JP2013033392A - Merchandise information provision system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable users and persons in charge in a shop to easily and quickly obtain appropriate merchandise information the harmonization degree of which is high for users.SOLUTION: A merchandise information provision system includes a user terminal and a server that stores and investigates merchandise information. The terminal includes: a photograph analyzing section that creates analysis results constituted by color distribution of pixel unit for each analysis reference color with respect to a photographic image of merchandise; an evaluation processing section that inputs evaluation of a user on merchandise the photographic image of which is displayed; a communication section that transmits the analysis results and evaluation of a user on the photographic image to the server; and a harmonization degree investigation requesting section that requests investigation of harmonization degree of the merchandise for which a user desires coordination. The server includes a purchase history information storage section that stores information of purchased merchandise for each user and a harmonization degree investigation execution section that calculates harmonization degree of the user who has requested investigation for merchandise satisfying prescribed conditions by using user evaluation information constituted by the received analysis results and evaluation of the user when an investigation request is received.

Description

  The present invention relates to a merchandise information providing system, and in particular, can provide information on merchandise that matches a user's preference, particularly information on clothing, using information on a large amount of merchandise included in an electronic book and purchase history information. It is related with the product information provision system which can be performed.

Traditionally, when searching for clothes that look good on you, based mainly on your own experiences and information that you happen to find, taking into account information that has been published in clothing magazines and websites, and clothing that you have purchased in the past. I chose clothes that matched my taste.
Also, consult with a store clerk at the store and consider the customer's tastes obtained through the customer's clothes and conversations with the customer while visiting the store. I was asked to propose an appropriate coordination for the design and color of the clothes.

In addition, at stores that store purchase history information for each customer, the customer uses the information on the clothing that the customer has purchased at the store in the past to determine the preferences of the customer and then purchases the clothing to be purchased. There was also a case of advice.
Furthermore, for all members registered at a certain store, we create a usage record database that pre-registers various wearing patterns that combine the purchased products and personal belongings of each member, and in response to requests from specific members, There has also been a system that provides information on clothing that is considered appropriate for the request by using utilization result data other than members (see FIGS. 3, 8, and 9 of Patent Document 1).

  In Patent Document 2, personal color information including evaluation scores indicating the degree of matching for combinations of color codes and information on past purchase clothing items for each customer in a clothing store are registered in a database in advance. In addition, a clothing sales system has been proposed in which a customer searches for a database by entering information on clothing that the customer wants to purchase or coordinate newly, and displays the customer's past purchased clothing in a suitable order. (See FIGS. 6 to 10 of Patent Document 2).

JP 2006-139706 A JP 2006-89883 A

However, if a customer owns a lot of clothes and wants to purchase new clothing that matches those belongings, it is difficult to recall the colors and designs of all the possessions at the time of purchase, There were many cases where it took time to select purchased items.
In addition, even if the customer's database stores historical information on clothing items that the customer has purchased in the past, the content of the historical information of the store alone is necessary and sufficient information for suggesting coordination. However, it is not always possible to propose products that can be appropriately coordinated with those purchased accessories.

  In addition, even if the data other than the specific member registered in the store is used to provide coordination information to the specific member, all of the wearing patterns other than the registered member Since it does not necessarily match the member's preferences, it is unclear whether the information provided is really appropriate for that particular member, and a lot of information with different preferences is registered. In many cases, appropriate information could not be provided.

  Therefore, the present invention has been made in consideration of the circumstances as described above, and in consideration of customer preferences, it is possible to appropriately harmonize the customer's and store staff with the products owned by the customer. An object of the present invention is to provide a product information providing system capable of selecting and proposing possible products.

The present invention includes a terminal owned by a user and a server that accumulates and investigates product information and provides desired information. The terminal and the server are connected via a network, and the terminal is connected to a display unit. A product storage unit that stores product information including a photographic image of the product, a photo analysis unit that generates an analysis result including a color distribution in pixel units for each predetermined analysis reference color with respect to the photographic image, and the display unit An evaluation processing unit that inputs a user's evaluation of the product of the photographic image displayed on the first communication unit that transmits an analysis result and a user's evaluation on the photographic image to the server; A purchase history information in which the server stores information on the products purchased for each user, the server including a harmony survey requesting unit that requests a survey of the degree of harmony for the products that the user desires to coordinate A storage unit; a second communication unit that receives an analysis result and user evaluation transmitted from the terminal; and a harmony degree survey request; and a user evaluation that stores the received analysis result and user evaluation as user evaluation information When the information storage unit and the harmony degree survey request are received, the harmony degree survey executing unit that calculates the harmony degree of the user who has requested the investigation for the product satisfying a predetermined condition using the user evaluation information A product information providing system characterized by comprising:
According to this, the information of the product purchased by the user is stored in the server, and the user's harmony with the product that the user desires to coordinate using the photograph image of the product transmitted from the terminal and the user's evaluation. Since it calculates, the product information useful for the user who has requested | required the harmony degree investigation can be obtained easily and rapidly.

Further, the server uses the harmony calculated by the harmony degree survey execution unit to select a product that is most preferable to be presented to the user from among products satisfying the predetermined condition, and information on the selected product. It is further characterized by further comprising a user response generation unit that transmits a response including the response to the terminal that has requested the survey.
According to this, in response to the user's harmony degree survey request, the information of the product that is most preferable to be presented to the user is sent, so the user can coordinate appropriately among a large number of products. Information about the product can be obtained easily and quickly.

Further, the predetermined condition is to make clothes manufactured for the purpose of wearing in the current season as a product to be investigated, and the harmony degree survey execution unit includes, among the clothes to be investigated, It is characterized by outputting a survey result indicating that clothes having a higher user evaluation have a higher degree of harmony with the user.
According to this, among the products manufactured for the current season, the user can obtain information related to products that are highly evaluated by the user and fit the user's preference.
In addition, the products satisfying the predetermined condition are limited to owned products already purchased by the user stored in the purchase history information storage unit.
According to this, it is possible to easily and quickly obtain information on a product that matches the user's preference from a group of products that the user has purchased and currently owns.

In addition, a store device owned by a store that sells the product is further provided, and the store device selects a product for which the degree of harmony for a specific user is to be investigated, and a product selected by the candidate selection unit. A store survey requesting unit that requests the server to investigate the degree of harmony of the specific user, and when the server receives the harmony degree survey request from the store device, it is stored in the user evaluation information storage unit. The store that has calculated the harmony degree of the specific user for the selected product using the user evaluation information of the specific user, generates an answer including the calculated harmony degree, and has made the survey request It is characterized by having a store answer generation unit that transmits to the apparatus.
According to this, when there is a request for checking the degree of harmony of a specific user for a selected product for which the degree of harmony is to be investigated from the store apparatus, the user evaluation information of the specific user stored in the server is used. Therefore, the person in charge at the store can easily and quickly obtain appropriate product information for a specific user.
Further, the product information stored in the product storage unit includes a photographic image of an electronic book distributed via a network or a photographic image of an electronic book acquired from a storage medium.

In the present invention, the analysis result is a histogram obtained for each analysis reference color, and the one histogram includes an analysis reference color name, a luminance value of the analysis reference color, and the number of pixels of the analysis reference color. It is characterized by.
Further, the degree of harmony is a correlation coefficient indicating a relationship between a histogram of products that satisfy the predetermined condition to be investigated and a histogram that is an analysis result of each product recorded in the user evaluation information storage unit, It is calculated from the product of the user's evaluation for the product satisfying the predetermined condition.

According to the present invention, information on the product purchased by the user is stored in the server, and the user for the product that the user desires to coordinate using the analysis result of the photograph image of the product transmitted from the terminal and the user's evaluation. Since the degree of harmony is calculated, it is possible to easily and quickly obtain information on an appropriate product having a high degree of harmony for the user who has requested the harmony degree survey.
In addition, when there is a request for a harmony degree survey for a specific user for the selected product from the store apparatus, the harmony degree is calculated using the user evaluation information of the specific user stored in the server. The person in charge can easily and quickly obtain appropriate product information for a specific user, and can provide useful information to the specific user.

It is a schematic block diagram of one Example of the merchandise information provision system of this invention. It is a block diagram of an embodiment of a user terminal (TE) of the present invention. It is a block diagram of an embodiment of the server (SV) of the present invention. 1 is a configuration block diagram of an embodiment of a store apparatus (PC) of the present invention. It is a flowchart of one Example of the evaluation input process performed with a user terminal (TE). It is a flowchart of one Example of the photo analysis process performed with a user terminal (TE). It is a flowchart of one Example of the harmony degree request | requirement process performed with a user terminal (TE). It is a flowchart of one Example of the user information collection process performed with a server (SV). It is a flowchart of one Example of the harmony degree investigation process performed with a server (SV). It is a flowchart of one Example of the store investigation request | requirement process performed in a store (PC). It is a flowchart of one Example of the investigation process performed when a server (SV) receives a store investigation request. It is explanatory drawing of one Example of the analysis result information produced | generated by the photograph analysis process of this invention. It is explanatory drawing of one Example of the purchase history information of this invention. It is explanatory drawing of one Example of the user evaluation information of this invention. It is explanatory drawing of one Example of the user reply information produced | generated with the server of this invention. It is explanatory drawing of one Example of the shop reply information produced | generated with the server of this invention. It is explanatory drawing of one Example of the investigation result displayed with the shop apparatus of this invention.

Hereinafter, the present invention will be described based on the embodiments shown in the drawings.
However, this does not limit the present invention.

<System configuration of the present invention>
FIG. 1 shows a schematic configuration diagram of an embodiment of a product information providing system of the present invention.
In FIG. 1, the system of the present invention is mainly composed of a user terminal (TE) 1, a server (SV) 2, and a store apparatus (PC) 3 that are connected to each other via a network 4.
As the network 4, various communication networks such as the Internet, a wired LAN, and a wireless LAN are used.

The user terminal (TE) 1 is a device owned by an individual user who is a user of this system, and for example, a personal computer, a mobile phone, a tablet terminal, or the like can be used. As will be described later, the user uses the user terminal 1 to input user information, product information purchased from a store, online browsing magazines, articles, evaluations of products published in catalogs, and the like. A request for investigating the degree of harmony (coordinating suitability) with the past purchased product is sent to the server 2 storing the history information of the purchased product.
The user terminal 1 may be limited to a specific personal computer or the like owned by the user. However, in order to connect to the server 2, user information including a predetermined user ID and password is input. The user terminal 1 to be used need not be limited to those owned by individuals.

The server (SV) 2 is an information processing apparatus that accumulates user information and product information of individuals who use this system, and performs surveys in response to requests from users and stores to provide information related to products.
For example, the server 2 performs a harmony survey (coordinate survey) on a predetermined product (especially clothing) in response to a request from the user, or a coordinate candidate product selected by the store in response to a request from the store. Conduct a coordinated survey.

The store apparatus (PC) 3 is installed in a store that sells products, and the selection of a coordinate candidate product recommended for a specific user from among the products handled by the store and the user's preference among a plurality of coordinate candidates And it is an apparatus which presents the goods considered to harmonize with the user.
Here, the store means a store that actually sells the actual product, a magazine company that provides a product catalog or sells products via the Internet, a store in a virtual shopping mall, and the like.

In the system of the present invention, when various information relating to products, users, etc. is accumulated in the server 2 and an inquiry request is transmitted from the user terminal (TE) 1 or store apparatus (PC) 3 to the server 2, the user TE1 or store Response information corresponding to the request from the PC 3 is generated, and the response information is transmitted to the requesting user or store.
The products handled in the present invention are not uniquely limited, but the products to be investigated in the following examples may be user's favorite items, for example, clothing items (hereinafter simply referred to as clothes). The target product.

For example, when the evaluation value of the clothing image of the magazine that user A is browsing online is input from the user TE1 and the harmony degree survey request is transmitted to the server SV2, the server SV2 is already stored in the server. Using the history information of clothes that the user A has purchased in a number of stores in the past, it calculates which clothes have the highest degree of harmony among the images of clothes for which investigation has been sought. Coordinate information (harmonic information about clothes) is transmitted.
As described above, by storing the purchase history information of the user's clothes on the server, the user is not limited to the purchase history at one store, and the degree of harmony with the user about the clothes that the user desires to coordinate with. Can be obtained easily and quickly.

In addition, for clothes handled by a certain store B, a survey request including an image of the coordinate candidate product is transmitted from the store PC 3 to the server SV2 in order to select an appropriate coordinate candidate product for a specific user C. In this case, the server SV2 uses the past purchase history information and user evaluation information of the user C already stored in the server to calculate the degree of harmony for each coordination candidate in consideration of the user's preference. Coordinate information including an index (weighted correlation coefficient) indicating which clothes should be recommended to the user C is transmitted to the PC 3.
As a result, the person in charge of the store can easily and quickly provide appropriate coordination information to the user C using the information of the product purchased by the user C at a store other than the store. Become.

<Configuration of user terminal TE>
FIG. 2 shows a configuration block diagram of an embodiment of the user terminal (TE) 1 of the present invention.
As shown in FIG. 2, the user terminal 1 (hereinafter abbreviated as “TE”) mainly includes a display unit 11, an input unit 12, a communication unit 13, a control unit 14, and a storage unit 15.
The TE1 may be a terminal having a communication function connectable to the server SV2, may be either wired or wireless communication, and may use any conventional communication protocol.
The display unit 11 is a part that displays information necessary to execute the function of TE1, and displays, for example, a photographic image of a product posted on an electronic book. As the display unit 11, various display devices such as LCD, organic EL, electronic paper, and plasma display can be used.

The input unit 12 is a part for inputting various types of information such as characters and symbols and requests to the server, and various input devices such as a covard, a mouse, and a touch panel can be used. Moreover, the display part 11 and the input part 12 may be integrated, and the hardware provided with both the input function and the display function may be provided like the tablet terminal and mobile telephone which are utilized today.
The communication unit 13 is a part that mainly communicates with the server 2 and corresponds to the first communication unit described above. For example, transmission / reception of the harmony degree request information transmitted to the server, the analysis result of the product photographic image, the user evaluation information, the server answer information received from the server, and the like is performed.
The communication unit 13 may use any currently used communication method as long as it uses a communication protocol and communication form unified with the server 2. In particular, when purchasing a product using an online store or providing coordinated information, it is preferable that communication is possible via the Internet.

The control unit 14 is a part that executes various functions of the TE1, and mainly uses a microcomputer including a CPU, a ROM, a RAM, an I / O controller, a timer, and the like.
The function of the user TE1 is realized when the CPU organically operates various hardware based on a program stored in advance in a ROM or the like.
When the operation of the control unit 14 is functionally classified, it mainly includes a photo analysis unit 14-1, an evaluation processing unit 14-2, and a harmony degree survey requesting unit 14-3.

The photo analysis unit 14-1 is a part that analyzes, for example, a photo image of a product published in an electronic catalog or electronic magazine distributed online. As will be described later, the analysis is performed by measuring the luminance value and the number of pixels (pixels) for each predetermined color component (analysis reference color) included in the photographic image, thereby obtaining an analysis result including a color distribution in pixel units. It means to generate. This analysis result is called a histogram.
The histogram of a photographic image is measured by being divided into, for example, three parts (head, upper body, and lower body), and is specified by information including a color name, a luminance value, and the number of pixels. That is, the analysis result is a histogram obtained for each analysis reference color, and one histogram includes the analysis reference color name, the luminance value of the analysis reference color, and the number of pixels of the analysis reference color. A histogram of one image is stored in the storage unit 15 as analysis result information 24. The photograph analysis unit 14-1 is a part that executes, for example, the process shown in FIG.

The evaluation processing unit 14-2 gives a degree of preference (referred to as evaluation or rating) of the user for each product image displayed in the electronic magazine or the like that the user is browsing in TE1, and performs the above-described photo analysis. It is a part which goes and transmits the evaluation information of the user for every goods to a server. This evaluation processing unit 14-2 is, for example, a part that executes processing shown in FIG.
In addition, the evaluation processing unit 14-2 inputs a user rating for the product of the photographic image displayed on the display unit.

  The harmony degree survey requesting unit 14-3 is activated based on a survey request input from the user, and investigates the user's harmony degree (coordinate information) for the product that the user desires to coordinate with the server 2. Is the part that requires When this survey request is transmitted to the server 2, the server checks the degree of harmony, and the server 2 returns the survey result (server answer information) to the TE1. This harmony degree investigation request | requirement part 14-3 is a part which performs the process shown in FIG. 7 mentioned later, for example.

  The storage unit 15 of the user TE1 is a part that stores information necessary for executing the function of the TE, and various storage devices such as a ROM, a RAM, a flash memory, and a hard disk are used. As shown in FIG. 2, the storage unit 15 mainly stores user information 21, electronic magazine information 22, analysis reference color information 23, analysis result information 24, transmission information 25, server answer information 26, and the like.

  The user information 21 is information that identifies a user who uses the TE1, and is information including a user ID, a user name, an address, an age, and the like, for example. In particular, at least the user ID and password are stored, the user is requested to input the user ID and password at the time of login, the user ID and password are confirmed, and the user who uses this TE1 is restricted. Is preferable in terms of security.

The electronic magazine information 22 stores electronic information such as product catalogs, fashion magazines, photographic images and articles related to products distributed via the network, and corresponds to product information stored in the product storage unit described above. To do. In particular, in the present invention, image information of product information included in the electronic magazine information 22 is used.
In addition, the product information includes not only an electronic book photo image distributed via a network but also information input by the user himself / herself and a photo image of an electronic book obtained from a storage medium such as a CD or a DVD. It may be.

  The analysis reference color information 23 is information relating to colors used in the photo analysis processing described later, and means information such as red, green, and blue, for example. In addition, as the number of colors used in the present invention, a large number of color components may be handled. However, if the number of colors to be handled increases, the time required for the analysis process increases. Therefore, it is preferable to limit the number of colors in order to obtain analysis results in real time as quickly as possible. For example, in addition to selecting an arbitrary color such as red, green, and blue, it is limited to 16 colors or 32 colors arbitrarily selected from colors often used in photographs published in magazine articles so far. That's fine.

The server reply information 26 is reply information returned from the server 2 in response to the survey request from TE1. The answer information 26 includes, for example, the user ID, the name of the investigated product, the photo number of the product, the degree of harmony of the product, and the like. This server answer information 26 corresponds to user answer information 53-4 described later. For example, FIG. 15 shows an example of the answer information.
The degree of product harmony is information indicating the degree of preference of the user, and means a scale indicating the suitability (coordinate suitability) between the product such as clothing and the user.

The analysis result information 24 is information generated by an analysis process performed by the photo analysis unit 14-1. It is mainly information consisting of a part name 24-1 and a histogram 24-2 obtained by processing as shown in FIG.
For example, as shown in FIG. 2, one histogram 24-2 includes a color name 31, a luminance value 32, and the number of pixels (pixels) 33, and is obtained for each preset analysis reference color. Further, the histogram is obtained for each product included in one image stored in the electronic magazine information 22.

In the following embodiments, assuming that clothes are to be analyzed, one product image is divided into images for each of three parts, and a histogram is obtained for each part.
Here, the three parts 24-1 are “head”, “upper body”, and “lower body”. That is, in photographic analysis, a product (clothing) in a photographic image is divided into three parts, and a histogram for each set color is calculated for each part. For example, when there are 16 analysis reference colors, 16 histograms (color, brightness, number of pixels) are calculated for each part. Further, when there are many images of products to be analyzed, photo analysis is performed for each product, and analysis result information 24 as shown in FIG. 2 is generated for each product.

The transmission information 25 is information transmitted from the TE 1 to the server 2 and corresponds to the user evaluation information 53-2 stored in the server 2 as shown in FIG.
The transmission information 25 is information generated in an evaluation input process shown in FIG. 5 to be described later, and mainly includes user information 41, a photo number 42, an analysis result, and a rating 44.
The user information 41 corresponds to the user information 21 and includes, for example, at least a user ID.
The photo number 42 is a number that identifies a photo of the product to be analyzed, and may be a mere number, but the magazine name, date, page number, and publication location (upper right, Lower left etc.) may be included.

The analysis result 43 corresponds to the analysis result information 24 for each photograph, and includes a histogram for each part.
The rating 44 means a user's evaluation of the product of the photographic image, and is an evaluation value (rating) given to the product by the user. In addition, when a page on which a photograph of a product is posted is turned without browsing for a certain period of time, an evaluation value of a predetermined level may be given.
This evaluation value serves as an index for determining whether the product is a favorite item for the user or not, and is used when the degree of harmony is calculated by the server, as will be described later.

<Configuration of server SV>
FIG. 3 shows a configuration block diagram of an embodiment of the server (SV) 2 of the present invention.
As shown in FIG. 3, the server 2 (hereinafter abbreviated as SV) mainly includes a communication unit 51, a control unit 52, and a storage unit 53.
The communication unit 51 is a part that performs data communication with the user terminal TE1 and the store apparatus PC3, and corresponds to the above-described second communication unit. For example, it is a part that receives the analysis result and user evaluation transmitted from the terminal TE1, and the harmony degree survey request. Data transmission / reception is performed between the server and the TE 1 and between the server and the PC 3 according to a predetermined communication format, communication protocol, and data format.

The control unit 52 is a part that uses a microcomputer including a CPU, a ROM, a RAM, an I / O controller, a timer, and the like and executes each function of the server. Functions shown in FIG. 3 are realized by operating various hardware based on a control program stored in a ROM or the like.
The functional blocks executed by the server 2 include, for example, a user information collection unit 61, a harmony degree survey execution unit 62, a user response generation unit 63, a request reception unit 64, a candidate image reception unit 65, a store response generation unit 66, and a correlation. There is a counting calculator 67.

The user information collection unit 61 is a part that collects information of users who use this system, and is a part that executes processing shown in FIG.
The user information collection unit 61 receives the information specifying the user input from the TE 1, the user evaluation information 53-2, and information on the product purchased by the user (purchased product information) and stores it in the storage unit 53.

The harmony degree investigation execution unit 62 is a part that is executed when a harmony degree investigation request is received from TE1, and is a part that executes processing shown in FIG. When the harmony degree survey request is received, as will be described later, the harmony degree of the user who has requested the investigation is calculated using the user evaluation information for a product that satisfies a predetermined condition.
For example, when a harmony degree survey request is received from a certain user A, the server 2 determines how many clothes suit the current season from the purchase history information 53-1 of the user A stored in the storage unit 53. In consideration of the preference of the user A, among the clothes owned by the user A, the clothes that are considered to be most appropriate with a high degree of harmony are presented to the user A.

Here, the predetermined condition is, for example, that clothing (clothes) manufactured for the purpose of wearing in the current season is set as a product to be investigated. In this case, the harmony degree survey execution unit outputs a survey result indicating that clothes with higher user evaluation among clothes to be investigated have higher degree of harmony with the user.
In addition, as a product satisfying a predetermined condition, for example, it may be limited to owned products already purchased by the user stored in the purchase history information.

  As described above, the degree of harmony is a numerical value indicating the degree of preference of the user. As shown in FIG. 9 and the like, the degree of harmony is stored in the server with a histogram of products that satisfy a predetermined condition to be investigated. It is a value calculated from the product of the correlation coefficient indicating the relationship with the histogram that is the analysis result of each product, which is one of the user evaluation information, and the user's evaluation of the product that satisfies the predetermined condition.

The user answer generation unit 63 is a part that generates an answer (user answer information 53-4) to be provided to the user and transmits the answer to the terminal TE1 that has requested the investigation. Using the degree of harmony calculated by the harmony degree survey execution unit 62, an answer including information on the item that is most preferably presented to the user among the items satisfying the predetermined condition is generated. Some of these answers include, for example, user information, information on the name of the product (clothes) selected as the survey target and information on the photo, the degree of harmony of the product, information on the clothing that is considered to be the most harmonious among the surveyed clothing, and selection Information such as a list of clothes arranged in order of harmony may be included.
However, instead of including all information about the product to be surveyed in the answer, for example, select the product (clothing) that is most preferable to present to the user, that is, select the clothing that is considered to be the most harmonious to the user, and select it. Only product information may be included in the answer. The information shown in FIG. 15 indicates answer information 53-4 including only information on the most preferable product to be presented to the user.

The request receiving unit 64 is a part that receives requests from the TE 1 and the PC 3. For example, it is a part that confirms the content of the received information and checks whether a harmony degree survey request from TE1 is received or a store survey request from PC3 is received. When a harmony survey request from TE1 is received, a harmony survey process as shown in FIG. 9 is executed, an answer to the user is generated by the user answer generating unit 63, and the answer is returned to TE1. The
When a store survey request is received from the PC 3, a store survey request process as shown in FIG. 11 is executed, and a weighted correlation coefficient is calculated by the correlation coefficient calculation unit 67, as will be described later. The store answer generation unit 66 generates an answer to the store, and the answer is returned to the PC 3.

  The candidate image receiving unit 65 is a part that receives an image of a coordinate candidate product sent from the store PC 3. The image of the coordinate candidate product is received together with a store survey request including information (user ID) of the user who requests the survey from the PC 3, for example.

The store response generation unit 66 is a part that generates a response (store response information 53-6) to be transmitted to the store PC 3 and transmits it to the store PC 3 that has made the survey request. Here, when the harmony degree survey request is received from the store PC 3, the harmony degree of the specific user for the selected product is calculated using the user evaluation information of the particular user stored in the server, and this harmony degree is calculated. And an answer including the weighted correlation coefficient calculated by the correlation coefficient calculation unit 67 is generated.
In this answer, for example, information such as information for specifying a store, surveyed user information (user ID), coordinate candidate product name and photo number, product harmony, weighted correlation coefficient for each coordinated product, etc. included.
However, not all of these pieces of information may be included in the answer, but only a part of necessary information (weighted correlation coefficient) may be included.

  The correlation coefficient calculation unit 67 is a part that calculates a correlation coefficient for each coordinate candidate product that has received the survey request based on a request from the store PC 3. For example, the correlation coefficient calculation unit 67 performs processing shown in FIG. This is the part for calculating the correlation coefficient. As will be described later, the correlation coefficient is a numerical value obtained by a predetermined mathematical expression using an arithmetic average of predetermined information obtained from two histograms.

The storage unit 53 of the server SV2 is a part for storing information used for realizing the function executed by the control unit 52 and information generated by executing the function, and includes ROM, RAM, flash Memory, hard disk, etc. are used.
As information memorize | stored in the memory | storage part 53, as shown in FIG. 3, for example, purchase history information 53-1, user evaluation information 53-2, analysis reference color information 53-3, user answer information 53-4, store There are received information 53-5, store answer information 53-6, and the like.

The purchase history information 53-1 is information stored in the purchase history information storage unit described above, and is stored for each user. The user purchases a past article by viewing an article or magazine of a store using this system. This is a record of the information on the product.
FIG. 13 shows an explanatory diagram of one embodiment of the purchase history information 53-1 used in the present invention.
In FIG. 13, the purchase history information 53-1 includes, for example, user information (user ID) 71, a product name 72, and a product image 73.
As the product name 72, a product code or the like may be used in addition to the name for specifying the purchased product.
The product image 73 means an image file of a purchased product, and for example, a photographic image of a product published in a magazine is used. The purchase history information 53-1 may include a purchase date, a store name, and the like.

In the example of FIG. 13, for example, the user with the user ID “00001” has purchased three products in the past, and the image files of the three products are stored in association with the product names. Show.
The purchase history information 53-1 may be input by the user himself / herself from TE1, but when purchasing a product from an online store via the Internet, the purchase information on the product is automatically sent from the PC 3 to the server 2 at the time of purchase. May be transferred and stored in the purchase history information 53-1 of the server 2.

The user evaluation information 53-2 is information stored in the above-described user evaluation information storage unit, and mainly stores the received analysis results and user evaluations. This information is information corresponding to the transmission information 25 sent from TE1, and includes user information 81, photo number 82, analysis result 83, and evaluation 84.
FIG. 14 is an explanatory diagram of an example of the user evaluation information 53-2. The information 53-2 includes analysis results and evaluation information about products previously purchased by the user.
The analysis result 83 corresponds to the analysis result information 24 generated by TE1 as shown in FIG.
Therefore, the analysis result 83 is a histogram obtained by analyzing the photograph of the product to be investigated for each user, and the histogram divides the photograph of the product into three parts (head, upper body, and lower body). Are generated respectively.
One histogram is generated for each predetermined analysis reference color, and includes, for example, a color name (color code) 31, a luminance value 32, and a pixel number 33 as shown in FIGS.

In FIG. 14, for example, as a result of analyzing the photo number “CL-AAA01” of the product owned by the user with the user ID “00001”, there are a histogram relating to the head, a histogram relating to the upper part of the body, and a histogram relating to the lower part of the body. Further, it is shown that the rating value (rating) of the user for this product is “5.0”.
For the user with the user ID “00002”, analysis results are stored for two products (CL-AAA01, CL-DDD03). For each product, three parts (head, upper body, lower body) are stored. ) Is stored as an analysis result.
Furthermore, it is shown that the rating values (ratings) for the two products for the user (00002) are “2.5” and “4.0”, respectively.

Here, the product (CL-AAA01) is common to two users and the analysis result information is the same, but the user's evaluation for the same product is “5.0” and “2”, respectively. .5 "and different.
Such a difference in evaluation is due to the user's preference. If the higher numerical value is higher, the user with the user ID “00001” has a better feeling about this product (CL-AAA01). It can be said that it is clothes that look good on you.
Further, the content of the histogram of each part shown in the analysis result information of FIG. 14 is, for example, information including three parameters (color name, luminance value, number of pixels) similar to the content of the analysis result 24 shown in FIG. It is.

Further, the color to be analyzed in each part is determined by the analysis reference color information 53-3.
The analysis reference color information 53-3 in FIG. 3 is set with colors used in the photo analysis processing.
For example, colors such as red, green, and blue are preset as analysis reference colors. Photo analysis processing is executed separately for all the set reference colors, and a histogram is generated for each reference color.
The user reply information 53-4 stores the reply to be transmitted to TE1, and the reply content to the harmony degree survey request from TE1 is stored. This user response information 53-4 corresponds to the server response information 26 of TE1 described above.

The store reception information 53-5 stores information sent from the store PC 3, and is, for example, information related to a coordinate candidate product.
The store reception information 53-5 includes user information (user ID) of a user who requests a coordinate survey, and an image file of a product for which a survey is requested.
The store reply information 53-6 is a result of a survey conducted on the survey request from the store PC 3, and stores the contents of the reply sent back to the PC 3. The store reply information 53-6 includes, for example, a store ID, a user ID, a product name to be surveyed, an image file name of the product, a survey result (weighted correlation coefficient), and the like.

<Configuration of store apparatus PC>
FIG. 4 shows a configuration block diagram of an embodiment of the store apparatus (PC) 3 of the present invention.
As shown in FIG. 4, the store apparatus 3 (hereinafter abbreviated as “PC”) mainly includes a display unit 101, an input unit 102, a communication unit 103, a control unit 104, and a storage unit 105.
The PC 3 may be a device having a communication function connectable to the server 2 as in the case of TE1, and may be a personal computer or the like. As the display unit 101, the input unit 102, and the communication unit 103, those having functions equivalent to the display unit 11, the input unit 12, and the communication unit 13 of the user TE1, respectively, may be used.
The control unit 104 may use a microcomputer in the same manner as TE1. However, in the PC 3, the control unit 104 mainly includes a candidate selection unit 121, a store survey request unit 122, a correlation coefficient reception unit 123, and a survey result generation unit 124.

The candidate selection unit 121 is a part that selects a coordinate candidate product for which the degree of harmony for a specific user is to be investigated, and the product is selected by an input operation of a person in charge at the store.
The input product information is stored in the storage unit 105 as selection candidate information 113.
The store survey request unit 122 is a part that executes survey request processing for requesting a survey from the server, and investigates the degree of harmony of a specific user with respect to a coordinate candidate product selected and input by the candidate selection unit. Is the part that requests the server.

The store survey requesting unit 122 executes, for example, the process shown in FIG.
When making this survey request, information (server transmission information 114) including user ID, information on candidate products for coordination, etc. is generated and transmitted to the server 2.
The correlation coefficient receiving unit 123 has a part that receives an answer (server answer information 115) sent from the server 2 in response to the survey request.
This server reply information 115 corresponds to the store reply information 53-6 described above, and this information 115 includes a weighted correlation coefficient that is a result of investigation by the server.

The survey result generation unit 124 is a part that generates information shown to the person in charge of the store based on the survey result (weighted correlation coefficient) for each coordinate candidate product sent from the server.
For example, according to the standard rules set in advance at the store, the coordination candidate products are rearranged based on the received correlation coefficient, the order of the products to be presented to the user and the products to be presented are determined, and the survey results are arranged. After that, the survey results are displayed.
For example, an investigation result as shown in FIG. 17 is displayed on the display unit 101.

In FIG. 17, there are 20 coordination candidate products for which a survey is requested for the user with the user ID “00001”, and these surveyed products are arranged in descending order of the weighted correlation coefficient as a result of the survey. It is changed and displayed.
The person in charge of the store can determine which product should be recommended for the user with the user ID “00001” by looking at the product list. This survey result is calculated in consideration of not only the product purchased by the user in the store in the past but also the purchase history information of the user stored in the server and an evaluation value indicating the user's preference. Therefore, it is possible to present a coordinated product that is more preferable for the user.

As shown in FIG. 4, the storage unit 105 of the PC 3 mainly stores user information 111, product image information 112, selection candidate information 113, server transmission information 114, and server answer information 115.
The user information 111 is the same as the user information stored in the TE 1 or the server 2.
The product image information 112 is a photographic image of the product.
The selection candidate information 113 is information relating to a product selected as a coordination candidate, and includes a product name, a photographic image of the product, and the like.
The server transmission information 114 is information transmitted to the server 2 and corresponds to information included in the investigation request.
The server reply information 115 is a reply received from the server and corresponds to the store reply information 53-6 as described above.

<Evaluation input process in user terminal TE>
FIG. 5 shows a flowchart of an embodiment of the evaluation input process executed in the user terminal (TE1).
Here, the user mainly inputs the evaluation of each image while viewing the image (clothing photo) published in the electronic magazine distributed in TE1, and the product (clothing) included in each browsed photo. ), And transmission information including the analysis result (histogram) generated by the photo analysis process and the user's evaluation is generated and the transmission information is transmitted to the server 2.
These processes are mainly executed by the evaluation processing unit 14-2 and the photo analysis unit 14-1.

In step S1 of FIG. 5, the user first receives an image or article included in an electronic magazine or the like distributed from a merchandise sales company or a magazine publishing company via the communication unit 13, and receives a photograph of the received magazine or The article is displayed on the display unit 111 and browsed.
At this time, the image and article of the electronic magazine to be distributed are stored in the storage unit 15 as the electronic magazine information 22.
The user performs a predetermined operation input by depressing an operation key of the keyboard that is the input unit 12 or touching an icon part of the operation key displayed on the display unit 11, for example, selection of a distribution image, magazine Browse magazines while turning pages, enlarging or reducing images and characters, and moving display areas. The electronic magazine may be downloaded by selecting a content that the user wants to browse while viewing the homepage of the seller or the like, or may be periodically distributed from the seller by a regular subscription.

In step S <b> 2, the control unit 14 checks whether the currently displayed photo is held in a format that can be cut out. Whether or not the format can be cut out can be determined, for example, by checking the file header and confirming whether a header character string or data string for determining the file type is recorded.
The clippable format is a general-purpose format related to a book layout such as XMDF, EPUB, or PDF.
If the photo portion in the currently browsed page image is not in a format that can be cut out as one image file, the process proceeds to step S3, and an error message indicating that the service cannot be used is displayed on the display unit 11. The process is terminated.
On the other hand, if the photo is in a format that can be cut out, the process proceeds to step S4.

In step S4, it is checked whether a page turning operation has been executed by the user. This may be done by confirming whether or not a predetermined key input by the input unit 13 corresponding to “page turning” or a predetermined touch panel operation has been performed.
If a page turning operation has been performed, the process proceeds to step S14. If not, the process proceeds to step S5.
In step S5, the control unit 14 uses a predetermined timer to check whether or not a certain time (for example, 2 to 3 minutes) has passed while the currently displayed page remains in the browsing state.
If the certain time has not yet elapsed, the process returns to step S4, and if the certain time has elapsed, the process proceeds to step S6.

In step S <b> 14, before the page turning operation is executed, a process of cutting out the photograph placed on the page (previous page) displayed on the display screen is performed. In this photograph, for example, it is assumed that a model wearing clothes that are products to be analyzed is photographed.
For example, if the photo to be cut out is an image file in JPEG format, an appropriate photo number is added as a single image file. The data is stored in the storage unit 15.

If a plurality of photos are posted on one page, all of these photos may be stored as image files with different photo numbers added to each photo.
In addition, there are cases where multiple products or multiple models are displayed in one photo. In this case, one photo is scanned to recognize the area where each product is displayed. The photograph image may be divided for each product, and the divided images for each product may be stored as individual image files.
In the following, for ease of explanation, it is assumed that only one product or one model is photographed in one photograph.

  In step S15, an area where one model is displayed is extracted from the cut out photograph. That is, an area where there is a product (clothes) to be analyzed is separated, and image data of the area is acquired. This region extraction may use person recognition as performed in conventional image processing. Thereby, for example, image data of only a model wearing clothes is acquired and temporarily stored in the storage unit 15.

In step S16, photo analysis processing is performed.
In the photo analysis process, as will be described later, histograms are calculated for each of the three parts, and the analysis result information 24 shown in FIGS. 2 and 12 is generated.
Details of this photographic analysis process are shown in FIG. In the photo analysis process, the acquired image data of one model is divided into three parts, and a histogram (color, luminance value, number of pixels) for each predetermined analysis reference color is calculated for each part. .

  In step S17, it is checked whether or not the analysis processing of the photos of all models existing in the same page has been completed. If there are still photographs to be analyzed, the process returns to step S15. If the analysis process has been completed for all the photos, the process proceeds to step S18.

In step S18, a predetermined first level 1 rating value (Rating) is added to all the cut-out photographs. That is, in addition to the analysis result information 24 for each photo, the same evaluation value is automatically given to all the photos.
In this case, since the page turning operation is performed, the same evaluation value is assigned to all the photos existing on the page displayed before the page turning. Thus, the reason why the same evaluation value is given is to reflect browsing in the evaluation. If a reader dislikes a category, he / she does not open the page, and conversely, even if he / she does not have a particular interest in the individual photos on the page, Is presumed not to hate. Therefore, the same evaluation value is given to the page once displayed.
Here, assuming that five levels of evaluation criteria are used, for example, a value of 1.0 is assigned as the first evaluation value (level 1). This evaluation value is used as a rating 44 in the transmission information 25 sent to the server. After step S18, the process proceeds to step S13, where transmission information 25 to the server is generated and transmitted to the server 2.

By the way, in step S5, when a certain time has already passed with a certain page opened, the user is interested in the product in the photograph posted on the page, and the user himself may try to evaluate it. It is thought that there is sex. Therefore, when a predetermined time has elapsed, the process proceeds to step S6, and the evaluation target photo in the currently displayed page is cut out.
This cut-out process is the same as that performed in step S14.

  In step S7, an icon of an evaluation button is displayed on the display unit 11. The evaluation button is a button for inputting the user's preference for a specific photo in the page. For example, a display such as ★★★ indicating the number of evaluations by the number of ★ is performed for each photo. When the evaluation button portion is pressed down by the user, the number of ★ displayed is increased or decreased, or a dialog for allowing the user to input a numerical value is displayed, and the user's evaluation is input. Here, a specific evaluation value is input by the number of ★ that increases or decreases with the number of times the button is pressed or a numerical value directly input by the user from a dialog in which the user can input a numerical value, and is evaluated (rating) in the storage unit 15. ) 44.

In step S8, it is checked whether or not analysis processing has been completed for all the photos in the same page. This process is the same as step S17.
In step S9, it is checked whether or not the displayed evaluation button has been pressed. If the evaluation button is depressed, the process proceeds to step S10. If the evaluation button is not depressed, the process proceeds to step S11.
In step S11, it is checked whether or not a page turning operation has been executed. This is the same as step S4. If the page turning operation has been performed, the process proceeds to step S12. If the operation has not been performed, the process returns to step S9.

In step S10, photo analysis processing is performed. This is the same process as step S16, and an analysis process is performed on the target photo for which the user has pressed the evaluation button to calculate a histogram.
Further, since the evaluation button is pressed by the user's own intention in step S9, a rating such as 1.0 is given to the analyzed photograph with 5.0 as the maximum. This evaluation value is determined based on the user's subjectivity, based on the evaluation value input by the evaluation button displayed in step S7.

  Since the page turning operation has been performed in step S12, a second evaluation value of a predetermined level is added to the photo cut out in step S6. As the second evaluation value (level 2), for example, a value of 1.0 is set in a five-step evaluation standard. Here, the second evaluation value (level 2) is added with the same value as the first evaluation value (level 1).

After step S10 and step S12, both proceed to step S13. In step S13, transmission information 25 as shown in FIGS. 2 and 14 is generated and transmitted to the server 2.
Thereby, the analysis result (histogram) and rating (rating) about the photographic image of the product appearing in the magazine that the user was browsing in TE1 are collectively transmitted to the server 2 and stored in the storage unit 53 of the server 2 It is stored as user evaluation information 53-2 (see FIG. 14).
Here, the analysis result of the product that has just been browsed by the user but has not been purchased and the evaluation information of the user are stored. Further, these pieces of information are used when there is an investigation request from a store later.

<Photo analysis processing at user terminal TE>
FIG. 6 shows a flowchart of an embodiment of the photo analysis process executed in the user terminal (TE) 1.
This process shows the detailed contents of steps S10 and S16 in FIG. Mainly, the photo analysis unit 14-1 divides an image of a person or clothes included in a photo into three parts (head, upper body, lower body), and generates a histogram as an analysis result for each part. The transmission information 25 including the analysis result is generated.

In step S31, when there are a plurality of cut-out photos, one photo is selected and an image file of the photo is read out. The selection criteria are not particularly limited, and may be selected based on, for example, the cutout order.
Hereinafter, the process from step S32 to step S43 is performed for one selected photograph, and when there are a plurality of cut-out pictures, the process from step S32 to step S43 is repeated.

In step S32, the clothing image or model image is divided into three parts from the read photo image file. Here, the three parts are a head, an upper body, and a lower body. In order to divide into these three parts, for example, a conventional technique such as a feature point extraction method may be used.
Alternatively, a photograph may be displayed on the display unit 11 and divided into three regions by a user instruction operation.
Thereby, the read image file data is divided into three so as to correspond to the three parts.

In step S33, it is checked whether or not the following analysis (steps S34 to S42) has been completed for all of the plurality of colors in the analysis reference color information 23 stored in advance in the storage unit 15.
When the analysis for all the reference colors is completed, the process proceeds to step S43. On the other hand, if not completed, the process proceeds to step S34.

In step S43, the transmission information 25 including the analysis result (histogram) 24 generated by the analysis (steps S34 to S42) is generated, and the process ends.
At this time, if the evaluation is already set for the photograph currently being analyzed by pressing the evaluation button in step S9, the rating is included in the transmission information 25.

In steps S34, S35, and S36, a histogram for each analysis reference color is generated for the region “head”.
In steps S37, S38, and S39, a histogram for each analysis reference color is generated for the region “upper body”.
Further, in steps S40, S41, and S42, a histogram for each analysis reference color is generated for the region “lower body”.

First, in step S34, a “head” divided image file is selected and read. In step S35, a histogram is generated for each of the predetermined reference colors for the read divided image file. For example, if one of the reference colors is “red”, a red pixel is searched from the divided image file, and the luminance value of the pixel is obtained.
When there are a plurality of red pixels, the number of pixels (number of pixels) is calculated together with the luminance value. For example, if there are 40 red pixels having a luminance value of “255”, the information “color, luminance, number of pixels” = “red, 255, 40” is generated as a red histogram.
Here, assuming that the luminance level is expressed by 256 gradations from 0 to 255, the luminance value is given by any value from 0 to 255.

In step S 36, the histogram for each analysis reference color generated as described above is temporarily stored as analysis result information 24. If the set analysis reference colors are, for example, “red, blue, yellow”, the histograms as described above are generated for each of these three colors.
A more detailed analysis result can be obtained as the number of analysis reference colors set is larger. However, for example, about 16 colors are preferably set from the viewpoint of shortening the processing time.
For example, in the case of 16 colors, 16 histograms are generated for “head”.

In step S37, the “upper body” divided image file is selected, and in step S40, the “lower body” divided image file is selected, which is different from step S34.
The histogram generation processing in steps S38 and S41 is the same processing as step S35, except that the divided image file to be analyzed is different.
Steps S39 and S42 are the same processing as step S36, and the histograms generated in steps S38 and S41 are temporarily stored for each analysis reference color.

In this way, a series of processing from step S34 to S42 is executed for an image file of one photograph, histograms for all preset reference colors are generated, and stored as the analysis result 43 of the transmission information 25. (Step S43).
This transmission information 25 is transmitted to the server, stored as user evaluation information 53-2, and then used to calculate the degree of harmony when there is a harmony degree survey request from the user.

<Harmony request processing in user terminal TE>
In FIG. 7, the flowchart of one Example of the harmony degree request | requirement process in the user terminal (TE) 1 is shown.
Here, a case is shown in which the user requests the server 2 for a harmony degree survey (coordinate analysis) for clothes that are owned by the user and stored as purchase history information 53-1 of the server. As a result, the user can obtain one clue for selecting clothes suitable for the current season from the clothes owned by the user by looking at the coordinate information returned from the server 2.

In step S51, the evaluation processing unit 14-2 of the control unit executes an evaluation input process. Here, for example, in addition to performing the process shown in FIG. 5 for a product published in a new magazine, the user's saved in the purchase history information 53-1 already stored in the server. Evaluation input processing is performed for past purchased products.
Thereby, the evaluation value (rating) of the user is added about the goods (clothes) purchased in the past, and the transmission information 25 as shown in FIG. 2 is produced | generated.

In step S52, the harmony degree survey requesting unit 14-3 transmits user information and a harmony degree survey request (coordinate analysis request) to the server. Here, the user information 21 may be any information that identifies the user. For example, only the user ID may be sent.
Upon receiving this request, the server 2 executes processing as shown in FIG. 9 to be described later, generates reply information 53-4 for the user, and sends it back to the TE 1 that sent the survey request.

In step S53, the server 2 waits for an answer to be transmitted. If an answer is received, the process proceeds to step S54. Otherwise, the process proceeds to step S55.
In step S54, the received reply from the server is stored as server reply information 26, and the reply is instructed to the display unit 11, and the process is terminated.
On the other hand, in step S55, since an answer to the survey request could not be received, an error indicating that there is no answer from the server is displayed on the display unit 11, and the process is terminated.

<User information collection processing at the server>
FIG. 8 shows a flowchart of an embodiment of user information collection processing executed in the server.
Here, the user information collection unit 61 receives information transmitted from the user and stores it in the storage unit 53. Information transmitted from the user includes user evaluation information 53-2 corresponding to the transmission information 25 and information related to the purchased product.

In step S101, it is checked whether or not transmission information 25 corresponding to the user evaluation information is received from the user TE1. If this information is received, the process proceeds to step S102; otherwise, the process proceeds to step S103.
In step S103, it is checked whether or not purchased product information is transmitted from the user TE1. If this information is received, the process proceeds to step S104. If not, the process returns to step S101. Purchased product information includes user information, product names, product images, and the like.

  In step S102, the received user evaluation information 53-2 is stored in the storage unit 53, and the process ends. In step S104, the received purchased product information is stored as purchase history information 53-1, and the process ends.

<Harmony investigation process on server>
FIG. 9 shows a flowchart of an embodiment of the harmony degree survey process executed by the server SV2. Here, mainly in response to a request from the user TE1, an answer about the harmony degree of the clothes owned by the user is generated and returned to the TE1.
In step S111, the harmony degree investigation execution unit 62 checks whether or not the harmony degree investigation request sent from TE1 has been received. This request is a request shown in step S52 of FIG. 7, and includes at least a user ID.
If there is no reception, step S111 is looped. If there is reception, the process proceeds to step S112.

In step S112, the purchase history information 53-1 of the user corresponding to this user ID is searched using the received user ID. That is, all purchase history information including the user ID is extracted.
Also, in this embodiment, information (harmonicity) that helps the user to select clothes suitable for the season by examining the products (clothes) that are appropriate to wear in the current season and considering the user's preference. Shall be provided to the user.

Accordingly, in step S112, a product that matches the current season is selected from the extracted purchase history information 53-1.
Although not shown in FIG. 3, assuming that the purchase date is stored in the purchase history information 53-1, it is determined from which purchase date the purchased product is a product manufactured for which season. . For example, if the month indicated by the purchase date is January or February, it is determined that the season matching the purchased product is spring.
Therefore, if the current season is spring, a product that is considered to be clothes manufactured for the purpose of wearing in spring (a product for spring) and has a purchase date of January or February is selected.
Alternatively, the purchase history information 53-1 stores information (seasonal information) indicating which season the product is manufactured for the purpose of wearing, and based on this seasonal information, Seasonal products may be selected.
Thus, a product group that matches the current season is selected from a large number of purchased products.

In step S113, it is checked whether or not a predetermined number (N) or more of products that match the current season have been selected. The predetermined number of N values is not limited to a fixed value, and the setting may be changed. For example, N = 2 may be set. Alternatively, when the number of purchased products stored as a history is large, a larger number may be set.
However, if the number of selected products is one, the product can only be selected and there is no room for further investigation.
If the number of selected products is N (≧ 2) or more, the process proceeds to step S115, and if not, the process proceeds to step S114.
In step S114, the user response generation unit 63 generates an error response indicating that the survey was not performed because there are few products to be surveyed, and transmits the error response to TE1. Thereafter, the process ends.

In step S115, information on one product among the selected products is read from the purchase history information 53-1. Here, there is no particular standard as to which product is read out. For example, the information of one product may be read out sequentially in the selected order.
Further, the read information includes a product image 73 of the product. The product image 73 may be stored as history information when the product is purchased, for example.

  In the product image 73, information (part information) indicating which part of the three parts described above corresponds to which part is stored is also stored at the time of purchase. For example, if the purchased product is, for example, a T-shirt, “upper body” is set as part information at the time of purchase. When purchasing a product, user information (user ID), purchase date, season information, product name (code), product image, and part information are associated with each other and stored as purchase history information 53-1. Shall.

In step S116, the part information of the selected product is extracted from the read product information.
In step S117, it is checked whether or not the analysis for all the analysis reference colors has been completed for the read product image of the product. If this analysis is completed, the process proceeds to step S120, and if not, the process proceeds to step S118.

In step S118, a histogram for each luminance is generated for each analysis reference color. The histogram generation processing here is the same processing as that shown in step S35 of FIG. Thereby, based on the site | part information extracted by step S116, the histogram regarding only the site | part is produced | generated.
In step S119, the generated histogram is temporarily stored together with the extracted part information. In this case, a histogram for one extracted part of the analysis result information as shown in FIG. 12 is stored.
Thereafter, the process returns to step S117, and the processes of steps S118 and S119 are repeated for all the preset analysis reference colors.

  In step S120, it is checked whether or not histogram generation has been completed for all the products selected in step S112. If the generation of histograms for all selected products is completed, the process proceeds to step S121. When that is not right, it returns to step S115 and the process from step S115 to step S119 is repeated about the remaining selection goods.

In step S121, for the extracted part information, a correlation coefficient considering the analysis result 83 reflecting the user's preference is calculated for each selected product.
The analysis result reflecting the user's preference is analysis result information in the user evaluation information 53-2 as shown in FIG. 14, and specifically, information of a histogram generated for each part. Of the information constituting this histogram, the number of pixels 33 obtained for each color is used to calculate the correlation coefficient.
This correlation coefficient is a parameter obtained by, for example, the following formula, and is obtained by using an arithmetic average for the number of pixels calculated for each analysis reference color.

First, a set X and Y of the number of pixels is defined as follows.
(X, Y) = {(Xi, Yi)}, i = 1, 2,... N
X: Set of the number of pixels for each reference color for the product (clothes) selected in step S112 Y: Each reference color for all products (clothes) included in the user evaluation information 53-2 stored in the server Xi: Number of pixels by luminance for analysis reference color (i) in set X Yi: Number of pixels by luminance for one analysis reference color (i) in set Y i: Number specifying the analysis reference color (i = 1, 2,... N)
n: Number of analysis reference colors set

この相関係数Ｋは、ステップＳ１１２で選択したすべての商品について、その商品ごとに計算する。 For such a set of two pixels, the correlation coefficient K for one selected product is calculated by the following equation.

The correlation coefficient K is calculated for each product for all the products selected in step S112.

In step S122, the product of the correlation coefficient K obtained by the above formula for each selected product and the rating of the user is calculated. Here, the user rating is an evaluation value for the selected product for which the correlation coefficient K is obtained, and is a numerical value stored as the rating 84 of the user evaluation information 53-2.
The product of the correlation coefficient K and the evaluation means an evaluation that takes into account the user's preferences so far, and corresponds to the degree of harmony in consideration of the user's preferences.
A larger product (harmonicity) means a higher degree of preference, and a smaller value means a lower degree of preference.
If there are m selected products, m products are calculated. Therefore, it means that the selected product having the maximum product among the m products is the most recommended product for the user as the product (clothes) in the current season.

In step S123, the user response generation unit 63 generates response information 53-4 including information on a product having the maximum product among the selected products. Here, the answer information includes, for example, user information, recommended product names, images of the products, harmony (product), and the like.
In step S124, the generated user response information 53-4 is transmitted to the user TE1.
Through the above processing, the server can check the degree of harmony, and can easily and quickly present a product preferable for the user from among the purchased products owned by the user.

<Store survey request processing by store PC>
FIG. 10 shows a flowchart of an embodiment of the store survey request process executed in the store apparatus (PC) 3.
Here, the person in charge of the store mainly requests the server to investigate the degree of harmony with the user for the candidate product (coordinate candidate product) to be purchased for a specific user, In response to the request, an answer from the server is obtained, and information for selecting which coordination candidate product is good as a product to be presented to the user is displayed.

In step S <b> 201, the store survey request unit 122 checks whether or not an input indicating a store survey request has been made by the person in charge of the store PC 3.
When the store survey request is to recommend a product (clothes) A sold at the store to a certain user B, whether or not the product A is appropriate for the user B is evaluated by the evaluation information of the user B. Means to query the server that stores.

In order to request this investigation from the server, the person in charge of the store selects and inputs the user information 111 for identifying the user to be investigated and the image information 112 of the candidate product for coordination.
The image information 112 of the coordinate candidate product is input as selection candidate information 113. Moreover, as a coordination candidate product, not only one but also a plurality of products may be selected and input.
If there is no input, step S201 is looped. If there is an input, the process proceeds to step S202.
In step S <b> 202, the store survey request unit 122 transmits a survey request including user information of the user to be surveyed to the server 2. Here, a user ID is transmitted to the server as user information.

In step S203, the candidate selecting unit 121 selects and inputs the image information 112 of the coordinate candidate products to be investigated one by one.
In step S <b> 204, the coordinate product image information 112 selected and input is transmitted to the server 2. As a result, the server 2 that has received the survey request calculates the degree of harmony for the requested coordinate candidate product for the requested user and includes the weighted correlation coefficient, as shown in FIG. The reply information 53-6 is generated and transmitted to the requested PC 3.

In step S205, the correlation coefficient receiving unit 123 receives the response information 115 sent from the server. As shown in FIG. 16, the answer information 115 includes at least user information, information for specifying a coordinated product, and a weighted correlation coefficient of the product.
In step S <b> 206, the received answer information 115 is stored in the storage unit 105.

  In step S207, it is checked whether or not all the coordinate candidate products selected and input have been transmitted to the server 2. If information on all candidate products has been transmitted to the server 2, the process proceeds to step S208. If candidate products that have not been transmitted still remain, the process returns to step S203, and the processing from steps S203 to S206 is repeated.

  In step S208, the survey result generation unit 124 generates a survey result according to the store standard rule. Here, the store standard rule means a policy when recommending a product to a customer, recommending a product similar to the customer's past purchased product, or a product that best matches the customer's preference contents that have been investigated in advance. This means a policy such as recommending trendy products regardless of the preference contents that have been investigated in advance.

In order to generate the survey result, the coordinate candidate products are rearranged in the order according to the store standard rule using the weighted correlation coefficient for each candidate product included in the received answer information.
In other words, according to the store standard rule, rearrangement is performed so that the priority order of the product that the store wants to recommend most is higher.
For example, considering the user's preference, when arranging the coordinate candidate products in the order that the user would like, they are rearranged in descending order of the weighted correlation coefficient (see FIG. 17).

In step S209, the survey result generation unit 124 displays the coordinate candidate products in the sorted order.
FIG. 17 shows one example of the coordinate survey results. Here, a list in which candidate products are rearranged in descending order of the weighted correlation coefficient is shown. The person in charge of the store can determine a product recommended to the user specified by the user information by looking at the product list.

<Store survey request processing in server SV>
Here, processing performed by the server when the server SV2 receives a store survey request from the store PC 3 will be described.
FIG. 11 shows a flowchart of an investigation process executed when the server receives a store investigation request.

In step S151, the request receiving unit 64 checks whether or not a store survey request has been received from the store PC3. If not received, step S151 is looped. If received, the process proceeds to step S152.
When the store survey request is received, the candidate image receiving unit 65 receives the image information of the coordinate candidate product and stores it as the store reception information 53-5. In the store reception information 53-5, information (user ID) for specifying a user to be investigated is also stored.

  In step S152, the received image of the coordinate candidate product is divided for each part. This process is the same as step S32 of the photo analysis process of FIG. Thereby, the product image data is divided into three parts (head, upper body, and lower body).

In steps S153 to S161, a histogram is generated and stored for each of the three parts. These processes are the same as the processes in steps S34 to S42 shown in FIG.
In steps S153, S154, and S155, a histogram for each analysis reference color is generated for the head image data, and a histogram (color name, luminance value, number of pixels) regarding the head as shown in FIG. 12 is stored. The
In steps S156, S157, and S158, a histogram is generated and stored for the upper body image data.
In steps S159, S160, and S161, a histogram is generated and stored for the lower body image data.
Through these processes, a histogram for one coordinate candidate product is generated and stored for each part.

  In step S162, the correlation coefficient calculation unit 67 calculates a correlation coefficient in consideration of the user's preference for each part from the histogram generated for each part. This process is the same process as step S121 of the harmony degree investigation process shown in FIG.

A set X, Y of the number of pixels is defined as follows.
(X, Y) = {(Xi, Yi)}, i = 1, 2,... N
X: Set of number of pixels for each reference color for coordinate candidate product (clothes) Y: Pixel for each reference color for all products (clothes) included in user evaluation information 53-2 stored in server Number set Xi: Number of pixels by luminance for one analytical reference color (i) in set X Yi: Number of pixels by luminance for one analytical reference color (i) in set Y i: Analysis Number specifying the reference color (i = 1, 2,... N)
n: Number of analysis reference colors set

For the two sets X and Y of the number of pixels, the correlation coefficient K for one coordinate candidate product is calculated by the following equation.

The correlation coefficient K is calculated for each part of one coordinate candidate product.
That is, three correlation coefficients K (K1, K2, K3) of the head, the upper body, and the lower body are obtained for one candidate product.
In addition, the correlation coefficient K is calculated for all products that are coordination candidates.

In step S163, the correlation coefficient calculation unit 67 calculates the product (harmonicity) C of the correlation coefficient K calculated for each region and the evaluation R (rating) (C = K × R).
Head product: C1 = K1 × R
Upper body product: C2 = K2 × R
Lower body product: C3 = K3 × R
Here, the rating R (rating) is an evaluation value for a candidate product for coordination, and is obtained by rating a product purchased in the past or rating automatically added when browsing a page. It is. If there is a product that has already been purchased in the past by the user as a candidate for coordination, since the rating is already stored in the user evaluation information 53-2 of the server, the evaluation value is used.

In addition, when a product that the user has not yet purchased and has not browsed the page is proposed as a coordinate candidate product, the rating is set by setting the value of R to 0 for the candidate product, for example. Ask.
As described above, the product C of the correlation coefficient K and the evaluation R corresponds to the degree of harmony considering the user's preference.

In step S164, the correlation coefficient calculation unit 67 calculates the weighted correlation coefficient Ka by adding the product (harmonicity) for each of the three parts for each coordinate candidate product.
That is, the weighted correlation coefficient Ka (= C1 + C2 + C3) is obtained by adding the product of the head (C1), the product of the upper body (C2), and the product of the lower body (C3).
The weighted correlation coefficient Ka is calculated for each selected coordinate candidate product.

In step S165, the store answer generation unit 66 generates answer information 53-6 including weighted correlation coefficients of all the coordinate candidate products.
The answer information 53-6 to this store is, for example, information as shown in FIG. 16, and when there are a plurality of coordinate candidate products, information from the survey product name to the weighted correlation coefficient that is the survey result is Are included.

In step S166, the store response generation unit 66 transmits the response information 53-6 to the store to the store PC 3 that has requested the survey.
The store PC 3 receives the response information 53-6 (server response information 115) to the store, and generates a coordinate survey result as shown in FIG. Appropriate product information can be presented and distributed to the user.

1 User terminal (TE)
2 servers (SV)
3 Store equipment (PC)
DESCRIPTION OF SYMBOLS 4 Network 11 Display part 12 Input part 13 Communication part 14 Control part 14-1 Photograph analysis part 14-2 Evaluation processing part 14-3 Harmony degree investigation request part 15 Storage part 21 User information 22 Electronic magazine information 23 Analysis reference color information 24 Analysis result information 25 Transmission information 26 Server response information 51 Communication unit 52 Control unit 53 Storage unit 53-1 Purchase history information 52-2 User evaluation information 53-3 Analysis reference color information 53-4 User response information 53-5 Store reception information 53-6 Store Response Information 61 User Information Collection Unit 62 Harmony Survey Execution Unit 63 User Response Generation Unit 64 Request Reception Unit 65 Candidate Image Reception Unit 66 Store Response Generation Unit 67 Correlation Coefficient Calculation Unit 101 Display Unit 102 Input Unit 103 Communication Unit 104 Control unit 105 Storage unit 111 User information 112 Product image information 113 Selection candidate information 14 server transmission information 115 server response information 121 candidate selecting section 122 stores examination request unit 123 correlation coefficient receiving unit 124 Findings generator

Claims (8)

Comprising a terminal owned by a user and a server for storing and investigating product information and providing desired information, the terminal and the server being connected via a network;
The terminal stores a display unit, a product storage unit that stores product information including a photographic image of the product,
A photo analysis unit that generates an analysis result including a color distribution in units of pixels for each predetermined analysis reference color for the photo image;
An evaluation processing unit for inputting a user's evaluation for the product of the photographic image displayed on the display unit;
A first communication unit that transmits an analysis result for the photographic image and a user's evaluation to the server;
A harmony degree survey requesting unit that requests the server to investigate the degree of harmony for the product that the user desires to coordinate;
A purchase history information storage unit in which the server stores information on products purchased for each user;
A second communication unit that receives the analysis result and user evaluation transmitted from the terminal, and a harmony survey request;
A user evaluation information storage unit for storing the received analysis result and user evaluation as user evaluation information;
When the harmony degree survey request is received, a harmony degree survey execution unit that calculates the harmony degree of the user who has requested the investigation for the product satisfying a predetermined condition using the user evaluation information. Product information providing system characterized by The server selects the most preferable product to be presented to the user from among the products satisfying the predetermined condition using the harmony calculated by the harmony degree survey execution unit, and includes the information on the selected product The product information providing system according to claim 1, further comprising: a user response generation unit that transmits a response to a terminal that has made the survey request. The predetermined condition is to make clothes manufactured for the purpose of wearing in the current season as a commodity to be investigated, and the harmony degree investigation execution unit may include the user's The product information providing system according to claim 1, wherein a survey result indicating that a higher-rated clothing has a higher degree of harmony with the user is output. The product information providing system according to any one of claims 1 to 3, wherein the products satisfying the predetermined condition are limited to owned products already purchased by a user stored in the purchase history information storage unit. . It further comprises a store device owned by the store that sells the product,
A store for which the store apparatus requests the server to check the harmony degree of the specific user with respect to the product selected by the candidate selection unit for selecting the product for which the degree of harmony for the specific user is to be investigated, and the product selected by the candidate selection unit With a survey requesting department,
When the server receives the harmony degree survey request from the store apparatus, the specific user with respect to the selected product using the user evaluation information of the specific user stored in the user evaluation information storage unit The merchandise information according to claim 1, further comprising: a store answer generation unit that calculates a degree of harmony of the store, generates an answer including the calculated degree of harmony, and transmits the answer to the store apparatus that has made the survey request. Offer system. 2. The product information stored in the product storage unit includes a photographic image of an electronic book distributed via a network or a photographic image of an electronic book acquired from a storage medium. Thru | or the merchandise information provision system in any one of 5. The analysis result is a histogram obtained for each analysis reference color, and the one histogram includes an analysis reference color name, a luminance value of the analysis reference color, and the number of pixels of the analysis reference color. The product information providing system according to any one of claims 1 to 6. The degree of harmony is a correlation coefficient indicating a relationship between a histogram of products that satisfy the predetermined condition to be investigated and a histogram that is an analysis result of each product recorded in the user evaluation information storage unit, and the predetermined The product information providing system according to claim 7, wherein the product information providing system is calculated from a product of a user's evaluation with respect to a product satisfying the above condition.

Images