recognition server. The document input proxy method according to claim 3, further comprising a process of requesting the document recognition server to perform a document recognition process.

  The invention according to claim 5 is the document input proxy method according to claim 1 or 2, wherein a fee required for document recognition is used as the selection criterion.

  A sixth aspect of the present invention is the document input proxy method according to the first, third or fourth aspect, wherein a fee required for document recognition is used as the processing condition.

  The invention according to claim 7 is the document input proxy method according to claim 1 or 2, wherein a time required for document recognition is used as the selection criterion.

  The invention according to claim 8 is the document input proxy method according to claim 1, 3 or 4, wherein a time required for document recognition is used as the processing condition.

  The invention described in claim 9 is the document input proxy method according to claim 1 or 2, wherein the quality of the document recognition result is used as the selection criterion.

  A tenth aspect of the present invention is the document input proxy method according to the first, third or fourth aspect, wherein the quality of the document recognition result is used as the processing condition.

  The invention described in claim 11 uses at least one of the congestion level of the document recognition server and the congestion level of a network for accessing the document recognition server as the selection criterion. It is an alternative method.

  According to a twelfth aspect of the present invention, at least one of a congestion degree of the document recognition server and a congestion degree of a network for accessing the document recognition server is used as the processing condition. This is a document input proxy method.

  The invention according to claim 13 is the document input proxy method according to claim 1 or 2, wherein a language supported by the document recognition server is used as the selection criterion.

  The invention described in claim 14 is the document input proxy method according to claim 1, 3 or 4, wherein a language supported by the document recognition server is used as the processing condition.

  A fifteenth aspect of the invention is the document input proxy method according to the first or second aspect, wherein an input image format supported by the document recognition server is used as the selection criterion.

  The invention described in claim 16 is the document input proxy method according to claim 1, 3 or 4, wherein an input image format supported by the document recognition server is used as the processing condition.

  The invention according to claim 17 is the document input proxy method according to claim 1 or 2, wherein an output format supported by the document recognition server is used as the selection criterion.

  The invention according to claim 18 is the document input proxy method according to claim 1, 3 or 4, wherein an output format supported by the document recognition server is used as the processing condition.

  According to a nineteenth aspect of the present invention, the second step includes a process of performing an analysis process on a given image and selecting an appropriate document recognition server from a plurality of document recognition servers based on the analysis result. The document input proxy method according to claim 1, 3 or 4.

  The invention according to claim 20 is the document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a connected rectangular component in a given image.

  The invention according to claim 21 is the document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a character size in a given image.

  The invention according to claim 22 is the document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing the number of characters in a given image.

  The invention according to claim 23 is the document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a quality of a given image.

  The invention described in claim 24 is the document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a layout in a given image.

  25. The document input proxy method according to claim 24, wherein the second step includes a process of sending only a character region obtained as a result of the process of analyzing the layout in the image to the document recognition server. It is.

  According to a twenty-sixth aspect of the present invention, the second step includes a process of selecting the document recognition server for each area according to the attribute of each area obtained as a result of the process of analyzing the layout in the image. 25. The document input proxy method according to claim 24.

  According to a twenty-seventh aspect of the present invention, in the second step, the given image is transmitted to the candidate document recognition server, and the electronic information data after the document recognition processing is transmitted from the document recognition server. 27. The document input proxy method according to claim 1, further comprising a process of receiving and synthesizing the received document recognition processing results and presenting the result to the client.

  A twenty-eighth aspect of the present invention is the document input proxy method according to the twenty-seventh aspect, wherein the received result of the document recognition process is converted into a state that is not converted into electronic information and presented to the client.

  The invention according to claim 29 is the document input proxy method according to any one of claims 1 to 28, wherein the third step includes a process of charging a requester for a compensation for the document recognition server selection.

  30. The document input proxy according to claim 1, wherein the third step includes a process of charging an administrator of the document recognition server for the value of the document recognition server selection. Is the method.

  In the invention described in claim 31, in the third step, the client selects the advertisement selected in accordance with the content of the image requested for the document recognition process by the client along with the electronic information data after the document recognition process. The document according to any one of claims 1 to 30, further comprising a fourth process of charging a fee for selection of an advertisement to an advertiser of the advertisement associated with the electronic information data. This is an input proxy method.

  According to a thirty-second aspect of the present invention, there is provided a first function that is installed in a computer and receives an image to be a document recognition process target from a client via a network, and a document recognition process for the received image among a plurality of document recognition servers. A second function for requesting a document recognition process by selecting an appropriate document recognition server to receive the request, and a requester for receiving electronically converted data after the document recognition process from the requested document recognition server A computer program for proxying document input that causes a computer to execute the third function provided in the above.

  According to a thirty-third aspect of the present invention, there is provided the computer program for proxying a document input according to the thirty-second aspect, wherein the second function includes a process of obtaining a selection criterion of the document recognition server from a requester.

  In the invention according to claim 34, the second function is to display a given image to a plurality of document recognition servers, collect processing conditions from each of the document recognition servers, and then select a most suitable server. 33. A computer program for proxying document input according to claim 32, including processing for selecting.

  According to a thirty-fifth aspect of the present invention, the process of selecting the most appropriate server is determined based on the requester's selection by prompting the client to select the process conditions collected from the document recognition server. 35. The computer program for proxying document input according to claim 34, including a process of requesting a document recognition process from the document recognition server.

  A thirty-sixth aspect of the invention is a computer program for proxying a document input according to the thirty-second or thirty-third aspect, wherein a fee required for document recognition is used as the selection criterion.

  A thirty-seventh aspect of the invention is the computer program for proxying a document input according to the thirty-second, thirty-four, or thirty-fifth aspect using a fee required for document recognition as the processing condition.

  A thirty-eighth aspect of the invention is a computer program for proxying a document input according to the thirty-second or thirty-third aspect, wherein a time required for document recognition is used as the selection criterion.

  A thirty-ninth aspect of the invention is a computer program for proxying a document input according to the thirty-second, thirty-four, or thirty-fifth aspect, wherein a time required for document recognition is used as the processing condition.

  The invention described in claim 40 is the computer program for proxying document input according to claim 32 or 33, wherein the quality of the document recognition result is used as the selection criterion.

  The invention described in claim 41 is the computer program for document input substitution according to claim 32, 34 or 35, wherein the quality of the document recognition result is used as the processing condition.

  The invention according to claim 42 is the document input according to claim 32 or 33, wherein at least one of the congestion level of the document recognition server and the congestion level of the network for accessing the document recognition server is used as the selection criterion. It is a computer program for substitution.

  The invention of claim 43 uses at least one of the congestion level of the document recognition server and the congestion level of the network for accessing the document recognition server as the processing condition. This is a computer program for document input agency.

  A 44th aspect of the present invention is the computer program for proxying a document input according to the 32nd or 33rd aspect, wherein a language supported by the document recognition server is used as the selection criterion.

  A 45th aspect of the present invention is the computer program for proxying a document input according to the 32nd, 34th or 35th aspect, wherein a language supported by the document recognition server is used as the processing condition.

  The invention described in claim 46 is the computer program for proxying document input according to claim 32 or 33, wherein the input image format supported by the document recognition server is used as the selection criterion.

  A 47th aspect of the present invention is the document input substitution computer program according to the 32nd, 34th or 35th aspect, wherein an input image format supported by the document recognition server is used as the processing condition.

  A 48th aspect of the present invention is the computer program for proxying a document input according to the 32nd or 33rd aspect, wherein an output format supported by the document recognition server is used as the selection criterion.

  A 49th aspect of the present invention is the document input substitution computer program according to the 32nd, 34th or 35th aspect, wherein an output format supported by the document recognition server is used as the processing condition.

  According to a 50th aspect of the present invention, the second function includes a process of performing an analysis process on a given image and selecting an appropriate document recognition server from a plurality of document recognition servers based on the analysis result. 36. A computer program for proxying document input according to claim 32, 34 or 35.

  The invention according to claim 51 is the computer program for proxying document input according to claim 50, wherein the analysis process includes a process of analyzing a connected rectangular component in a given image.

  The invention according to a 52nd aspect is the computer program for proxying a document input according to the 50th aspect, wherein the analyzing process includes a process of analyzing a character size in a given image.

  The invention according to claim 53 is the computer program for proxying document input according to claim 50, wherein the analysis processing includes processing for analyzing the number of characters in a given image.

  The invention according to claim 54 is the computer program for proxying document input according to claim 50, wherein the analysis processing includes processing for analyzing quality of a given image.

  The invention according to claim 55 is the computer program for document input substitution according to claim 50, wherein the analysis process includes a process of analyzing a layout in a given image.

  56. The document input proxy according to claim 55, wherein the second function includes a process of sending only a character region obtained as a result of the process of analyzing the layout in the image to the document recognition server. It is a computer program.

  The invention according to claim 57 is characterized in that the second function performs a process of selecting the document recognition server for each area according to an attribute of each area obtained as a result of the process of analyzing the layout in the image. 56. A computer program for proxying document input according to claim 55.

  The invention according to claim 58 is characterized in that the second function transmits an image given to the candidate document recognition server, and converts the electronic information data after the document recognition processing from the document recognition server. 58. A computer program for proxying a document input according to any one of claims 32 to 57, including a process of receiving and synthesizing the received document recognition results and presenting them to a client.

  The invention according to claim 59 is the computer program for proxying document input according to claim 58, wherein the received result of the document recognition process is converted into a state that is not converted into electronic information and presented to the client.

  The invention according to claim 60 is the computer program for proxying a document input according to any one of claims 32 to 59, wherein the third function includes a process of charging a requester for a fee for selecting the document recognition server. is there.

  61. The document input proxy according to claim 32, wherein the third function includes a process of charging an administrator of the document recognition server for the document recognition server selection. It is a computer program for.

  According to the invention of claim 62, the third function is that the client selects the advertisement selected in accordance with the content of the image requested by the client for the document recognition process together with the electronic information data after the document recognition process. 62. The document according to any one of claims 32 to 61, further comprising: a fourth process of charging a consideration for selection of an advertisement to an advertiser of the advertisement associated with the electronic information data. It is a computer program for input substitution.

  The invention according to claim 63 is suitable for firstly receiving an image to be subjected to document recognition processing from a client via a network, and suitable for document recognition processing of the received image among a plurality of document recognition servers. A second means for selecting the document recognition server and requesting a document recognition process; and a third means for receiving electronically converted data after the document recognition process from the requested document recognition server and providing the data to the requester. And a document input proxy device.

  The invention according to claim 64 is the document input proxy device according to claim 63, wherein the second means includes processing for obtaining a selection criterion of the document recognition server from a client.

  According to the invention of claim 65, the second means presents a given image to a plurality of the document recognition servers, collects processing conditions from each of the document recognition servers, and then selects the most suitable server. 64. A document input proxy device according to claim 63, comprising a selection process.

  According to a 66th aspect of the present invention, the process of selecting the most appropriate server is determined based on the requester's selection by presenting the requester with the process conditions collected from the document recognition server. 66. The document input proxy device according to claim 65, further comprising a process of requesting the document recognition server to perform a document recognition process.

  The invention according to claim 67 is the document input proxy device according to claim 63 or 64, wherein a fee required for document recognition is used as the selection criterion.

  The invention according to claim 68 is the document input proxy device according to claim 63, 65 or 66, wherein a fee required for document recognition is used as the processing condition.

  The invention according to claim 69 is the document input proxy device according to claim 63 or 64, wherein a time required for document recognition is used as the selection criterion.

  The invention according to claim 70 is the document input proxy device according to claim 63, 65 or 66, wherein a time required for document recognition is used as the processing condition.

  The invention according to claim 71 is the document input proxy device according to claim 63 or 64, wherein the quality of the document recognition result is used as the selection criterion.

  The invention according to claim 72 is the document input proxy device according to claim 63, 65 or 66, wherein the quality of the document recognition result is used as the processing condition.

  73. The document input according to claim 63 or 64, wherein at least one of the congestion level of the document recognition server and the congestion level of a network for accessing the document recognition server is used as the selection criterion. It is a substitute device.

  The invention according to claim 74 is characterized in that at least one of the congestion degree of the document recognition server and the congestion degree of the network for accessing the document recognition server is used as the processing condition. This is a document input proxy device.

  The invention described in claim 75 is the document input proxy device according to claim 63 or 64, wherein a language supported by the document recognition server is used as the selection criterion.

  The invention according to claim 76 is the document input proxy device according to claim 63, 65 or 66, wherein a language supported by the document recognition server is used as the processing condition.

  The invention according to claim 77 is the document input proxy device according to claim 63 or 64, wherein an input image format supported by the document recognition server is used as the selection criterion.

  A 78th aspect of the present invention is the document input proxy device according to the 63rd, 65th or 66th aspect, wherein an input image format supported by the document recognition server is used as the processing condition.

  The invention according to claim 79 is the document input proxy device according to claim 63 or 64, wherein an output format supported by the document recognition server is used as the selection criterion.

  The invention according to claim 80 is the document input proxy device according to claim 63, 65 or 66, wherein an output format supported by the document recognition server is used as the processing condition.

  The invention according to claim 81, wherein the second means includes a process of performing an analysis process on a given image and selecting an appropriate document recognition server from a plurality of document recognition servers based on the analysis result. A document input proxy device according to claim 63, 65 or 66.

  The invention according to claim 82 is the document input proxy device according to claim 81, wherein the analysis process includes a process of analyzing a connected rectangular component in a given image.

  The invention according to claim 83 is the document input proxy device according to claim 81, wherein the analysis process includes a process of analyzing a character size in a given image.

  The invention according to claim 84 is the document input proxy device according to claim 81, wherein the analysis processing includes processing for analyzing the number of characters in a given image.

  The invention according to claim 85 is the document input proxy device according to claim 81, wherein the analysis process includes a process of analyzing a quality of a given image.

  The invention according to claim 86 is the document input proxy device according to claim 81, wherein the analysis process includes a process of analyzing a layout in a given image.

  87. The document input proxy device according to claim 86, wherein the second means includes a process of sending only a character area obtained as a result of the process of analyzing the layout in the image to the document recognition server. It is.

  The invention according to claim 88 is characterized in that the second means performs a process of selecting the document recognition server for each area according to the attribute of each area obtained as a result of the process of analyzing the layout in the image. 90. A document input proxy device according to claim 86.

  In the invention according to claim 89, the second means transmits an image given to the document recognition server that is a candidate, and the electronic information data after the document recognition processing is transmitted from the document recognition server. The document input proxy device according to any one of claims 63 to 88, further comprising a process of receiving and synthesizing the received document recognition processing result and presenting the result to the client.

  The invention according to claim 90 is the document input proxy device according to claim 89, wherein the received result of the document recognition process is converted into a state that is not converted into electronic information and presented to the client.

  The invention according to claim 91 is the document input proxy device according to any one of claims 63 to 90, wherein the third means includes a process of charging a requester for the document recognition server selection.

  92. The document input proxy according to claim 63, wherein the third means includes a process of charging an administrator of the document recognition server for the value of the document recognition server selection. Device.

  The invention according to claim 93 is characterized in that the third means is configured such that the advertisement selected according to the contents of the image requested by the client for the document recognition process is displayed together with the electronic information data after the document recognition process. 93. A document according to any one of claims 63 to 92, further comprising a fourth process of charging a consideration for selection of an advertisement to an advertiser of the advertisement associated with the electronic information data. It is an input proxy device.

  According to the present invention, there is an effect that it is possible to carry out digitization of paper documents, which is expected to increase in demand in the future, with high accuracy and high speed.

  Further, according to the present invention, there is an effect that an optimal document input environment can be provided for each object to be digitized.

FIG. 1 is a schematic diagram showing a schematic configuration of the system. FIG. 2 is a schematic diagram showing an example of a selection criterion of a document recognition server by a user as the first embodiment of the present invention. FIG. 3 is a schematic diagram illustrating an example of processing conditions of the document recognition server. FIG. 4 is a schematic diagram showing another example of selection criteria for the document recognition server by the user as the second embodiment of the present invention. FIG. 5 is a schematic diagram showing another example of processing conditions of the document recognition server. FIG. 6 is a schematic diagram illustrating an example of a document image sent by the user to the service providing apparatus. FIG. 7 is a schematic diagram illustrating an example of an image presented to the user after the document recognition process is performed by the document recognition server on the document image sent by the user and the image is converted again by the service providing apparatus. FIG. 8 is a schematic diagram showing another example of an image presented to the user after the document recognition process is performed by the document recognition server on the document image sent by the user and the image is converted again by the service providing apparatus. FIG. 9 is a schematic diagram illustrating another example of an image presented to the user after the document recognition process is performed by the document recognition server on the document image sent by the user and the image is converted again by the service providing apparatus. FIG. 10 is a schematic diagram showing a characteristic example of the document recognition server as the third exemplary embodiment of the present invention. FIG. 11 is a schematic diagram illustrating an example of a document image analysis result. FIG. 12 is a schematic diagram illustrating an example of an image including a document, a table, and a photograph sent by the user to the service providing apparatus as the fourth exemplary embodiment of the present invention. FIG. 13 is a schematic diagram illustrating an example of a document image analysis result. FIG. 14 is a schematic diagram illustrating an example of characteristics of the document recognition server. FIG. 15 is a schematic diagram showing a schematic configuration of a system as a fifth embodiment of the present invention. FIG. 16 is a schematic view illustrating the contents of an advertisement database (advertisement DB).

  Exemplary embodiments of a document input proxy method, a document input proxy computer program, and a document input proxy device according to the present invention will be explained below in detail with reference to the accompanying drawings.

  A schematic configuration constituting the embodiment of the present invention will be described with reference to FIG.

  A system configuration example is shown in FIG. In this system, a user user terminal handled by a user, a service providing apparatus managed by a document input proxy service provider, and a plurality of document recognition servers are connected to be accessible via a communication line (for example, the Internet). Yes.

  Each of the user device, the service providing device, and the document recognition server has an architecture as a computer, and a computer program necessary for realizing each function is installed in a storage area of each device. In this case, the hardware program and the software program, which is a software resource, are connected by a general relationship similar to the mutual relationship between a general computer and an application program, and there is no particularity there. . Therefore, hereinafter, the function itself executed by the processor of the user device, the service providing device, and the document recognition server that have a computer architecture according to the computer program will be described.

  The basic operation will be described next.

  First, a user who requests a document input operation converts a document on paper into a document image using an image input device such as a scanner or a digital camera.

  Next, the user sends the converted image to the service providing apparatus via the communication line.

  In the service providing apparatus, the document image is digitized using an appropriate document recognition server based on the sent image, and the obtained result is sent back to the user (user terminal).

  Since the operation of the service providing apparatus is a central part of the present invention, it will be described in detail in the following embodiments.

  A first embodiment of the present invention will be described with reference to FIGS.

  In the first embodiment, an example will be described in which a user desires an input condition as a selection criterion and selects a document recognition server based thereon.

  The input conditions used here are shown as input conditions (1) in FIG.

  The user assigns priorities to the input conditions and sends them to the service providing apparatus together with the images. Here, it is assumed that the user has prioritized the order of “fee” and “delivery date”, and the service providing apparatus has the order “4-3-1-2-5” of images and input conditions and the input condition “Japan”. "Word (language)", "color (input image format)" and "html (output image format)" are sent.

  In the service providing apparatus, an image is sent to each document recognition server, and information on a delivery date and a fee required for digitizing the sent image is obtained. Here, it is assumed that the charge and delivery date of each document recognition server are obtained as shown in the processing condition example (1) illustrated in FIG.

  Here, the service providing apparatus interprets the priority order “4-3-1-2-5” obtained from the user, and performs “input data format”, “language”, “delivery date”, “charge” and Select the document recognition server in the order of “Output Data Format”.

  First, when “input data format” is selected as a reference, document recognition servers 1, 3, and 4 corresponding to “color” are selected. Since there are still three candidates here, when “Language” is selected as a reference, the document recognition servers 1, 4 correspond to “Japanese” among the document recognition servers 1, 3, 4. is there. Since two candidates still remain at this point, the next time the “delivery date” is selected as the reference, the document recognition server 4 is earlier, so the service providing apparatus finally selects the document recognition server 4. Of course, if it cannot be narrowed down to one even at this time, the comparison is performed in the order of “charge” and “output data format”. Then, a signal for requesting processing is sent to the document recognition server 4.

  When the document recognition server 4 receives a signal requesting processing, the document recognition server 4 digitizes the document image and returns the result to the service providing apparatus.

  The service providing apparatus sends the obtained electronic information data to the user terminal and charges a fee. Alternatively, there is also a method for requesting the compensation associated with the selection from the document recognition server 4. For example, the document recognition server 4 is charged 24 yen, which is 10% of the charge 240 yen, as a consideration for selection. That is, out of 240 yen obtained from the requester, 216 yen is paid to the document recognition server 4 as a price for digitization.

  A second embodiment of the present invention will be described with reference to FIGS.

  In the present embodiment, an example will be described in which an input condition is presented to a user and a document image is digitized using a document recognition server selected by the user.

  First, the user sends a document image to be digitized to the service providing apparatus using the user terminal. When the service providing apparatus receives the image, it sends it to each document recognition server to obtain the input conditions. The input conditions used here are shown in FIG. 4 as input conditions (2).

  In the service providing apparatus, an image is sent to each document recognition server, and an input condition for digitizing the sent image is obtained. Here, it is assumed that the input conditions of each document recognition server are obtained as shown in the processing condition example (2) illustrated in FIG. Here, regarding “quality”, it is assumed that actually digitized data is sent to the service providing apparatus.

  Here, the service providing apparatus sends information other than “quality” to the user terminal. The user terminal presents the input conditions sent to the user and prompts the user to select. Of course, the “quality” information may also be sent to the user terminal. However, since the “quality” information has a large amount of data, the method described below often does not take much processing time. At this point, the user can select a plurality of candidates. In the case of this example, it is assumed that the user has selected a server other than the document recognition server 3 that is heavily congested in the “network congestion level”. The result is sent from the user terminal to the service providing apparatus.

  In the service providing apparatus, the digitized data obtained by the document recognition server selected by the user is converted again into image data. The reason for re-converting the image data is that if the data that has been digitized is sent to the user terminal, the user will get the digitized data at that time. This is because there is a risk of damage. The obtained image data is sent to the user terminal and presented to the user. Here, an example of a document image sent by the user is shown as an input image example (1) in FIG. 7), FIG. 8 shows the document recognition server 2 quality example (1), and FIG. 9 shows the document recognition server 4 quality example (1).

  The user looks at the presented image, confirms “quality” (recognition performance), and considers other input conditions such as “charge” and selects the most appropriate document recognition server. In this example, it is assumed that the document recognition server 2 is selected. This selection result is sent from the user terminal to the service providing apparatus.

  The service providing apparatus sends the digitized information created by the selected document recognition server 2 to the user terminal and charges a fee.

  A third embodiment of the present invention will be described with reference to FIGS.

  In the present embodiment, when the service providing apparatus grasps the characteristics of each document recognition server, a simple analysis is performed on the input image input by the user, and the most appropriate document is determined based on the result. A method for selecting a recognition server will be described.

  FIG. 10 shows the characteristics of the document recognition server used in this embodiment as a characteristic example (1) of the document recognition server.

  Here, the charge is the charge per character (yen), the character recognition speed is the number of characters that can be recognized per second, the character size characteristic is the character size that the character recognition rate can achieve above a certain standard, and the quality characteristic is also It is the quality of characters that can achieve the character recognition rate above a certain standard. Several methods for evaluating the quality of characters have been proposed. Here, the contour length 2 ÷ the number of pixels (area) is used as a parameter representing blurring and crushing of a character image.

  If a document image is sent from the user via the user terminal, the service providing apparatus calculates the number of connected rectangular components in the image.

  The number of connected components is roughly proportional to the number of characters, but in the case of a Japanese document, there are many characters in which one character is composed of a plurality of connected components, so by adding processing that integrates connected components close to each other, Furthermore, the number of characters can be estimated with high accuracy. As the simplest method, when the distance between the centers of the connected component circumscribed rectangles is smaller than a predetermined threshold value, the connected components may be integrated.

  Then, by obtaining statistics on the size of the rectangle obtained as a result of this integration processing, the size of characters in the document can be obtained. As the simplest method, the average value of the circumscribed rectangles of the connected components may be used as the character size in the document.

  Further, the total contour length (number of edges) and the total number of black pixels in the document image are obtained, and quality = total contour length 2 ÷ total number of black pixels.

  In the present embodiment, it is assumed that a result illustrated in FIG. 11 is obtained as a document image analysis result example (1) as a result of the above analysis.

  Next, the service providing apparatus selects a document recognition server within the range of the characteristics regarding the character size and quality. In this example, the document recognition server 2 does not satisfy the character size characteristic condition, and the document recognition server 3 does not satisfy the quality characteristic. For the document recognition servers 1 and 4 remaining as candidates, a server with a low processing fee is selected within a certain processing time. Here, the threshold of the processing time is 10 seconds. The processing time in the document recognition server 1 is 963/100 = 9.63 seconds, the processing time in the document recognition server 4 is 963/80 = 12.03 seconds, and the document recognition server 4 sets the threshold of processing time. Therefore, the document recognition server 1 is finally selected. Here, if a plurality of document recognition servers do not exceed the processing time threshold, the document recognition server with the lower charge may be selected. The same applies when the processing time of all the document recognition servers exceeds the threshold value.

  The service providing apparatus sends a document image to the document recognition server 1 and sends the obtained digitized information to the user terminal.

  A fourth embodiment of the present invention will be described with reference to FIGS.

  In the present embodiment, the service providing apparatus performs area identification processing and sends only necessary portions to the document recognition server, thereby shortening the image transfer time.

  The service providing apparatus performs area identification processing on the image sent from the user, and divides the image into a photo area and a character area. Many methods have been proposed as this method. For example, a method as disclosed in JP-A-7-334614 may be used. Since the document recognition server performs character area recognition processing, the service providing apparatus sends only the obtained document area portion to the document recognition server to obtain digitized information.

  More specifically, processing when the image shown in FIG. 12 is sent from the user will be described as a region identification example (1). It is assumed that, as a result of performing area identification processing in the service providing apparatus, two document areas, one table area, and one photo area are obtained as shown in the area identification example (1).

  Here, it is assumed that the result of analyzing each document area in the same manner as in the third embodiment is obtained as a document image analysis result example (2) as shown in FIG. Further, it is assumed that the characteristics of each document recognition server are as shown in FIG. 14 as a characteristic example (1) of the document recognition server.

  First, an appropriate document recognition server is selected for the table area 1. From the characteristic example (2) of the document recognition server shown in FIG. 14, the document recognition server 3 is the only document recognition server that can process the table. Therefore, the service providing apparatus sends an image of the table area to the document recognition server 3 and digitizes it. get information.

  For the document area 1, an appropriate document recognition server is selected. It is assumed that the document area can be processed by all servers. Similar to the third embodiment, a document recognition server is selected that is within the range of its characteristics with respect to character size and quality. In this example, only the document recognition server 2 satisfies this condition. Therefore, the service providing apparatus sends the image of the document area 1 to the document recognition server 2 to obtain digitized information.

  When the same processing is performed on the document area 2, only the document recognition server 1 satisfies the character size and quality conditions, so the image of the document area 2 is sent to the document recognition server to obtain digitized information.

  If digitized information is obtained for each document area, necessary information such as the coordinates of each document area is added and sent to the user terminal.

  A fifth embodiment of the present invention will be described with reference to FIGS.

  In the present embodiment, a system configuration illustrated in FIG. 15 is used. However, the processing of the part other than the advertisement search device is the same as that of the other embodiments, so the description of the time part will be omitted.

  In the service providing apparatus, the digitized information for the image sent from the user terminal is obtained by the methods shown in the first to fourth embodiments. Next, the service providing apparatus sends the computerized information to the advertisement retrieval apparatus. The advertisement search apparatus searches for an appropriate advertisement from an advertisement DB (advertisement database) based on the digitized information, and sends the obtained advertisement to the service providing apparatus. The service providing apparatus attaches the obtained advertisement to the digitized information and sends it to the user terminal.

  Many methods have been proposed for selecting advertisements based on computerized information, but in the simplest example, advertisements with relevant keywords in the advertisement DB are extracted using words appearing in the digitized document as keys. Just do it. When registering an advertisement, it is only necessary to register the keyword to which the advertisement is given and the fee paid by the advertiser together with the advertisement. An example of the advertisement DB is shown in FIG. 16 as an advertisement DB example (1).

  Here, an example will be described in which the image shown in the region identification example (1) illustrated in FIG. 12 is given from the user terminal and the digitized information obtained by the service providing unit is input to the advertisement search device.

  The advertisement retrieval apparatus performs morphological analysis and extracts words that can be keywords. In a simple example, only nouns need be extracted. Here, it is assumed that “summer vacation”, “Izu plateau”, and “serasegi” are extracted. Next, an advertisement in which these words are keywords is searched from the advertisement DB. In the case of the advertisement DB example, the keywords No. 123 and 124 have “summer vacation”, so this advertisement is extracted and sent to the service provider, and the advertiser is charged.

JP 2000-172428 A JP-A-7-28946

Claims (93)

A first step of receiving an image to be subjected to document recognition processing from a requester;
A second step of requesting document recognition processing by selecting an appropriate document recognition server for document recognition processing of the received image from a plurality of document recognition servers;
A third step of receiving electronically converted data after document recognition processing from the requested document recognition server and providing the data to the requester;
A document input proxy method comprising:   The document input proxy method according to claim 1, wherein the second step includes a process of obtaining selection criteria of the document recognition server from a client.   The second step includes a process of selecting a most appropriate server after presenting a given image to a plurality of the document recognition servers and collecting processing conditions from each of the document recognition servers. Document input proxy method described.   In the process of selecting the most appropriate server, the processing conditions collected from the document recognition server are presented to the requester to prompt the selection, and the document recognition process is determined based on the requester's selection. 4. The document input proxy method according to claim 3, further comprising:   3. The document input proxy method according to claim 1, wherein a fee required for document recognition is used as the selection criterion.   5. The document input proxy method according to claim 1, wherein a fee required for document recognition is used as the processing condition.   3. The document input proxy method according to claim 1, wherein a time required for document recognition is used as the selection criterion.   5. The document input proxy method according to claim 1, wherein time required for document recognition is used as the processing condition.   3. The document input proxy method according to claim 1, wherein the quality of the document recognition result is used as the selection criterion.   The document input proxy method according to claim 1, 3 or 4, wherein the quality of the document recognition result is used as the processing condition.   3. The document input proxy method according to claim 1, wherein at least one of a congestion degree of the document recognition server and a congestion degree of a network for accessing the document recognition server is used as the selection criterion.   5. The document input proxy method according to claim 1, wherein at least one of a congestion level of the document recognition server and a congestion level of a network for accessing the document recognition server is used as the processing condition.   3. The document input proxy method according to claim 1, wherein a language supported by the document recognition server is used as the selection criterion.   5. The document input proxy method according to claim 1, wherein a language supported by the document recognition server is used as the processing condition.   3. The document input proxy method according to claim 1, wherein an input image format supported by the document recognition server is used as the selection criterion.   5. The document input proxy method according to claim 1, wherein an input image format supported by the document recognition server is used as the processing condition.   3. The document input proxy method according to claim 1, wherein an output format supported by the document recognition server is used as the selection criterion.   6. The document input proxy method according to claim 1, wherein an output format supported by the document recognition server is used as the processing condition.   6. The second step includes a process of performing an analysis process on a given image and selecting an appropriate document recognition server from a plurality of document recognition servers based on the analysis result. Document input proxy method.   The document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a connected rectangular component in a given image.   The document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a character size in a given image.   The document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing the number of characters in a given image.   The document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a quality of a given image.   The document input proxy method according to claim 19, wherein the analysis process includes a process of analyzing a layout in a given image.   25. The document input proxy method according to claim 24, wherein the second step includes a process of sending only a character region obtained as a result of a process of analyzing a layout in the image to the document recognition server.   25. The document input according to claim 24, wherein the second step includes a process of selecting the document recognition server for each area according to an attribute of each area obtained as a result of the process of analyzing the layout in the image. Acting method.   In the second step, the given image is transmitted to the document recognition server that is a candidate, the digitalized data after the document recognition process is received from the document recognition server, and the received document recognition process 27. The document input proxy method according to claim 1, further comprising a process of synthesizing the results and presenting the result to the client.   28. The document input proxy method according to claim 27, wherein the received document recognition processing result is converted into a state not converted into electronic information and presented to the client.   The document input proxy method according to any one of claims 1 to 28, wherein the third step includes a process of charging a requester for a compensation for the document recognition server selection.   The document input proxy method according to any one of claims 1 to 29, wherein the third step includes a process of charging an administrator of the document recognition server for compensation for the document recognition server selection. The third step includes a process of sending an advertisement selected according to the content of the image requested for the document recognition process from the requester to the requester together with the digitalized data after the document recognition process,
The document input proxy method according to any one of claims 1 to 30, further comprising a fourth process of charging a fee for selection of an advertisement to an advertiser of the advertisement associated with the digitalized data. Installed on the computer,
A first function for receiving an image as a document recognition processing target from a client via a network;
A second function of requesting document recognition processing by selecting an appropriate document recognition server for document recognition processing of the received image from a plurality of document recognition servers;
A third function for receiving electronically converted data after document recognition processing from the requested document recognition server and providing the data to the requester;
A computer program for document input substitution that causes a computer to execute.   33. The computer program for proxying document input according to claim 32, wherein the second function includes a process of obtaining a selection criterion of the document recognition server from a requester.   The second function includes a process of selecting a most suitable server after presenting a given image to a plurality of the document recognition servers and collecting processing conditions from each of the document recognition servers. A computer program for the document input agent described.   In the process of selecting the most appropriate server, the processing conditions collected from the document recognition server are presented to the requester to prompt the selection, and the document recognition process is determined based on the requester's selection. 35. The computer program for proxying document input according to claim 34, further comprising a process for requesting a document.   34. The computer program for proxying document input according to claim 32 or 33, wherein a fee required for document recognition is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein a fee required for document recognition is used as the processing condition.   34. The computer program for proxying document input according to claim 32 or 33, wherein a time required for document recognition is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein time required for document recognition is used as the processing condition.   34. The computer program for proxying document input according to claim 32 or 33, wherein the quality of the document recognition result is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein the quality of the document recognition result is used as the processing condition.   34. The computer program for proxying a document input according to claim 32 or 33, wherein at least one of a congestion degree of the document recognition server and a congestion degree of a network for accessing the document recognition server is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein at least one of a congestion level of the document recognition server and a congestion level of a network for accessing the document recognition server is used as the processing condition.   34. The computer program for proxying document input according to claim 32 or 33, wherein a language supported by the document recognition server is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein a language supported by the document recognition server is used as the processing condition.   34. The computer program for proxying a document input according to claim 32 or 33, wherein an input image format supported by the document recognition server is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein an input image format supported by the document recognition server is used as the processing condition.   34. The computer program for proxying document input according to claim 32 or 33, wherein an output format supported by the document recognition server is used as the selection criterion.   36. The computer program for proxying document input according to claim 32, 34, or 35, wherein an output format supported by the document recognition server is used as the processing condition.   36. The second function includes a process of performing an analysis process on a given image and selecting an appropriate document recognition server from a plurality of document recognition servers based on the analysis result. Computer program for document input agency.   51. The computer program for proxying document input according to claim 50, wherein the analysis processing includes processing for analyzing connected rectangular components in a given image.   51. The computer program for proxying document input according to claim 50, wherein said analysis processing includes processing for analyzing a character size in a given image.   51. The computer program for proxying document input according to claim 50, wherein said analysis processing includes processing for analyzing the number of characters in a given image.   51. The computer program for proxying document input according to claim 50, wherein the analysis processing includes processing for analyzing quality of a given image.   51. The computer program for proxying document input according to claim 50, wherein the analysis process includes a process of analyzing a layout in a given image.   56. The computer program for proxying document input according to claim 55, wherein the second function includes a process of sending only a character area obtained as a result of a process of analyzing a layout in the image to the document recognition server.   56. The document input according to claim 55, wherein the second function includes a process of selecting the document recognition server for each area according to an attribute of each area obtained as a result of the process of analyzing the layout in the image. Computer program for substitution.   The second function transmits the image given to the candidate document recognition server, receives the digitized data after the document recognition process from the document recognition server, and receives the received document recognition process. 58. A computer program for proxying document input according to any one of claims 32 to 57, which includes a process of synthesizing the results and presenting them to the client.   59. The computer program for proxying document input according to claim 58, wherein the received document recognition processing result is converted into a state that is not converted into electronic information and presented to the client.   The computer program for proxying a document input according to any one of claims 32 to 59, wherein the third function includes a process of charging a requester for a fee for selecting the document recognition server.   The computer program for proxying document input according to any one of claims 32 to 60, wherein the third function includes a process of charging an administrator of the document recognition server for a fee for selecting the document recognition server. The third function includes a process of sending an advertisement selected according to the content of the image requested for the document recognition process from the requester to the requester together with the digitalized data after the document recognition process,
62. The computer program for proxying document input according to any one of claims 32 to 61, further comprising a fourth process of charging a fee for selection of an advertisement to an advertiser of the advertisement associated with the electronic information data. A first means for receiving an image to be subjected to document recognition processing from a requester via a network;
Second means for requesting document recognition processing by selecting an appropriate document recognition server for document recognition processing of the received image from a plurality of document recognition servers;
Third means for receiving electronically converted data after document recognition processing from the requested document recognition server and providing the data to the requester;
A document input proxy device comprising:   64. The document input proxy device according to claim 63, wherein the second means includes processing for obtaining a selection criterion of the document recognition server from a requester.   64. The second means includes a process of selecting a most appropriate server after presenting a given image to a plurality of document recognition servers and collecting processing conditions from each of the document recognition servers. Document input proxy device described.   In the process of selecting the most appropriate server, the processing conditions collected from the document recognition server are presented to the requester to prompt the selection, and the document recognition process is determined based on the requester's selection. 66. The document input proxy device according to claim 65, further comprising:   65. The document input proxy device according to claim 63 or 64, wherein a fee required for document recognition is used as the selection criterion.   67. The document input proxy device according to claim 63, 65, or 66, wherein a fee required for document recognition is used as the processing condition.   The document input proxy device according to claim 63 or 64, wherein a time required for document recognition is used as the selection criterion.   67. The document input proxy device according to claim 63, 65, or 66, wherein a time required for document recognition is used as the processing condition.   The document input proxy device according to claim 63 or 64, wherein the quality of the document recognition result is used as the selection criterion.   67. The document input proxy device according to claim 63, 65, or 66, wherein the quality of the document recognition result is used as the processing condition.   65. The document input proxy device according to claim 63, wherein at least one of a congestion degree of the document recognition server and a congestion degree of a network for accessing the document recognition server is used as the selection criterion.   67. The document input proxy device according to claim 63, 65, or 66, wherein at least one of a congestion degree of the document recognition server and a congestion degree of a network for accessing the document recognition server is used as the processing condition.   The document input proxy device according to claim 63 or 64, wherein a language supported by the document recognition server is used as the selection criterion.   67. The document input proxy device according to claim 63, 65, or 66, wherein a language supported by the document recognition server is used as the processing condition.   The document input proxy device according to claim 63 or 64, wherein an input image format supported by the document recognition server is used as the selection criterion.   67. The document input proxy device according to claim 63, 65, or 66, wherein an input image format supported by the document recognition server is used as the processing condition.   The document input proxy device according to claim 63 or 64, wherein an output format supported by the document recognition server is used as the selection criterion.   67. The document input proxy device according to claim 63, 65 or 66, wherein an output format supported by the document recognition server is used as the processing condition.   66. The process according to claim 63, 65, or 66, wherein the second means includes a process of performing an analysis process on a given image and selecting an appropriate document recognition server from a plurality of document recognition servers based on the analysis result. Document input proxy device.   The document input proxy device according to claim 81, wherein the analysis processing includes processing for analyzing a connected rectangular component in a given image.   The document input proxy device according to claim 81, wherein the analysis processing includes processing for analyzing a character size in a given image.   The document input proxy device according to claim 81, wherein the analysis processing includes processing for analyzing the number of characters in a given image.   The document input proxy device according to claim 81, wherein the analysis processing includes processing for analyzing quality of a given image.   The document input proxy device according to claim 81, wherein the analysis processing includes processing for analyzing a layout in a given image.   87. The document input proxy device according to claim 86, wherein the second means includes a process of sending only a character area obtained as a result of a process of analyzing a layout in the image to the document recognition server.   The document input according to claim 86, wherein the second means includes a process of selecting the document recognition server for each area according to an attribute of each area obtained as a result of the process of analyzing the layout in the image. Proxy device.   The second means transmits a given image to the document recognition server that is a candidate, receives the digitized data after the document recognition process from the document recognition server, and receives the received document recognition process The document input proxy device according to any one of claims 63 to 88, further comprising a process of synthesizing the results and presenting the result to the client.   90. The document input proxy device according to claim 89, wherein the received document recognition processing result is converted into a state not converted into electronic information and presented to the client.   The document input proxy device according to any one of claims 63 to 90, wherein the third means includes a process of charging a requester for a fee for selecting the document recognition server.   The document input proxy device according to any one of claims 63 to 91, wherein the third means includes a process of charging an administrator of the document recognition server for a value for the document recognition server selection. The third means includes a process of sending an advertisement selected according to the content of the image requested for the document recognition process from the requester to the requester together with the digitalized data after the document recognition process,
93. The document input proxy device according to claim 63, further comprising a fourth process of charging a fee for advertisement selection to an advertiser of the advertisement associated with the digitalized data.

Images