KR20000054824A - System for Searching Ideal Partners Using Face Image and Controlling Method - Google Patents

Abstract

PURPOSE: An ideal type search system using a facial image and control method therefor is provided to maximize information utilization by searching information having a high similarity to facial image information inputted from an authenticated user in a database having various facial image information and transmitting the searched information to the user along with other requested information, and to accurately search image information of an ideal type that the user wants by performing image searching after fine-adjusting parts of the facial image information so that a facial image matches the user's preference. CONSTITUTION: An ideal type search system using a facial image and control method therefor includes the following units. A plurality of user interface units(1) generates membership information and facial image information and searches required images and membership information. A web server system unit(2) searches membership information and optimized facial image information according to an image and character information search signal inputted from the plurality of user interface units(1). A central arithmetic processing unit (9) includes a data control unit (13), which analyzes inquiries, according the search of a user, inputted through a communication device unit (8) and controls the production of information of n of facial images.

Description

System for Searching Ideal Partners Using Face Image and Controlling Method}

The present invention relates to an ideal type search system using a face image and a method of controlling the same. More particularly, the present invention searches for image information having high similarity with respect to face image information input from an authenticated user in a database in which various face image information is stored. The user to select the information from n people with high similarity, and then use it to maximize the usability of the information. The present invention relates to an ideal search system using a face image and a control method thereof, which can be finely adjusted and stored in a database, thereby maximizing the processability of face image information.

In general, information retrieval systems for retrieving specific information generally provide a search service based on text information. As an example, looking at an information retrieval system executed by a marriage information company, the control method of such a system determines whether a specific user accesses the Internet site of the system in the site access check step (S1) as shown in FIG. If a specific user is connected to the Internet site, the initial menu screen transmission step (S2) is performed to display the initial menu screen on the terminal of the currently connected user.

After the initial menu screen transmission step S2, the process proceeds to a specific search determination step S3 to determine whether the current user inputs a signal for requesting a specific search on the initial menu screen. At this time, if the current user does not input a signal for requesting a specific search on the initial menu screen during the specific search determination step (S3), the user proceeds to the user information input step (S4). Personal information such as education, address, contact information etc. is input according to the program guide and stored in the database of the server.

On the other hand, if the current user inputs a signal requesting to search for a specific person on the initial menu screen during the specific person search determination step (S3), the process proceeds to the specific person search condition input step (S5) and the current user desires it. The search condition of a specific person is input sequentially according to the program guide.

After the specific search condition input step (S5), the search judgment step (S6) is performed to perform a search according to the condition input by the step (S5), and then there is a specific person meeting the search condition set by the user. To judge. At this time, as a result of the determination in the search determination step (S6), if there is no specific person in the database that matches the search condition set by the user, the process returns to the step (S3) and repeats the loop.

However, as a result of the determination in the search determination step S6, if a specific person matching the search condition set by the user exists in the database, the process proceeds to the information output step S7 and outputs the personal information on the selected specific person and ends. .

That is, the control method of the marriage information retrieval system as described above receives information on the ideal type of opposite sex as well as its own information at the time of membership registration or after registration and stores it in a database.

At this time, the user's picture may be input together, and the input picture information is not used for the ideal type search but merely provides additional image information.

In the text-based system, the search for the ideal type is performed based on the assumptions, educational background, etc. of the ideal type that the user desires, and the other party's information can be confirmed based on the search result.

However, the conventional information retrieval system as described above cannot search for an ideal type similar to the desired face type because the user performs a search based on text information and performs an ideal type matching the desired condition. Recently developed face image retrieval system also needs to generate a limited face shape by creating a combination of predetermined face elements, there is a problem that the user can not search for the ideal face image information.

Accordingly, the present invention has been invented to solve the above-mentioned problems, and searches for information similar to the face image information input from the authenticated user in a database in which various face image information is stored, and uses the retrieved information as other required information. By transmitting to the user along with the face image information of the top n people with high similarity, the object of the present invention is to provide an ideal search system and a control method using the face image that can maximize information usability. .

Another object of the present invention is to fine-tune a part of the face image information to match the user's preferences and then stored in the database, so the ideal search system using a face image that can maximize the processability of the face image information accordingly and Its purpose is to provide a control method.

In order to achieve the above object, the present invention provides a user interface unit for generating member registration information and face image information and searching for necessary images and member information, and the image and text information search signals inputted from the plurality of user interface units. Accordingly, the present invention provides an ideal type search system using a face image including a web server system unit for searching for membership registration information and optimal face image information and transmitting the same to the corresponding user interface unit.

In still another aspect of the present invention, in the idealized search system using a face image, when a user sets an information search function, an information search and communication execution step of transmitting the set text and image search information to a web server system unit, and the information search and A user authentication determination step of determining whether a user currently input from the user interface unit is an authenticated user when the user information including text and image search information is input from the user interface unit to the web server system unit by the communication execution step; As a result of judging during the authentication decision step, when the user currently connected to the web server system unit is an authenticated user, the character and image search information input from the current user interface unit are compared with the image search information stored in the database, and the number of n similarities is high. Video information video information day A character and face image information search execution step of searching in the database and outputting the retrieved n pieces of image information on a display of the user interface unit; and n having high similarity searched in the database after the character and face image information search execution step. If there is an ideal type that the user satisfies among the person image information, the personal information request determination step of judging whether the user wants to transmit the information about the personal image information that matches the person image information selected by the user among the searched n person images And when the user wants to transmit the information about the personal image information matching the person image information selected by the user as a result of the determination of the specific person information request, transmitting the personal image information corresponding to the selected person image information to the user interface unit. Specific Provided is a control method of an ideal search system using a face image formed by information transmission.

1 is an explanatory diagram illustrating a control method of a conventional marriage information retrieval system.

2 is a convex diagram illustrating an ideal search system of the present invention.

3 is a flowchart of the present invention.

4 is an explanatory diagram of a control point display screen for face image search information according to the present invention;

FIG. 5 is an explanatory diagram for explaining face image search information in which the control points of FIG. 4 are finely adjusted; FIG.

FIG. 6 is an explanatory diagram showing attributes of data included in a user information database and a video information database used for operating a system; FIG.

7 is an explanatory diagram of a feature vector of a face image used in an embodiment of the face image information search unit of the present invention;

<Detailed Description of Codes>

1: interface unit 2: web server system unit

3: key panel unit 4: face image information generating unit

5: database 6: face element adjuster

7: output device part 8: communication device part

9: central operation processing unit 10: user management unit

11: face image information retrieval unit 12: database unit

13: Data control unit 14: Temporary buffer

15: Romans 16: Ram

17: user information database 18: video information database

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

As shown in FIG. 2, the system of the present invention generates user registration information and face image information, and inputs user interface unit 1 for retrieving necessary images and member information. Web server system unit (2) for retrieving the member registration information and the top n face image information with high similarity according to the image and text information search signal and transmitting the same to the corresponding user interface unit (1) and overall control of the system of the present invention. Is made of.

In addition, the user interface unit 1 includes a key panel unit 3 for inputting a user's input control signal, for example, a member registration information and a face image information generation control signal, and the query via the key panel unit 3. Face image information generation of the user A face that generates a desired face image information based on an existing face or a virtual face image according to a control signal or generates face image information by using previously stored face images and face element information. And a database 5 for storing a plurality of virtual faces and face elements, for example, a celebrity or a plurality of predetermined virtual faces, which are connected to the image information generating unit 4 and the face image information generating unit 4. By using the control points predefined in the face elements of the face image information set by the face image information generating unit 4. A face element adjusting unit 6 for finely adjusting face elements and a query face image information or character information connected to one end of the face image information generating unit 4 and generated based on information input by a user. The communication device unit 8 which transmits a communication signal of text and image information input and output connected to one end of the output device unit 7 and the face image information generation unit 4 to the web server system unit 2. Is made of.

For example, the key panel unit 3 may be configured of a scanner, a keyboard, a digital camera, or the like so that a user can directly input a face image or information to be used as a query. The output unit unit 7 may be, for example, Monitors and printers can be used.

On the other hand, the web server system unit 2 is connected to the user interface unit 1 and the network communication device unit 8 for transmitting a communication signal of text and video information input and output, and the communication device unit 8 A central processing unit (9) for analyzing a query according to a user's text and image information search input via the user's request information through each processing device unit, for example, controlling the calculation of the upper n face image information having high similarity; According to the control command of the central processing unit 9, the user management unit 10 for charging the user according to the calculation of the request information of the authenticated user and the number of times of continuous management and retrieval, and the image retrieval control of the central processing unit 9 A face image information retrieval unit 11 for retrieving n face image information similar to a queried face image according to a signal, and a store for storing text and image information of the user. It consists of a one database portion 12.

In addition, the central processing unit 9 analyzes a query according to a user's text and image information search input through the communication device unit 8, and then requests user information through each processing unit, for example, n face image information. A data control unit 13 for controlling the calculation of the data, a temporary buffer 14 for temporarily storing data processed by the data control unit 13, and a ROM 15 for storing a drive program of the data control unit 13. And a RAM 16 for storing data generated during the execution of the function of the data controller 13.

In addition, the database unit 12 includes a user information database 17 storing user information, and an image information database 18 storing image information such as face image information.

The following describes embodiments of the method of the present invention applied to such a system.

According to an exemplary embodiment of the present invention, as shown in FIG. 3, the process proceeds to the information input determination step S2 in the initial state S1 to determine whether the current user's desired condition information or user's personal identification information is input. To judge. In this case, if the user does not execute the information input as a result of the determination in the information input determination step (S2), the process proceeds to the character and image information generation function determination step (S3) to allow the user to perform the image information generation function for searching for specific image information. Determine whether it is set. If the user does not set the image information generation function for retrieving the character and the specific image information within a predetermined time during the character and image information generation function determination step (S3), the process returns to the previous step (S1) and repeats the loop. Perform.

However, when the user sets the image information generation function for retrieving the character information and the specific image information as a result of the determination in the character and image information generation function determination step (S3), the character and image information direct input determination step (S4) is performed. Proceed to proceed with the image information that he wants to be optimal, for example, it is determined whether to directly enter the photo image information and query information containing the optimal image information to look for.

At this time, as a result of the character and image information direct input determination step (S4), as a result of the image information that the user currently wants to be optimal, for example, when directly inputting photographic image information and text query information including the optimal image information he wants to find The facial element satisfaction determination step S10 is performed.

However, when the character and image information direct input determination step (S4) is determined as a result, if the user does not directly input the image information that he / she wants to be optimal within a predetermined time, the process proceeds to the image DB use determination step (S6) and the current image information database. It is determined whether to set image search information using the stored image information.

At this time, when the image DB use determination step (S6) is determined as the image search information among the image information stored in the image information database, the process proceeds to the image information search step using the image DB (S7) and the image information stored in the image information database. Search for the video search information you are looking for.

However, if the user does not use the image information stored in the image information database as the image search information as a result of the determination during the image DB use determination step (S6), proceed to the virtual face type selection execution step (S8) to find the user. Image retrieval information For example, the ideal face information is set, for example, by using montage photograph information in which a face pattern is set in advance.

Meanwhile, after the image information retrieval step using the image DB (S7), the process proceeds to the image DB setting step S9 to determine whether there is image retrieval information that the user finds among the image information stored in the image information database, If does not exist in this video information database, the virtual face type selection execution step S8 is reached.

However, if the image search information is found among the image information stored in the image information database as the result of the determination in the image DB setting step S9, the information is set as the image search information and the face element satisfaction determination step S10 is performed. It is determined whether the face element of the currently set image search information is satisfactory.

If the face element of the currently set image retrieval information is satisfactory as determined in the face element satisfaction determination step (S10), the process proceeds to the information retrieval and communication execution step (S5) and transmits the currently set text and image retrieval information to the web server system unit. send.

However, if the face element of the currently set image retrieval information is not satisfactory as a result of the determination in the face element satisfaction determination step (S10), the process proceeds to the face element adjustment execution step (S11) so that some face elements of the currently set image retrieval information are themselves The control operation is performed to the desired shape, and the process proceeds to the information search and communication execution step S5 to transmit the currently set text and image search information to the web server system unit.

That is, when the user sets the image information generation command signal for searching for text and image information using the key panel unit 3 of the user interface unit 1, the face image information generation unit of the user interface unit 1 ( 4) recognizes this, and selectively receives information such as a query face image or query information from the user, for example, a face shape, a head length, whether glasses are worn, and a mouth shape.

In this case, if the user directly inputs the face image, the face image information generation unit 4 selects the input face image, otherwise the face image or image of an existing person such as an entertainer stored in the database 5 is already present. Select one of a plurality of specified virtual face images.

In this case, the face image information generation unit 4 replaces specific face elements such as eyes, nose, mouth, and face contour on the display screen of the output device 7 based on the desired face image. At this time, each face component is replaced using the components stored in the database 5 of the user interface unit 1.

Then, the face image information generating unit 4 performs fine adjustment through the face element adjusting unit 6, for example, when the set image search information is not satisfied by the user. The image search information is displayed on the display screen of the output device 7 as shown in FIG. Then, the user moves the control points of each facial element existing in the query face image using the key panel unit 3 on the displayed image search information divided into the shape of the control point displayed on the display screen of the output device 7. The element adjusting unit 6 recognizes this and finely adjusts the selected control point to complete the user's desired facial image.

Then, the face image information generation unit 4 inputs the set text and image search information to the communication device unit 8 of the web server system unit 2 through the communication device unit 8.

For example, as shown in FIG. 4, when the control point such as “A” is moved using the key panel unit 3, the face element adjusting unit 6 recognizes this and finely adjusts the image search information as shown in FIG. 5. .

Here, in the case of the image input by the user, the control points of the face elements may be automatically generated by the user interface, and the user may manually specify the fine points for more effective fine adjustment.

On the other hand, after the fine adjustment using the control points by the face element adjustment unit 6 is set to the weight of the face components to be used for the search.

In other words, each control point of the image search information is automatically given a pattern shape when the virtual face is selected in the above process, and can be designated and adjusted by the user when the photograph is input. You can also use the Control and Shift keys to change the left / right symmetry or maintain the aspect ratio. In this process, the control range of each control point is generated by using various threshold values to exclude various changes in the shape of the face.

On the other hand, as a result of the determination during the information input determination step (S2), if the user executes the information input proceeds to the general information input determination step (S12), the user is currently general information, for example, the ideal condition information or the user desired It is determined whether the personal information input function of the user is set.

In addition, when the user determines the registration information, for example, when the user sets the personal identification information input function as a result of the general information input determination step (S12), the information search and communication execution step is performed after proceeding to the general information input execution step (S13). Proceed to step S5 and transmit the registration information to the web server system unit.

On the other hand, as a result of the determination in the general information input determination step (S12), when the user sets the search information, for example, if the user sets the ideal condition information input function desired by the user, the process proceeds to the information search determination step (S14) and the current user performs the information search function. Determine whether it is set.

If the current user does not set the information search function as a result of the determination in the information search determination step S14, the process returns to the previous step S1 and repeats the loop.

However, if the current user sets the information search function as a result of the determination in the information search determination step (S14), the process proceeds to the information search and communication execution step (S5) and transmits the currently set text and image search information to the web server system unit. do.

Meanwhile, when the user information including text and image search information or registration information is input from the user interface unit to the web server system unit by the information search and communication execution step (S5), the process proceeds to the user authentication determination step (S15) and the current user. It is determined whether the user input from the interface unit is an authenticated user.

When the user authentication determination step (S15) determines that the user currently connected to the web server system unit is not an authenticated user, the general function execution step (S16) proceeds to the general function, for example, site introduction or membership invitation. It transmits the information to the user interface or stores the registration information received from the user.

However, as a result of the determination in the user authentication determination step (S15), if the user currently connected to the web server system unit is an authenticated user, proceed to the text and face image information search execution step (S17) and input from the current user interface unit. The text and image search information is compared with the image search information stored in the database, and the image information of an optimally similar image information database is searched and output on the display of the user interface unit.

On the other hand, after the character and face image information search execution step (S17) to proceed to the search information satisfaction determination step (S18) and according to the image search information requested by the current user of the high n similarity person image information searched in the database After transmitting to the user interface unit, it is determined whether the user satisfies such images.

If the user is dissatisfied with the n similarity image information having a high similarity with respect to the user query searched in the database, the process proceeds to the initial step S1.

However, as a result of the determination in the search information satisfaction determination step (S18), if there is information that the user is satisfied with the person image information for his query among the n person image information searched in the database, the specific person information request determination step (S19) is performed. Proceed to determine whether the user wants to transmit information on the personal information that matches the currently selected person image information.

In this case, as a result of the determination in the specific personal information request determination step (S19), if the user does not want to transmit the information about personal image information that matches the currently selected person image information, the process proceeds to the end step.

However, as a result of the determination in the specific personal information request determination step (S19), if the user wants to transmit information on personal image information that matches the currently selected person image information, the user proceeds to the specific information transmission step (S20) and the selected person image. Personal information corresponding to the information is transmitted to the user interface unit.

That is, when the user information including the text and image information search signal is transmitted to the communication device unit 8 of the web server system unit 2 through the communication device unit 8 of the user interface unit 1, the central processing unit ( The data control unit 13 of 9) recognizes this in accordance with the operation program contained in the ROM 15 and transmits the input user information as shown in FIG. 6 to the user management unit 10. FIG.

The user manager 10 then compares the user information stored in the user information database 17 to determine whether the user attempting to connect is a registered user. As a result of judging by the user management unit 10, if the user is a non-member, the user interface unit 1 receives a subscription invitation message that provides a description of the registration process, terms, and facial image search service. It displays on the output device part 7. However, as a result of the determination by the user manager 10, if the user is an authenticated member, the remaining number of searches along with the welcome message is transmitted to the user.

Meanwhile, when the current user information is authenticated by the user manager 10, the data controller 13 may retrieve image search information input from the current user interface unit 1 through the face image information search unit 11, for example, a query face image. The search is performed by comparing with the image information database 18. In this case, a general face image method may be used for the image information search, and one embodiment is performed as shown in FIG. The method of FIG. 7 converts the face image information stored in the image information database 18 into a feature vector value divided into partial regions of n × m, and the image retrieval information input from the user interface unit 1 is also n × m. The similarity is determined by converting the feature vector value divided into partial regions of and comparing the transformed two vector values.

The face image information retrieval unit 11 displays the result of the retrieval as in the above process on the output device unit 7 of the user interface unit 1 through the central processing unit 9.

Then, the user may look at the n face images received, and if there is a desired ideal type, the user may request personal identification information data on the person to the web server system unit 2. Therefore, when the central processing unit 9 of the web server system unit 2 requests information about a specific person from the user interface unit 1, the user management unit 10 uses the user management unit 10 to recall the personal information of the person. Transfer to the user interface unit (1).

Here, the temporary buffer 14 is a portion that is stored before processing the query face image or the face image and text information at the time of registration. The temporary buffer 14 is configured to ensure a large capacity, whereas the RAM 16 is a data controller 13 rather than a large capacity. ) And are processed to ensure fast access speeds.

Meanwhile, when the function of the face image information retrieval unit 11 is described again in the above process, the face image information retrieval unit 11 searches for a similar face image with respect to the face image queried by the user as shown in FIG. 7. It performs the function. The face image information retrieval unit 11 searches a face image of a registered user as a target, extracts feature information of a query face image, and compares the feature information of a registered member with the feature information of a registered member.

As a result of the search, the top n IDs are transmitted in the order most similar to the query face image, and the data control unit 13 of the central processing unit 9 receives this and transmits the face image of the corresponding ID to the user. When the user receives a search result and examines the face images and has a desired ideal type, the user may request detailed information of the person through the user management apparatus.

The face image registration or face image search is performed through the face image information search unit 11. Typical face image retrieval methods include a principal component method using a eigenvector, a template matching method, a method using a neural network, and the like. In the present invention, the face image retrieval apparatus may be one of the above methods.

FIG. 7 illustrates a feature vector generation method for performing registration or face image search by extracting a feature vector from an image as one embodiment of a face image search method.

In the case of face image search, a query for a face image can be performed purely as in the above procedure, and the search range can be reduced by using text information (the condition of the person to be searched). This information is also temporarily stored using the temporary buffer 14 of the central processing unit 9, and by limiting the information in the user information database 17, it is possible to attempt to search only the face image of the member meeting the corresponding condition.

As described above, the present invention searches for n pieces of information having a high similarity to the face image information input from an authenticated user in a database in which various face image information is stored, and transmits the retrieved information along with other request information to the user. By using the face image information with high similarity, it is possible to maximize the usability of the information accordingly, and finely adjust a part of the face image information to match the user's preferences, and then perform image search. Accordingly, there is an effect that the user can accurately search for the desired face image information.

Other features of the present invention operate a user information database that holds user's personal information and desired ideal information and an image information database storing only face images for efficient retrieval. The two databases are based on IDs assigned to users. Configured to perform synchronization.

By doing this, in addition to the conventional search using the text-based information, there is an effect that can search in the order of the ideal type most similar to the desired face image.

Claims (10)

A user interface unit which generates member registration information and face image information and retrieves necessary images and member information, and face image information having high membership registration information and similarity according to image and text information search signals inputted from the plurality of user interface units. Ideal search system using a face image, characterized in that consisting of a web server system for searching for and transmitting to the corresponding user interface. The method of claim 1, wherein the user interface unit generates a desired face image information based on an existing face or a virtual face image according to a query signal for controlling face image information generation of a user, or previously stored face images and faces. A face image information generation unit for generating face image information using element information, a database which is connected to the face image information generation unit and stores a plurality of virtual faces and face elements which are actually specified or are already designated; And a face element adjusting unit for finely adjusting each face element using control points predefined in face elements of the face image information set by the face image information generating unit, and connected to one end of the face image information generating unit. Communication signals of text and video information Ideal search system using a face image, characterized in that it comprises a communication device unit for transmitting to the web server system unit. The ideal type search using the face image of claim 2, wherein the face element adjusting unit adjusts each control point by adjusting control points for each face component such as eyes, nose, mouth, and face contour in the face image information. system. The apparatus of claim 1, wherein the web server system unit is connected to a network with the user interface unit to transmit a communication signal of text and image information input and output, and a text and image of a user input through the communication device unit. A central processing unit for controlling the calculation of n face image information having high similarity through each processing unit by analyzing a query according to information retrieval, and calculating and continuously managing request information of an authenticated user according to a control command of the central processing unit A user management unit for executing a; a face image information retrieval unit for retrieving face image information similar to a face image queried according to an image retrieval control signal of the central processing unit; and a database unit for storing text and image information of the user; Ideal search using facial image, characterized in that provided System. The data control unit of claim 4, wherein the central processing unit analyzes a query according to text and image information search of a user input through the communication unit and processes user request information through each processing unit, and the data control unit. And a temporary buffer for temporarily storing data processed by the target image. The ideal type search system according to claim 4, wherein the database unit comprises a user information database storing user information and an image information database storing face image information. The method of claim 4, wherein the face image information retrieval unit focuses on face elements by weighting each face element such as eyes, nose, mouth, and face contour of the query face image information during face image retrieval. Ideal search system using face images. When the user sets the information search function in the ideal type search system using the face image, an information search and communication execution step of transmitting the set text and image search information to the web server system unit, and the user interface by the information search and communication execution step. When the user information including text and image search information is input to the web server system unit, the user authentication determination step of determining whether the user currently input from the user interface unit is an authenticated user and the result of the determination during the user authentication determination step If the user connected to the web server system unit is an authenticated user, the text and image search information input from the current user interface unit are compared and analyzed with the image search information stored in the database. Searched in Then, the user is selected from the character and face image information retrieval step of outputting the retrieved n pieces of image information on the display of the user interface unit and the n person image information having high similarity searched in the database after the character and face image information retrieval step. If there is an ideal type satisfies the specific person information request determination step of judging whether the user wants to transmit information on the personal image information that matches the person image information selected by the user among the searched n person images, and the specific person information As a result of the determination during the request determination step, when the user wants to transmit the information about the personal image information corresponding to the person image information selected by the user, the user information is sent to the user interface unit. Made up The control method of the ideal type search system using the face image, characterized by. 10. The method of claim 8, wherein when the user sets the image information generation function for searching for specific image information during the information search and communication execution step and directly inputs the photo image information containing the optimal image information that the user wants to find, And a character and image information direct input determination step of directly transmitting the current image information directly input to the web server system unit. 10. The method according to claim 8 or 9, wherein the image information stored in the current image information database is stored when the image information that is determined to be optimal is not directly inputted within a predetermined time as a result of the character and image information direct input determination step. The image DB use determination step of determining whether to set the image search information by using the image information and the image information stored in the image information database when set to the image search information among the image information stored in the image information database If the user does not use the image information stored in the image information database as a result of the image information search step for searching whether there is the image search information that the user finds from the image DB use determination step, ideal type A virtual face type selection execution step of setting image retrieval information using montage photo information, etc., in which a face pattern is preset for face information of the face information, and among the image information stored in the image information database during the image information retrieval step using the image DB. If the image search information is found, the information is set as the image search information, and the face element satisfaction determination step of determining whether the face element of the currently set image search information is satisfactory and the face element satisfaction determination step are determined. If the face element of the image search information is satisfactory, the information search and communication execution step of transmitting the currently set text and image search information to the web server system unit and the face element satisfaction determination step are determined. If the element is not satisfactory, the current The control method of the image search system is ideal search their face some elements of the information using a face image, it characterized in that it further comprises a face element control execution step of executing a control operation in a desired shape.