JP2007034549A - Proposal design management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To form a proposal with an appropriate content in a short time with less burdens of cost and labor hour by selecting better article and price without confirmation of a customer's purchase requirement or without the need for experience, intuition or knowledge. <P>SOLUTION: This system comprises an imaging means inputting an image of a form; a character recognition means accessing a character recognition dictionary database and reading characters of the form from the image; a data conversion means accessing an inter-commodity correspondence database and determining a commodity corresponding to a commodity name described in the form read by the character recognition means; a data retrieval means accessing a commodity information database and retrieving data of the commodity determined by the data conversion means to determine proposal contents; a format conversion means accessing a format information database and creating proposal data with the contents determined by the data retrieval means; and a data output means outputting the proposal data. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a proposal design management system.

  Traditionally, companies that belong to the service industry, such as stores and financial institutions, and companies that belong to the manufacturing industry that manufactures industrial products, etc., have the purpose of encouraging purchases to sell products to customers. Proposals and quotations are presented. In that case, the customer receives proposals and quotations from multiple sellers who sell similar products, compares the content and price of the products, and selects the seller that best suits their purchase conditions as the purchase destination. Select.

Therefore, for the seller, proposing an optimal combination of products according to the purchase conditions of the customer is an effective means that leads to a product order. In particular, in order to propose an optimal combination of products in an industry with many types of products and assortments, a high level of product knowledge and experience is required. An estimate making system has been used (for example, see Patent Document 1). Such a system creates a database of product attributes such as product name, price, and manufacturer name, provides a template for a proposal form, and further includes product selection, automatic price tabulation, and a standard form printing function. provide.
JP-A-7-129691

  However, in the conventional system described above, the product information of the products handled by the seller is simply made into a database. Therefore, in order to match the purchase requirements of the customer with the products, work such as customer hearing and order form input This is a burden on both customers and sellers. In addition, it is necessary to select products and prices that are in better conditions than the proposals of other sellers competing with the customer's purchase requirements, but this requires the experience, intuition, and knowledge of sales representatives. Therefore, it is necessary to make an investment for training sales staff, and it takes time to create a proposal, which increases the burden of cost and labor. Furthermore, for products that have already been contracted with customers, there are timings that are suitable for proposals, such as the timing for renewing contracts, but a great deal of effort is required to manage such timings individually. It was. Also, if the timing cannot be managed properly, important sales opportunities will be lost.

  The present invention solves the problems of the conventional system so that a proposal can be automatically created based on the contents of other proposals, without checking the purchase requirements of the customer, In addition, it is possible to select better products and prices without the need for experience, intuition and knowledge, and it is possible to create proposals with appropriate contents in a short time with less burden on costs and labor. The purpose is to provide a proposal design management system.

  Therefore, in the proposal design management system of the present invention, the imaging means for inputting the image of the form, the character recognition means for accessing the character recognition dictionary database and reading the characters of the form from the image, and the inter-product correspondence database A data conversion means for obtaining a product corresponding to the product name described in the form read by the character recognition means, and accessing a product information database and searching for the data of the product obtained by the data conversion means Data search means for determining the description contents of the document, format conversion means for accessing the format information database and creating proposal data of the description contents determined by the data search means, and data for outputting the proposal data Output means.

  In another proposal design management system of the present invention, an imaging means for inputting a form image, a character recognition means for accessing a character recognition dictionary database and reading a form character from the image, and the character recognition means Data storage means for storing the product data described in the form in the proposal information database, and sequentially comparing the product data stored in the proposal information database with the notification criteria stored in the notification criteria database. , Timing measurement means for determining whether or not to notify the proposal information, and when converting the proposal information, the format conversion is performed by accessing the format information database and creating the notification data for notifying the proposal information. Means and data output means for outputting the data of the notice.

  In another proposal design management system of the present invention, a form edge detecting means for detecting the coordinates of the four corners of the form on the image, and a distortion for calculating the distortion correction parameter of the image based on the coordinates of the four corners. The image processing apparatus further includes: a quantity calculation unit; and a distortion correction unit that corrects the distortion of the image based on the distortion correction parameter. The character recognition unit reads a character of the form from the image whose distortion is corrected.

  According to the present invention, in the proposal design management system, a proposal can be automatically created based on the contents of other proposals. Therefore, it is possible to select better products and prices without confirming the purchase requirements of the customer and without the need for experience, intuition and knowledge. Proposals with appropriate content can be created.

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

  FIG. 1 is a diagram showing the configuration of a proposal design management system according to the first embodiment of the present invention.

  As shown in the figure, the proposal design management system according to the present embodiment includes an image input apparatus 10 that performs image acquisition and transmission, a proposal design apparatus 20 that automatically creates a proposal, and the proposal design. A proposal output device 40 that outputs the proposal received from the device 20 is provided. The image input apparatus 10 includes an imaging unit 11 and a data communication unit 12. Further, the proposal design apparatus 20 includes a data communication unit 21, a form edge detection unit 22, a distortion amount calculation unit 23, a distortion correction unit 24, a character recognition unit 25, a data conversion unit 26, a character recognition dictionary database 27, an inter-commodity A correspondence database 28, a data search means 31, a format conversion means 32, a product information database 33, a format information database 34, a charging means 35, a data communication means 36, and a charging database 37 are provided. Further, the proposal output device 40 includes data communication means 41 and data output means 42.

  Here, the proposal design management system is a system for designing and managing proposals presented to customers by sellers of various products including service products in the fields of industrial products, financial products, insurance and the like. . Note that the proposal includes an estimate of a product. The seller is generally a company, an organization, or the like, but may be an individual or a person who sells any kind of product. Further, the customer may be any person such as an individual, a company, or an organization as long as the customer purchases the product.

  The proposal design management system acquires description items such as proposals, quotations, contracts, etc. created by other sellers regarding products that the customer desires to purchase as an image, and is included in the acquired description items It is a system that reads product requirements by OCR (Optical Character Reader) technology and automatically designs better combinations and prices of products sold by the customer. The proposal design management system uses a digital camera and a camera as a means for obtaining images of these documents when a customer presents proposals, estimates, contracts, etc. of other sellers. It has an image input means such as a cellular phone, and further has a function of correcting paper distortion caused by the angle of view of the acquired image.

  Here, the image pickup means 11 of the image input device 10 is an image input means, and may be a field angle free device such as a digital camera, a camera of a camera-equipped mobile phone, a scanner, a handy image reader, An apparatus with a fixed angle of view such as a facsimile may be used. Further, the image pickup means 11 may be an input device of image files acquired by these devices or application software for input.

  The data communication means 12 is means for transmitting information as digital data through a communication circuit. The communication circuit may be wired or wireless. The data communication means 12 also has communication control functions such as retransmission control and error correction control according to communication traffic. The data communication unit 12 transmits the image data input by the imaging unit 11 to the proposal design apparatus 20.

  The data communication means 21 of the proposal design apparatus 20 is a digital data transmission means similar to the data communication means 12 of the image input apparatus 10 and receives input image data transmitted from the image input apparatus 10. .

  Further, the form edge detection unit 22, the distortion amount calculation unit 23, and the distortion correction unit 24, such as a proposal, an estimate, a contract, etc., when the input image is taken by an angle-free device. This is means for performing various processes for correcting trapezoidal distortion caused by the angle of view that occurs when the front side of the form surface is photographed obliquely. Here, the reason for performing these processes is to correct the distortion of characters due to image distortion and reduce the occurrence of misreading and unreading in the subsequent OCR process.

  The form edge detecting means 22 is means for calculating the coordinates of the four corners of the form on the image to be calculated by the distortion amount calculating means 23 by image processing. Specifically, first, the four sides of the form are detected, and a linear expression at the coordinates on the image is obtained. Thereafter, the coordinates of the four corners of the form can be calculated by obtaining the intersection of each straight line. As a method for detecting the four sides, a straight line detection method such as a Huff transform can be applied to the original image after applying an edge extraction filter typified by the Sobel operator and using a point having a strong edge strength. .

  The distortion amount calculation means 23 is a means for calculating distortion correction parameters for perspective transformation from the coordinates of the four corners of the form calculated by the form edge detection means 22. The perspective transformation is a known process for generating a two-dimensional image from a specific viewpoint in consideration of a three-dimensional camera position with respect to the form surface. Then, the distortion amount calculation means 23 calculates distortion correction parameters to be converted into an image without distortion obtained by photographing the original image from the front by applying the coordinates of the four points.

  Further, the distortion correction unit 24 is a unit that performs perspective transformation on each pixel of the original image based on the distortion correction parameter calculated by the distortion amount calculation unit 23 to correct the distortion of the image. In this way, since the trapezoidal distortion caused by photographing from an oblique direction is corrected, OCR processing can be applied in the same manner as an image photographed by a device with a fixed angle of view.

  The character recognition unit 25 is a unit that performs OCR processing on the image output from the distortion correction unit 24 and reads the written characters. Specifically, a character feature extraction process, a matching process with a dictionary registered in advance in the character recognition dictionary database 27, a selection process for selecting a recognition result from candidates, and the like are performed.

  Further, the data conversion means 26 is a means for performing processing for obtaining the corresponding own product based on the association between the products from the product name or the like of the proposal read by the character recognition means 25. Assume that associations between products are created in advance and stored in the inter-product correspondence database 28.

  Further, the data search means 31 accesses the product information database 33 using the product name replaced by the product name of another seller by the data conversion means 26 as a key, and the minimum price, characteristics, etc. of the product sold by itself. Is a means for searching. The data search means 31 determines the description content of the proposal created by the proposal design apparatus 20.

  The format conversion unit 32 is a unit that reads out the document format, layout, and the like stored in the format information database 34 and sets the data obtained by the data search unit 31 as a corresponding item. Thereby, a proposal including the appearance can be created.

  The charging unit 35 is a unit that performs a process of adding a count to the usage history counter of the charging database 37 for each process. The counter value of the usage history counter becomes usage history data that is subject to metered accounting in a business model that charges each time a proposal is designed.

  The data communication means 36 is a digital data transmission means similar to the data communication means 12 and the data communication means 21, and transmits the proposal data created by the proposal design apparatus 20 to the proposal output apparatus 40.

  The data communication means 41 of the proposal output device 40 is a digital data transmission means similar to the data communication means 12, the data communication means 21, and the data communication means 36, and is transmitted from the proposal design apparatus 20. Receive proposal data.

  Further, the data output means 42 may be software such as application software or browser software that displays the proposal automatically designed by the proposal design apparatus 20 on a display, or hardware such as a printer that outputs to a medium such as paper. It may be a wear device.

  The character recognition dictionary database 27 stores data such as registration dictionary data used when recognizing characters on the image, format registration data defining where information such as product names and prices are on the form. It is a stored database.

  The inter-product correspondence database 28 is a database that lists associations between product names of other sellers and corresponding product names sold by the seller.

  The merchandise information database 33 is a database of merchandise information of merchandise to be sold for searching using merchandise names as keys, and stores list information such as merchandise names, prices, users, features, and special notes. .

  Furthermore, the format information database 34 is a database that defines the format, layout, etc. of the proposal.

  The accounting database 37 is a database for recording the number of accesses to the proposal design apparatus 20, and the charge obtained by accessing the proposal design apparatus 20 and acquiring information by multiplying the recorded number of accesses by the unit price. Can be charged as a fee.

  Next, the operation of the proposal design management system configured as described above will be described.

  FIG. 2 is a diagram showing an example of the data configuration of the inter-product correspondence database of the proposal design management system according to the first embodiment of the present invention, and FIG. 3 is a proposal design management system according to the first embodiment of the present invention. FIG. 4 is a flowchart showing the operation of the proposal design management system according to the first embodiment of the present invention.

  The proposal design management system in the present embodiment is a system for designing and managing proposals related to various products, and is not limited to a specific product. The case where life insurance is used will be described as an example.

  In life insurance, when a contract is established, it is common to write the contract content on an insurance policy and deliver it to the customer. In the present embodiment, an example is given in which insurance company A as a seller presents his / her proposal to customer C who has taken out life insurance of insurance company B as another seller. I will explain. Then, it is assumed that the customer C who wants to review the life insurance presents the insurance policy of the insurance company B under contract to the insurance company A.

  First, the operation of the image input apparatus 10 will be described. In this case, it is assumed that the image input device 10 is installed in the sales office of the insurance company A. Then, using the image pickup means 11, the sales staff of the insurance company A or the customer C himself photographs the insurance policy of the insurance company B. The imaging unit 11 is a digital camera, a camera-equipped mobile phone camera, a scanner, a handy image reader, a facsimile, or the like. The photographed data becomes digital image data and is transmitted from the imaging means 11 to the data communication means 12.

  The data communication means 12 is software such as a dedicated device, a dedicated application that runs on the OS (operating system) of the computer, or a plug-in on a general-purpose browser. The received digital image data is transmitted to the proposal design apparatus 20 via the communication circuit.

  Next, the operation of the proposal design apparatus 20 will be described. The proposal design device 20 receives the digital image data of the insurance policy of the insurance company B from the image input device 10, and transmits the created insurance company A proposal to the proposal output device 40. In this case, first, the data communication means 21 receives digital image data from the image input device 10.

  Subsequently, the form edge detection unit 22, the distortion amount calculation unit 23, and the distortion correction unit 24 perform distortion correction processing on the digital image data received from the image input device 10. Normally, when taking a picture of a form such as an insurance policy using a device such as a digital camera that has a degree of freedom for the angle of view, the paper is not rectangular on the image, and the front is large depending on the distance. , Trapezoidal distortion occurs in the back side. In addition, there is also an inclination when it is oblique to the paper surface. The distortion correction processing in the present embodiment corrects such distortion by perspective transformation using the coordinates of the four corners of the form.

  In this case, the form edge detecting means 22 is means for calculating the coordinates of the four corners of the form on the image to be calculated by the distortion amount calculating means 23 by image processing. Specifically, first, the four sides of the form are detected, and a linear expression at the coordinates on the image is obtained. Thereafter, the coordinates of the four corners of the form can be calculated by obtaining the intersection of each straight line. As a four-sided detection method, an edge extraction filter represented by the Sobel operator is applied to the original image, and a known straight line detection method such as a Huff transform is applied to a point having a strong edge strength. To detect.

  Subsequently, the distortion amount calculation unit 23 calculates perspective transformation parameters from the coordinates of the four corners of the form calculated by the form edge detection unit 22. Here, the distortion amount calculating means 23 calculates parameters to be converted into an image without distortion obtained by photographing the original image from the front by applying the coordinates of the four points.

  Subsequently, the distortion correction unit 24 performs perspective transformation on each pixel of the original image based on the distortion correction parameter calculated by the distortion amount calculation unit 23 to correct the distortion of the image. In this way, by correcting the trapezoidal distortion and tilt caused by photographing from an oblique direction, an image similar to an image photographed by a device having a fixed angle of view such as a scanner can be obtained. Therefore, a conventional OCR process can be applied to an image photographed by a scanner.

  Subsequently, the character recognition unit 25 performs OCR processing on the image output by the distortion correction unit 24 and reads the written characters. In this case, a known OCR technique is used to access the character recognition dictionary database 27 and perform OCR processing. In the character recognition dictionary database 27, information such as form layout information, form format information, and a character recognition dictionary is registered in advance. This information can be accumulated by accumulating the form image obtained as described above and registering the form layout and form format each time a new form is entered. it can.

  The input image is compared with a plurality of layouts registered after analyzing the form layout, and form identification is performed. For the form identification, existing techniques such as a ruled line feature of an insurance policy and a character string at a specific position are used. As a result, when it is recognized that the document is a specific form, character recognition is performed for each area defined in advance as a form format. Character recognition is a process of determining a character by performing a character region extraction and a feature extraction process and performing a feature matching process with a plurality of registered character recognition dictionaries. In addition, since attributes such as an insurance company name, product name, quantity, and price are defined in each area, the contents of the insurance policy of insurance company B can be read by performing character recognition. The insurance company name can also be specified by the form identification result.

  Subsequently, the data conversion unit 26 obtains the product name of the corresponding insurance company A from the product name of the insurance policy of the insurance company B read by the character recognition unit 25 based on the inter-product correspondence database 28. Here, in the inter-product correspondence database 28, as shown in FIG. 2, the correspondence between the product names of other insurance companies and the company name, that is, the product name of the insurance company A is created and stored in advance as a list. It shall be. The correspondence stored in the inter-product correspondence database 28 does not necessarily relate to the same product, but relates to the product of the insurance company A that can be proposed to the customer C for the product of another insurance company. I just need it. Then, the data conversion unit 26 converts the product name output from the character recognition unit 25 into the product name of the insurance company A according to the association stored in the inter-product correspondence database 28. The matching products are registered in the “own product name” column, but if there is no such product, the “corresponding similar product name” column is referred to. In the “corresponding similar product name” column, when there are a plurality of similar products, a list of all products that can be presented is registered.

  In the example shown in FIG. 2, the column of “insurance company name” stores the name of each insurance company and is used to specify the name of the insurance company identified from the contents of the insurance policy. If the digital image data received from the image input apparatus 10 is for the company, that is, the insurance policy of the insurance company A and for the old insurance, “own company (old product)” is referred to. Thereby, the proposal about the new insurance which is a corresponding new product can be created.

  Subsequently, the data search means 31 accesses the product information database 33 using the product name of the insurance company A replaced by the data name of the insurance company B by the data conversion means 26 as a key, and the minimum of the products sold by the insurance company A Search prices, features, etc. The data search means 31 determines the description of the insurance company A proposal created by the proposal design apparatus 20.

  Here, in the product information database 33, as shown in FIG. 3, contents to be proposed to the customer C, such as the minimum price and features, are registered for the products sold by the insurance company A. By printing these contents, a proposal to the customer C can be created.

  Subsequently, the format conversion unit 32 reads the document format, layout, etc. stored in the format information database 34 and performs processing for setting the data obtained by the data search unit 31 to the corresponding item. Thereby, a proposal including the appearance can be created. Here, as described above, when there are a plurality of corresponding similar products, a proposal is created for all the products. Thereby, even if there is no product of the same company, that is, the insurance company A, a plurality of candidates can be presented to the customer C, and the possibility of closing can be increased. In addition, if an area for setting “other company's product name” is provided in the template, it is possible to clearly indicate which insurance the customer C is currently contracting, so that the proposal is easier to understand. It can be.

  Subsequently, the charging unit 35 counts up the usage history counter of the charging database 37 for each process. As a result, pay-as-you-go billing based on the billing database 37 is possible, so that a business form in which the proposal design apparatus 20 is operated by an organization different from the insurance company A can be realized.

  Subsequently, the data communication unit 36 transmits the proposal data created by the proposal design apparatus 20 to the proposal output apparatus 40.

  Next, the operation of the proposal output apparatus 40 will be described. The proposal output device 40 outputs the proposal received from the proposal design device 20. In this case, first, the data communication means 41 receives the proposal data transmitted by the proposal design apparatus 20.

  Subsequently, the data output means 42 prints and outputs the proposal data automatically designed by the proposal design apparatus 20 on a medium. In addition, you may output by displaying on the application, browser, etc. on a display.

Next, a flowchart will be described.
Step S <b> 1 The customer's insurance policy is photographed and sent to the proposal design apparatus 20.
Step S2 The four sides of the insurance policy image are detected.
Step S3: Calculate the distortion amount of the insurance policy image.
Step S4 The distortion of the insurance policy image is corrected.
Step S5: The insurance policy layout is analyzed to identify the form.
Step S6: It is determined whether the form identification is successful. If the form identification is successful, the process proceeds to step S8, and if the form identification is not successful, the process proceeds to step S7.
Step S7 The form layout and the form format are additionally registered using the corrected image of the insurance policy.
Step S8 Character recognition of each area (insurance company name, product name, quantity, price) is performed according to the form format corresponding to the specified type of insurance policy.
Step S9: The inter-product correspondence database 28 is searched to identify the corresponding product name of the company.
Step S10: Information on the specified company product is acquired from the product information database 33.
Step S11 According to the template of the format information database 34, the acquired information is embedded and a proposal is created.
Step S12: The usage history counter is counted up.
Step S13 The proposal is printed and the process is terminated.

  As described above, in the present embodiment, the proposal design management system acquires, as an image, description items such as proposals, quotations, contracts, and the like created by other sellers regarding the products that the customer desires to purchase, Read the product requirements included in the acquired entries and automatically create proposals for the products you sell. Therefore, since it is not necessary to confirm the product requirements with the customer, it is possible to save the trouble and labor of the customer and the seller. In addition, since it is possible to select their own products corresponding to the products of other sellers and to present conditions that are more advantageous than other sellers, product knowledge and experience are not required for sales representatives. Education investment and personnel cost investment can be suppressed.

  Next, a second embodiment of the present invention will be described. In addition, about the thing which has the same structure as 1st Embodiment, the description is abbreviate | omitted by providing the same code | symbol. The description of the same operation and the same effect as those of the first embodiment is also omitted.

  FIG. 5 is a diagram showing a configuration of a proposal design management system according to the second embodiment of the present invention.

  The proposal design management system in the first embodiment immediately sells itself when a description item such as an image of a proposal, an estimate, a contract, etc. created by another seller is acquired as an image. The proposal design management system in the present embodiment is designed to automatically create proposals, quotations, contracts, etc. created by other sellers. When items such as images are acquired as images, they are stored as customer data and managed in a time series, and sales representatives are notified at the most appropriate time when proposals for customers are created. Yes. In other words, the proposal design management system in the first embodiment creates a proposal for each request, whereas the proposal design management system in the present embodiment assigns an appropriate sales person to the sales person according to the scheduling. Notify and create proposals for customers.

  Therefore, the proposal design management system in the present embodiment, for example, the product is a product such as life insurance, time deposits, various loans, etc., depending on conditions such as reserve period, maturity of contract period, change of interest rate, etc. This is particularly effective when the product has a time suitable for contract renewal or contract modification. In this case, the input contract information is accumulated, and the proposal design management system automatically notifies the optimal sales staff at the optimal timing.

  As shown in FIG. 5, in the configuration of the proposal design management system in the present embodiment, the proposal design apparatus 20 further includes a data storage means 44, a timing measurement means 45, a proposal information database 46, and a notification reference database 47. The other points are the same as the configuration of the proposal design management system in the first embodiment.

  Here, the data storage unit 44 is a unit that accumulates and stores various conditions such as the product name, contract expiration date, and interest rate of the proposal read by the character recognition unit 25 in the proposal information database 46.

  The timing measuring means 45 is a means for repeating the determination of the presence / absence of the proposal information notification at regular intervals. The timing measuring unit 45 sequentially compares information such as contract expiration dates, interest rates, and prices stored in the proposal information database 46 with the notification criteria. Based on the jurisdiction information and the like stored in the database 47, a sales representative to whom the proposal information is notified is determined, and only when the sales representative is determined, processing of the means after the format conversion means 32 Start up.

  Further, the proposal information database 46 is a database for contract information between customers and other sellers, and stores a list of product names, contract expiration dates, interest rates, and the like.

  The notification reference database 47 stores a list of data that serves as a reference for determining whether or not there is a proposal notification when the content of the information stored in the proposal information database 46 is determined. Specifically, the list is a list of notification date criteria, notification interest rates, notification prices, and the like.

  In the present embodiment, the format information database 34 defines a template for the notification document, and the format conversion means 32 reads the template for the notification document stored in the format information database 34 to obtain customer information and salesperson information. Etc., and a notification document including the appearance is created. The data communication means 36 transmits the notification document data created by the proposal design apparatus 20 to the proposal output apparatus 40, and the data output means 42 notifies the customer about the proposal made by the proposal design apparatus 20. Output the document.

  The other points are the same as those in the first embodiment, and a description thereof will be omitted.

  Next, the operation of the proposal design management system in this embodiment will be described.

  FIG. 6 is a first diagram showing an example of the data structure of the notification reference database of the proposal design management system according to the second embodiment of the present invention, and FIG. 7 is a proposal design according to the second embodiment of the present invention. FIG. 8 is a flowchart showing an example of the data structure of the notification standard database of the management system, and FIG. 8 is a flowchart showing the operation of the proposal design management system in the second embodiment of the present invention.

  The proposal design management system in the present embodiment is a system for designing and managing proposals related to various products for which a contract period exists, and is not limited to specific products. For convenience, the case where the product is an investment trust will be described as an example.

  For fixed-term contract products such as investment trusts, it is common to review products based on the investment yield at the same time as the maturity. In the present embodiment, the case where financial institution A as a seller presents his / her proposal to customer C who has joined an investment trust of financial institution B as another seller is taken as an example. I will explain. Then, it is assumed that the customer C presents the insurance company A with the contract of the financial institution B that is contracting the investment trust as part of a questionnaire of the financial institution A or the like.

  First, the operation of the image input apparatus 10 will be described. In this case, it is assumed that the image input device 10 is installed at the sales office of the financial institution A. Then, using the image pickup means 11, the sales representative of the financial institution A or the customer C himself / herself photographs the investment trust contract of the financial institution B. The imaging unit 11 is a digital camera, a camera-equipped mobile phone camera, a scanner, a handy image reader, a facsimile, or the like. The photographed data becomes digital image data and is transmitted from the imaging means 11 to the data communication means 12.

  The data communication means 12 is a dedicated device or software such as a dedicated application running on the OS of the computer, a plug-in on a general-purpose browser, and the digital image data received from the imaging means 11 is transferred to the communication circuit. To the proposal design apparatus 20.

  Next, the operation of the proposal design apparatus 20 will be described. The proposal design apparatus 20 receives the digital image data of the contract of the financial institution B from the image input apparatus 10 and notifies the proposal output apparatus 40 used by an appropriate salesperson of the financial institution A at a suitable timing. Send data. In this case, first, the data communication means 21 receives digital image data from the image input device 10.

  Subsequently, the form edge detection unit 22, the distortion amount calculation unit 23, and the distortion correction unit 24 perform distortion correction processing on the digital image data received from the image input device 10. Normally, when a sheet of a document such as a contract is photographed using a device that has a degree of freedom for the angle of view such as a digital camera, the sheet is not rectangular on the image, and the front is large depending on the distance. , Trapezoidal distortion occurs in the back side. In addition, there is also an inclination when it is oblique to the paper surface. The distortion correction processing in the present embodiment corrects such distortion by perspective transformation using the coordinates of the four corners of the form.

  In this case, the form edge detecting means 22 is means for calculating the coordinates of the four corners of the form on the image to be calculated by the distortion amount calculating means 23 by image processing. Specifically, first, the four sides of the form are detected, and a linear expression at the coordinates on the image is obtained. Thereafter, the coordinates of the four corners of the form can be calculated by obtaining the intersection of each straight line. As a four-sided detection method, an edge extraction filter represented by the Sobel operator is applied to the original image, and a known straight line detection method such as a Huff transform is applied to a point having a strong edge strength. To detect.

  Subsequently, the distortion amount calculation unit 23 calculates perspective transformation parameters from the coordinates of the four corners of the form calculated by the form edge detection unit 22. Here, the distortion amount calculating means 23 calculates parameters to be converted into an image without distortion obtained by photographing the original image from the front by applying the coordinates of the four points.

  Subsequently, the distortion correction unit 24 performs perspective transformation on each pixel of the original image based on the distortion correction parameter calculated by the distortion amount calculation unit 23 to correct the distortion of the image. In this way, by correcting the trapezoidal distortion and tilt caused by photographing from an oblique direction, an image similar to an image photographed by a device having a fixed angle of view such as a scanner can be obtained. Therefore, a conventional OCR process can be applied to an image photographed by a scanner.

  Subsequently, the character recognition unit 25 performs OCR processing on the image output by the distortion correction unit 24 and reads the written characters. In this case, a known OCR technique is used to access the character recognition dictionary database 27 and perform OCR processing. In the character recognition dictionary database 27, information such as form layout information, form format information, and a character recognition dictionary is registered in advance. This information can be accumulated by accumulating the form image obtained as described above and registering the form layout and form format each time a new form is entered. it can.

  The input image is compared with a plurality of layouts registered after analyzing the form layout, and form identification is performed. For the form identification, existing techniques such as a ruled line feature of a contract and a character string at a specific position are used. As a result, when it is recognized that the document is a specific form, character recognition is performed for each area defined in advance as a form format. Character recognition is processing for determining a character by performing character segmentation, feature extraction processing, and the like, and then performing feature matching processing with a plurality of registered character recognition dictionaries. Each area also has attributes such as financial institution name, product name, quantity, price, contract expiration date, interest rate, etc., so the contents of financial institution B's contract can be read by character recognition. Can do. The financial institution name can also be specified by the form identification result.

  Subsequently, the data storage unit 44 accumulates and stores various conditions such as the product name, contract expiration date, and interest rate of the contract read by the character recognition unit 25 in the proposal information database 46 together with the customer ID.

  Subsequently, the timing measurement unit 45 sequentially compares the information stored in the proposal information database 46 with the reference data stored in the notification reference database 47, and repeats the determination of the presence or absence of the proposal information notification at regular intervals. In addition, as shown in FIGS. 6 and 7, the notification standard database 47 includes information such as presence / absence of notification at the expiration date, notification condition formula at price / interest rate, jurisdiction information of sales staff, interest rate, exchange rate, price, etc. Necessary market price information is saved and updated. Then, the timing measuring means 45 collates the conditional expressions stored in the notification standard database 47 based on the product names, prices, interest rates, contract expiration dates, etc. for the plurality of customers stored in the proposal information database 46, Decide whether to notify.

  For example, when the contract is half a year before the expiration date and the price of the corresponding company stored in the inter-product correspondence database 28, that is, the product of the financial institution A is lower or the interest rate is higher. Conditional expressions that indicate that there is a notification are registered. When the notification is determined, the sales representative of the notification destination is determined according to the jurisdiction information in the notification reference database 47 based on the customer address and the product classification described in the proposal information database 46. And only in the case of notification, the process after the format conversion means 32 is started.

  Subsequently, the data conversion unit 26 performs the same operation as in the first embodiment, and based on the product name of the contract of the financial institution B read by the character recognition unit 25, based on the inter-product correspondence database 28, The product name of the corresponding financial institution A is obtained. In addition, instead of recognizing the character at each notification timing, the data that has been recognized once may be stored and reused.

  Subsequently, the data search means 31 performs the same operation as in the first embodiment, and uses the product name of the financial institution A replaced by the data conversion means 26 from the product name of the financial institution B as a key. 33 is accessed to search for the lowest price, features, etc. of the products sold by financial institution A.

  Subsequently, the format conversion means 32 reads the template of the notification document stored in the format information database 34 and performs processing for setting the data obtained by the data search means 31 to the corresponding item. As a result, a notification document including the appearance can be created.

  Subsequently, the charging unit 35 counts up the usage history counter of the charging database 37 for each process. As a result, pay-as-you-go billing based on the billing database 37 is possible, so that a business form in which the proposal design apparatus 20 is operated by an organization different from the financial institution A can be realized.

  Subsequently, the data communication unit 36 transmits the data of the notification document created by the proposal design device 20 to the proposal output device 40.

  Next, the operation of the proposal output apparatus 40 will be described. The proposal output device 40 outputs the notification document received from the proposal design device 20. In this case, first, the data communication unit 41 receives the data of the notification document transmitted by the proposal design apparatus 20.

  Subsequently, the data output means 42 prints and outputs the notification document automatically designed by the proposal design apparatus 20 on a medium. In addition, you may output by displaying on the application, browser, etc. on a display.

Next, a flowchart will be described.
Step S21: A customer contract is photographed and sent to the proposal design apparatus 20.
Step S22: The four sides of the contract image are detected.
Step S23: The distortion amount of the contract document image is calculated.
Step S24: The distortion of the contract document image is corrected.
Step S25: The layout of the contract is analyzed to identify the form.
Step S26: It is determined whether the form identification is successful. If the form identification is successful, the process proceeds to step S28, and if the form identification is not successful, the process proceeds to step S27.
Step S27 The form layout and the form format are additionally registered using the corrected image of the contract.
Step S28 Character recognition of each area (financial institution name, product name, quantity, price) is performed according to the form format corresponding to the specified contract type.
Step S29 The read information is stored in the proposal information database 46 together with the customer ID.
Step S30: Determine whether there is a notification at regular intervals.
Step S31: Whether or not to notify is determined. If not notified, the process returns to step S30. If notified, the process proceeds to step S32.
Step S32 Character recognition of each area (financial institution name, product name, quantity, price) is performed in accordance with the form format corresponding to the specified contract type.
Step S33 The read information is stored in the proposal information database 46 together with the customer ID.
Step S34 According to the template of the format information database 34, the acquired information is embedded to create a notification document.
Step S35: The usage history counter is counted up.
Step S36: Prints the notification document and ends the process.

  As described above, in this embodiment, the proposal design management system acquires the description items such as proposals, estimates, contracts, etc. created by other sellers as images, and sells the products It is designed to automatically determine and notify a suitable time and a suitable notification destination for making a proposal. Therefore, it is possible to make a proposal suitable for the customer in a timely manner and contribute to the improvement of the order accuracy.

  In the first embodiment and the second embodiment, the example in which the image input device 10, the proposal design device 20, and the proposal output device 40 are formed as separate devices has been described. The input device 10, the proposal design device 20, and the proposal output device 40 may be integrally formed as a single device. Further, although the example in which the proposal design apparatus 20 includes the charging unit 35 has been described, the charging unit 35 may be omitted.

  Furthermore, the person who owns the proposal design apparatus 20 and the person who owns the image input apparatus 10 and the proposal output apparatus 40 may be different persons. For example, in the first embodiment of the present invention, the proposal design apparatus 20 communicates with the image input apparatus 10 and the proposal output apparatus 40 via the data communication means 21 and the data communication means 36, so that the remote place Can be installed. Therefore, the sales staff can carry the image input device 10 and the proposal output device 40 and install the proposal design device 20 in the office centralized center for centralized management. In addition, the proposal design apparatus 20 may be an ASP (Application Service Provider) service managed by a manager or the like through the Internet.

  In addition, this invention is not limited to the said embodiment, It can change variously based on the meaning of this invention, and does not exclude them from the scope of the present invention.

It is a figure which shows the structure of the proposal design management system in the 1st Embodiment of this invention. It is a figure which shows the example of a data structure of the correspondence database between goods of the proposal design management system in the 1st Embodiment of this invention. It is a figure which shows the example of a data structure of the merchandise information database of the proposal design management system in the 1st Embodiment of this invention. It is a flowchart which shows operation | movement of the proposal design management system in the 1st Embodiment of this invention. It is a figure which shows the structure of the proposal design management system in the 2nd Embodiment of this invention. It is a 1st figure which shows the example of a data structure of the notification reference | standard database of the proposal design management system in the 2nd Embodiment of this invention. It is a 2nd figure which shows the example of a data structure of the notification reference | standard database of the proposal design management system in the 2nd Embodiment of this invention. It is a flowchart which shows operation | movement of the proposal design management system in the 2nd Embodiment of this invention.

Explanation of symbols

11 Image pickup means 22 Form edge detection means 23 Distortion amount calculation means 24 Distortion correction means 25 Character recognition means 26 Data conversion means 27 Character recognition dictionary database 28 Inter-product correspondence database 31 Data search means 32 Format conversion means 33 Product information database 34 Format information Database 42 Data output means 44 Data storage means 45 Timing measurement means 46 Proposal information database 47 Notification standard database

Claims (3)

(A) imaging means for inputting a form image;
(B) character recognition means for accessing the character recognition dictionary database and reading the characters of the form from the image;
(C) Data conversion means for accessing a product correspondence database and obtaining a product corresponding to the product name described in the form read by the character recognition means;
(D) data search means for accessing the product information database, searching for the product data obtained by the data conversion means, and determining the description content of the proposal;
(E) a format conversion unit that accesses the format information database and creates data of a proposal with the description content determined by the data search unit;
(F) A proposal design management system comprising data output means for outputting the data of the proposal. (A) imaging means for inputting a form image;
(B) character recognition means for accessing the character recognition dictionary database and reading the characters of the form from the image;
(C) data storage means for storing the product data described in the form read by the character recognition means in the proposal information database;
(D) timing measurement means for judging whether or not to notify the proposal information by sequentially comparing the product data stored in the proposal information database with the notification standards stored in the notification standard database;
(E) When notifying the proposal information, format conversion means for accessing the format information database and creating data of the notification letter for notifying the proposal information;
(F) A proposal design management system comprising data output means for outputting the data of the notice. (A) form edge detecting means for detecting coordinates of the four corners of the form on the image;
(B) distortion amount calculating means for calculating distortion correction parameters of the image based on the coordinates of the four corners;
(C) distortion correction means for correcting distortion of the image based on the distortion correction parameter;
(D) The proposal design management system according to claim 1 or 2, wherein the character recognition means reads a character of a form from an image with corrected distortion.

Images