JP4467719B2 - Database construction system and recording medium on which program is recorded - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a database construction system capable of constructing a database by collecting associated data of individuals prefer type eyeglass frame and the components each type of face the individual.
[0002]
[Prior art]
Traditionally, at a optician, when a customer visits, a store clerk recommends a spectacle frame that the customer likes, or recommends a spectacle frame that seems to suit the customer, but some of them are not necessarily recommended by the store clerk. Some customers do not.
[0003]
Such a customer looks into a display shelf of spectacle lenses and selects an appropriate spectacle frame, but a large number of spectacle frames are arranged on the display shelf, which is not necessarily preferable for the customer.
[0004]
In such a case, if the spectacle frame that seems to suit the customer or the spectacle frame that the customer seems to like can be automatically picked up, it will lead to shortening the selection time of the spectacle frame for the customer, and also to the spectacle store. The store clerk doesn't have to deal with customers more than necessary. In addition, when there are many customers coming to the store, there is a service limit to accepting multiple customers by one person, but if customers can automatically pick up eyeglass frames that they like, customer service will be improved. Can also be expected.
[0005]
[Problems to be solved by the invention]
To meet these demands, eyeglass stores have traditionally created a customer management database, but because the number of individuals is limited, it is necessary to grasp general trends between face types and eyeglass frames. Was inadequate. In addition, since the database is created by manual input, there is a problem that much time is required and an input error occurs. This problem exists not only in eyeglass stores, but also in handbag and shoe stores, accessory stores, and the like.
[0006]
The present invention has been made in view of the above problems, and can naturally collect information from a large number of individuals, not limited to customers who visit the store, and associates personal belongings with personal characteristics and personal characteristics. It is an object to provide a database construction system that can be constructed.
[0007]
[Means for Solving the Problems]
In order to solve the above-described problem, a database construction system according to claim 1 of the present invention is a web server that presents spectacle frame information, and a personal face image transmitted to the web server using a web browser. Extraction means for extracting the facial component from the information, and based on the facial component and information on the spectacle frame selected by the individual, the type of the personal face and the selected spectacle frame In order to collect data for associating, the face image of the individual is automatically decomposed into components and classified into types for each face component, and the type for each component of the face is selected. statistical analysis means for statistically analyzing the relationship between the eyeglass frame, and a storing means for storing in association with each component for each type and the selected spectacle frame of the face And wherein the door.
[0010]
According to the first aspect of the present invention, since the type of the face of the individual who accesses the homepage of the spectacle store and the type of spectacle frame preferred by the individual are related and classified, A database that correlates with the tendency of the type of eyeglass frame preferred by the individual can be constructed naturally and efficiently. In addition, since an extraction unit that extracts facial components from a face image is provided, facial components can be automatically extracted, and a large amount of databases can be constructed more efficiently.
[0012]
Database construction system according to claim 2, in database construction system according to claim 1, Bei display means for displaying the relationship between the chosen spectacle frame and the type of each element of the face in the frequency distribution Eteiru It is characterized by that.
[0013]
According to the second aspect of the present invention, since the display means is provided, the relationship between the type of each face component and the selected spectacle frame can be displayed in detail, and the personal preference tendency Can be easily grasped visually.
[0016]
The database construction system according to claim 3 is the database construction system according to claim 1 or 2 , wherein the facial components are facial contour, head hair, facial color, eyes, eyebrows, nose, ears, It is one of the mouth and chin.
[0017]
The database construction system according to claim 4 is characterized in that, in the database construction system according to claim 3 , at least the age and sex are transmitted together with the face image, and the age and sex are reflected in the construction of the database. To do.
[0018]
The database construction system according to claim 5 is characterized in that, in the database construction system according to claim 4 , the race information is transmitted together with the face image, and the race information is reflected in the construction of the database.
[0019]
Database construction system according to claim 6, in a database construction system according to any one of claims 3 to 5, the spectacle frame is selected, the shape, color, material, manufacturer name, designer name It is characterized by subdividing into any one of the components and associating it with the type of the facial component.
[0020]
According to the invention described in claim 6 , since the selected spectacle frame is subdivided into any one of its shape, color, material, manufacturer name, and designer name, the spectacle frame component Each can be classified in association with the face component type.
[0021]
Database construction system according to claim 7, in a database construction system according to any one of claims 3 to 5, the spectacle frame previously selected, the time difference between the selected spectacle frame after the Based on this, the degree of preference of the previously selected spectacle frame is determined.
[0022]
According to the seventh aspect of the present invention, since the degree of preference of the spectacle frame is determined based on the selection time during which the spectacle frame is selected, it is easy to determine how much the individual likes the spectacle frame. Can be judged quantitatively.
[0023]
Database construction system according to claim 8, in a database construction system according to any one of claims 3 to 5, after the eyeglass frame different from the spectacle frame selected previously is selected, again above When the same frame as the selected eyeglass frame is selected, it is assumed that the previously selected eyeglass frame is liked and the favorite degree is determined.
[0024]
According to the eighth aspect of the present invention, since the re-selected spectacle frame is regarded as a favorite and the favorite degree is determined, it is possible to easily quantitatively determine how much the individual likes the spectacle frame. .
[0027]
The program recording medium according to claim 9 is based on a program for executing an extraction step of extracting a facial component from a personal face image transmitted to a web server using a web browser, and an execution result of the extraction program. Statistically analyze the relationship between a program that executes a step of comparing and collating facial components with predetermined criteria and classifying them by component type, components classified by type, and selected spectacle frames A program for executing a statistical analysis step and a program for executing a storage step for storing a statistical analysis result are recorded.
[0028]
DETAILED DESCRIPTION OF THE INVENTION
In FIG. 1, 1 is a personal computer owned by an individual, 2 is a monitor, 3 is a keyboard, and 4 is a mouse. The personal computer 1 is connected to the Internet 6 via the provider 5.
[0029]
Reference numeral 7 denotes a spectacle store. The spectacle store 7 is provided with a customer management database server 8, a face image information storage database server 9, a spectacle frame information storage database server 10, a web server 11, and a television camera 12. Each server 8-11 is connected by LAN, and the television camera 12 is connected to the face image information storage database server 9.
[0030]
The web server 11 is connected to the Internet 6. The spectacle frame information is provided to the spectacle store 7 via the Internet 6 from the spectacle frame manufacturer 13, and the spectacle frame information is sequentially stored in the spectacle frame information storage database server 10 of the spectacle store 7.
[0031]
The spectacle store 7 has a homepage, and when an individual accesses the homepage of the spectacle store 7 using the web browser of the personal computer 1 and selects the spectacle frame fitting corner, the screen 2A of the monitor 2 is displayed, for example, As shown in FIG. 2A, when the spectacle frame fitting corner of the homepage of “XX Spectacle Store” is displayed and the selection screen of the spectacle frame 14 is designated, the screen transitions and a plurality of spectacle frames 14 are displayed on the monitor screen 2A. Is displayed.
[0032]
A screen for inputting the material, shape, size, color, manufacturer name, designer name, etc. of the spectacle frame is displayed on the screen before the spectacle frame 14 is directly displayed on the monitor screen 2A, and the material, shape, size, frame is displayed. color, after allowed to enter the manufacturer name or the like, may be configured to navigate to the selection screen of the spectacle frame 14.
[0033]
It should be noted that the spectacle frame color may be input on another screen instead of this screen. In addition, when it is desired to see a different spectacle frame in the selected spectacle frame, it may be possible to search and display what other spectacle frame is available.
[0034]
Here, the spectacle frame information storage database server 10 is provided in the spectacle store 7, but the spectacle frame information storage database server 10 is installed in the spectacle frame manufacturer 13 and the spectacle frame manufacturer is displayed on the homepage opened by the spectacle store 7. The 13 eyeglass frame information storage database servers 10 may be linked to display the eyeglass frame information on the monitor screen 2A.
[0035]
The image composition is performed on the server side of the spectacle store 7 and the synthesized image is transmitted to the personal computer 1.
[0036]
That is, as shown in FIG. 3, when the personal computer 1 accesses the homepage of the spectacle store 7 (S.1), the screen transitions to a face image input screen (S.2). (S.3), when the spectacle frame 14 is selected on the spectacle frame selection screen, the screen shifts to the image composition consent screen (S.4), where an individual such as an address, name, age, telephone number, password, etc. When the information is input (S.5) and the input is completed (S.6), the upload of the face image is executed (S.7).
[0037]
In the eyeglass store 7, the web server 11 combines the selected eyeglass frame 14 with the face image 15 by using the selected eyeglass frame 14 and the uploaded face image 15 (S.8). When a password is input from the personal computer 1 side (S.9), a face image 15 on which the selected spectacle frame 14 is tried is displayed on the monitor screen 2A as shown in FIG.
[0038]
The spectacle store further stores the personal component 22c, personal information 22b such as age and sex, and the selected spectacle frame 14 in association with each other (customer reserve army or person interested in spectacle frame interest). The relationship between the selected spectacle frame 14 and the spectacle database server 21 as a means, the personal database server 22 storing the personal face image 22a and personal information 22b, the personal face component 22c and the personal information 22b, and the selected spectacle frame 14 is statistically analyzed. The statistical analysis means 24 for analyzing is connected. The statistical analysis unit 24 includes an extraction unit 24a that extracts a personal face component 22c from the face image 22a, and is connected to the hard disk 23 as a temporary storage unit for temporarily storing a large amount of data during statistical analysis processing. Has been.
[0039]
The face image 22 a transmitted from the personal computer 1 is stored in the personal database server 22. The statistical analysis means 24 takes out the face image 22a stored in the personal database server 22, performs statistical analysis according to the flowchart shown in FIG. 4, associates the type of each face component 22c with the selected spectacle frame 14, The result is stored in the glasses database server 21. Since it is necessary to process a large amount of information during the statistical analysis, the data is temporarily stored in the hard disk 23.
[0040]
FIG. 4 is a diagram for explaining an example of a statistical analysis procedure by the statistical analysis means 24, and shows a procedure for statistically analyzing the contour of the face component 22c.
[0041]
When the face image 22a is input to the statistical analysis means 24 (S.11), the extraction means 24a extracts the face outline by the image recognition program (S.12). The statistical analysis unit 24 performs statistical analysis while sequentially comparing and comparing the contour extracted from the face image 22a and the round face, square face, and thin surface determination criteria possessed by the statistical analysis unit 24 itself.
[0042]
That is, the inputted face image 22a is compared with the round face determination standard (S.13), and if it is determined that it meets the standard, the face image 22a is determined to be a round face (S.14). If the face image 22a cannot be determined to be a round face, the face image 22a is then compared with the square face determination standard (S.15), and if the face image 22a is determined to meet the standard, the face image 22a is determined to be a square face. (S.16). If it cannot be determined that the face is a square face, it is then compared with a thin surface determination standard (S.17), and if it is determined that it meets the standard, the face image 22a is determined to be a thin surface (S.18). If it cannot be determined that the surface is narrow, it is determined as other (S.19). As described above, the statistical analysis means 24 statistically analyzes the contours of the constituent elements 22c of the input face image 22a and classifies them into any type. The statistical analysis means 24 performs such statistical analysis each time a face image 22a is newly input.
[0043]
FIG. 5 is a conceptual diagram showing the concept of association between the component 22c of the face image 22a and the selected spectacle frame 14 by the statistical analysis means 24. As shown in FIG. 5, the statistical analysis unit 24 determines and classifies the face type of the individual not only the outline but also each component of the face, that is, hair, eyebrows, eyes, nose, mouth, and the like. Here, for example, the face image of male A as an individual has a short hair, contour is a square face, eyebrows are fine eyebrows, eyes are cut long, nose is vaginal, mouth is large, etc. Is determined and classified. If the eyeglass frames selected by the male A are frame numbers 3 and 12, the eyeglass frames of frame numbers 3 and 12 have short hair, square face outlines, narrow eyebrows, cut-length eyes, vaginal nose, It is associated with the type of each facial component, the large mouth.
[0044]
On the other hand, the face image of woman B as another individual is semi-long in terms of hair, semi-long in terms of hair, thin outlines, thin eyebrows, round eyes, high nose, small mouth, etc. If it is determined and classified, and this woman B selects glasses with frame numbers 25, 28, and 12, glasses frames with frame numbers 25, 28, and 12 have semi-long hair, thin outlines, thin eyebrows, circles Associated with each type of facial component: open eyes, high nose, small mouth.
[0045]
Here, both male A and female B have selected the eyeglass frame of frame number 12. Further, when comparing the constituent elements of the faces of male A and female B, only the eyebrows are determined to be thin eyebrows and are common. Thereby, the statistical analysis means 24 determines that “a person with a thin eyebrows prefers a spectacle frame of frame number 12”. The more these pieces of association information are, the higher the accuracy of the association is.
[0046]
In this way, each individual face component 22c and the selected spectacle frame 14 are statistically analyzed by the statistical analysis means 24, and the type of each face component and the selected spectacle frame are associated with each other in the database. Can be obtained, for example, as shown in FIG. 6, to obtain a frequency distribution of the association between the face component and the spectacle frame. By displaying this frequency distribution on the screen, the relationship between the type of each face component and the spectacle frame can be grasped instantaneously.
[0047]
Further, for example, by statistically analyzing and classifying based on the transmitted personal information 22b such as age, sex, race, etc., instead of the face component 22c, each age group, sex, race, etc. You can also know the type of glasses frame you like.
[0048]
When the relationship between the face component 22c and the spectacle frame 14 is displayed in a frequency distribution, the face component 22c and the spectacle frame 14 are associated with each other by manual input using a keyboard, a mouse, or the like. It is good also as making it input into the spectacles database server 21 as a preservation | save means. The statistical analysis means 24 can also automatically analyze without displaying the frequency distribution. For example, the statistical analysis unit 24 may be provided with a determination program that automatically determines a selected high-frequency spectacle frame, and associates the type of each face component with the spectacle frame.
[0049]
The eyeglass frame 14 associated with the face component 22c may be not every eyeglass frame number but every eyeglass frame component subdivided by shape, color, material, manufacturer name, designer name, and the like. In addition, when the spectacle frame is associated with the type of each individual facial component, if the amount of data to be processed is too inconvenient, only the characteristic components of the individual among the facial components 22c. By associating the spectacle frames, the processing data can be reduced, and the data processing time can be shortened.
[0050]
Further, as shown in FIG. 7, when an individual who accesses the homepage of the spectacle store selects a plurality of spectacle frames, the individual counts the selection time during which the spectacle frame is selected, so that the individual can select the spectacle frame. It is possible to determine how much the user likes the eyeglass frame and to easily and quantitatively grasp the degree of favorite of the selected spectacle frame (see S.38 to S.44 of S.31 to S.44). ).
[0051]
In the flowchart shown in FIG. 7, when an individual accesses the homepage of a spectacle store (S.31), inputs a face image (S.32), and consents to image composition (S.33), the screen transitions and the address , Name, age, password, telephone number, gender, race, etc. are entered (S.34). When the input of the personal information is completed (S.35) and the transmission of the face image and the personal information is completed, the screen transitions again and the spectacle frame selection screen is displayed. When the individual selects a spectacle frame (S.37), counting of the selection time of the selected spectacle frame is started (S.38), image composition is performed (S.39), and the individual uses a password. By inputting (S.40), a composite image of the selected spectacle frame and the face image is displayed (S.41). When the individual checks the composite image and then selects the next spectacle frame (S.42), the count of the selection time of the previously selected spectacle frame ends, and at the same time the selection time of the next selected spectacle frame Is started (S.43). The time difference between the count start time of the next selected spectacle frame and the count start time of the spectacle frame selected before is calculated and calculated as the selection time of the previously selected spectacle frame. Again, when the individual inputs the password (S.40), a composite image of the next selected spectacle frame and face image is displayed (S.41), and the individual determines whether or not to select the spectacle frame. Judgment is made (S.42).
[0052]
As described above, by repeating the selection of a plurality of spectacle frames, the selection time of each selected spectacle frame can be calculated, and the longer the selection time, the more the individual decides that he / she likes the spectacle frame. can do.
[0053]
In addition, when an individual who accesses a spectacle store homepage selects a plurality of spectacle frames, by managing the selection order and selection history of the selected spectacle frames, for example, After selecting a different spectacle frame, if you select the same spectacle frame that you selected earlier, you can consider that the individual likes the previous spectacle frame that you selected multiple times, and the selected spectacle frame You can also know how much you like.
[0054]
Furthermore, in the present invention, the selected spectacle frame 14 is tried on the face image 15, but as shown in FIG. 8, an individual accesses the homepage of the spectacle store 7 to try on the spectacle corner of the spectacle frame 14. , The face image 16 of the model, the hairstyle 17 and the like are displayed on the monitor screen 2A together with the plurality of eyeglass frames 14, and the eyeglass frame 14, the face image 16 of the model 16, the hairstyle 17 and the like are sequentially selected to display the composition. Then, the selected spectacle frame 14, model face image 16, and hairstyle 17 face image may be displayed on the monitor screen 2A.
[0055]
Here, if the facial image of the model is prepared with an emphasis on the facial feature, the aesthetic harmony relationship between the facial feature, the facial feature and the spectacle frame can be automatically collected. desirable. In this case, on the screen where the face image of the model and the selected eyeglass frame are combined, if the selection of matching or not matching is selected by clicking the radio button, the collected aesthetic harmony relationship Reliability is improved.
[0056]
The construction of such a database for glasses can be performed by an individual using a personal computer 1 at home or the like to access the homepage of a spectacle store. Since it can be obtained naturally and the association by statistical analysis and the storage of data can be performed almost automatically, almost no effort is required and an eyeglass database can be efficiently constructed.
[0058]
【The invention's effect】
According to the present invention, glasses for each type of component of the individual's face, since the type of spectacle frame the person likes are classified almost automatically associated to that person and the type of person's face prefer The tendency of frames can be easily grasped, and a large amount of eyeglass database can be constructed naturally and efficiently.
[Brief description of the drawings]
FIG. 1 is a conceptual diagram of a database construction system for glasses according to an embodiment of the present invention.
FIGS. 2A and 2B are explanatory diagrams showing a display state of a monitor screen of a personal computer according to an embodiment of the present invention, wherein FIG. 2A shows a state in which a plurality of eyeglass frames are displayed, and FIG. 3 shows a monitor screen on which an image obtained by combining a spectacle frame with a face image is displayed.
FIG. 3 is a flowchart for explaining a database construction procedure of the eyeglass database construction system according to the embodiment of the present invention.
FIG. 4 is a flowchart illustrating an example of a statistical analysis procedure of a statistical analysis unit according to an embodiment of the present invention.
FIG. 5 is an explanatory diagram for schematically explaining the association between the constituent elements of the face image statistically analyzed by the statistical analysis means and the selected spectacle frame.
FIG. 6 is an example of a frequency distribution display of the eyeglass database construction system according to the embodiment of the present invention.
FIG. 7 is a flowchart for explaining another example of the database construction procedure of the eyeglass database construction system according to the embodiment of the present invention.
FIG. 8 is a diagram showing a spectacle frame and a model face image displayed on a monitor screen of a personal computer accessing a spectacle store homepage.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Personal computer 2 ... Monitor 2A ... Monitor screen 3 ... Keyboard 4 ... Mouse 5 ... Provider 6 ... Internet 7 ... Glasses store 8 ... Customer management database server 9 ... Face image information storage database server 10 ... Glasses frame information storage database server 11 ... Web server 12 ... TV camera 13 ... Spectacle frame manufacturer 14 ... Spectacle frame 15 ... Face image 16 ... Face image 17 ... Hairstyle 21 ... Spectacle database server (storage means)
22 ... Personal database server 22a ... Face image (personal face image)
22b ... Personal information 22c ... Face component 23 ... Hard disk (temporary storage means)
24 ... Statistical analysis means 24a ... Extraction means

Claims (9)

A web server for presenting information on the spectacle frame ;
Extraction means for extracting a facial component from a personal face image transmitted to the web server by using a web browser, the extracted facial component and information on the spectacle frame selected by the individual Based on this, the personal face image is automatically decomposed into components to collect data for associating the personal face type with the selected spectacle frame, and the facial component A statistical analysis means for statistically analyzing a relationship between the type of each face component and the selected spectacle frame ,
A database construction system comprising storage means for storing the type of each face component and the selected spectacle frame in association with each other. 2. The database construction system according to claim 1, further comprising display means for displaying a relationship between the type of each component of the face and the selected spectacle frame in a frequency distribution . The component of the face is any one of face outline, hair, face color, eyes, eyebrows, nose, ears, mouth, and chin. Database construction system. 4. The database construction system according to claim 3 , wherein at least age and sex are transmitted together with the face image, and the age and sex are reflected in construction of the database. 5. The database construction system according to claim 4, wherein race information is transmitted together with the face image, and the race information is reflected in construction of the database. 4. The selected spectacle frame is subdivided into any one of a shape, a color, a material, a manufacturer name, and a designer name, and associated with the type of the face component. 6. The database construction system according to any one of items 5 . 6. The preference degree of the previously selected spectacle frame is determined based on a time difference between the spectacle frame selected earlier and the spectacle frame selected later. 2. A database construction system for eyeglasses according to item 1 . After a spectacle frame different from the previously selected spectacle frame is selected, when the same spectacle frame as the previously selected spectacle frame is selected again, the previously selected spectacle frame is liked. The database construction system according to any one of claims 3 to 5, wherein the degree of likeness is determined on the assumption . A program for executing an extraction step of extracting a facial component from a personal face image transmitted to a web server using a web browser ;
A program for performing a step of comparing and collating the constituent elements of the face with a predetermined criterion based on the execution result of the extraction program, and classifying the constituent elements according to the type of the constituent elements;
A program for performing a statistical analysis step for statistically analyzing a relationship between the components classified by type and the selected spectacle frame;
A program recording medium on which a program for executing a storing step for storing the statistical analysis result is recorded.

Images