KR20190114084A - Membership management and visitor monitoring apparatus using facial recognition and the method thereof - Google Patents

Abstract

The present invention relates to a membership management and visitor monitoring apparatus using facial recognition, capable of performing membership management and visitor monitoring by generating facial recognition information as facial characteristic information, not an image, and replacing ID and passwords, and a method thereof. The membership management and visitor monitoring apparatus using facial recognition comprises: a photographing unit (100) transmitting the photographed image of visitors; and a facial recognition management unit (300), after extracting a facial area of a visitor from the photographed image transmitted from the photographing unit (100) or a recorded image, generating facial characteristic information which converts the facial area into number or text data forms, not image data, and applying the generated facial characteristic information as unique identifiers to membership management or monitoring for a person on the blacklist.

Description

MEMBERSHIP MANAGEMENT AND VISITOR MONITORING APPARATUS USING FACIAL RECOGNITION AND THE METHOD THEREOF}

The present invention relates to a member management and visitor monitoring technology using face recognition, and more particularly, to perform facebook management and visitor monitoring by generating face recognition information as face characteristic information instead of images to replace ID and password. The present invention relates to a member management and visitor monitoring apparatus and method using facial recognition.

In general, operators of for-profit or non-profit companies or institutions, such as manufacturers, service providers, distributors, retailers and retailers, recruit members, pay points, and provide additional services, such as events, to increase sales by securing fixed customers. Use marketing strategies to manage In addition, stores and public offices monitor a lot of visitors by installing a number of surveillance cameras to prevent and cope with accidents.

When the general public wants to receive additional services provided by such marketing activities, the general public registers their personal information, generates an ID and password, and receives a membership card provided for use of the additional services if necessary. Become a member by

In the case of the member management marketing of the prior art, it is time-consuming and inconvenient to perform member registration because the general public enters personal information and performs authentication during the member registration process. You may not lose.

In addition, since the customer information is entered at the time of membership registration, personal information is leaked when the member information is exposed to hacking.

Accordingly, the Republic of Korea Patent No. 10-1388439 discloses an 'e-mail based authentication and confirmation device and method that can minimize the exposure of personal information by generating an authentication key based on the email account.

In addition, in the prior art, it is common to install and operate a plurality of surveillance cameras in order to manage visitors in stores, government offices, and the like, but there is a problem that a monitor must always reside in a monitoring room.

Accordingly, the Republic of Korea Patent Publication No. 10-2012-0105885 transmits the video and audio obtained through the recording means and the microphone installed in the store to the user terminal through the web server, so that the user can manage the store without restrictions on the place. Initiate a 'store surveillance system' that enables

Republic of Korea Patent Registration 10-1388439 Republic of Korea Patent Publication No. 10-2012-0105885

However, in the case of the 'e-mail-based authentication device and method' of the above-mentioned prior art, the security of personal information can be improved. However, when registering as a member, the personal information must be entered, and if a user wants to use additional services, he / she has a membership card or ID. There was an inconvenience, such as having to remember the password.

In addition, the security of personal information is increased, but since personal information is input and recorded at the time of membership registration, there is a problem that cannot fundamentally solve the problem of personal information leakage.

And in the case of the "store monitoring system" of the above-described prior art, it provides a convenience to monitor the store regardless of the place, but there is a problem that can not monitor the person of interest. Accordingly, there is also a problem that does not properly respond to frequent theft actors, VIP customers that require prior response.

Accordingly, the present invention is to solve the above-mentioned problems of the prior art, by generating the facial recognition information as facial feature information rather than an image to replace the ID and password, the member subscribers without entering their personal information Membership can be registered using the feature information, and after membership registration, membership management such as point or event provision can be performed by simply performing facial recognition without using a separate membership card or membership number. It is an object of the present invention to provide an apparatus and method for monitoring members and monitoring visitors using face recognition, which significantly improves convenience of member registration and member management.

Another object of the present invention is to provide a member management and visitor monitoring apparatus and method using face recognition, by which entry of personal information is excluded from membership registration, thereby fundamentally preventing the leakage of personal information.

In addition, the present invention stores the facial feature information of the attention person and attribute information, such as the alerter or VIP, and then automatically identify and notify the alerter or VIP during the monitoring of visitors, thereby responding to the attention It is another object of the present invention to provide an apparatus and method for monitoring members and monitoring visitors using facial recognition that can easily perform the task.

Member management and visitor monitoring apparatus using the facial recognition of the present invention for achieving the above object,

A photographing unit 100 transmitting a photographed image of the visitor; And

After extracting the visitor's face region from the captured image or the recorded image transmitted from the photographing unit 100, the facial feature information is generated by converting the face region into a number or text data format instead of image data, and generating the facial feature information. It is characterized in that it is configured to include; facial recognition management unit 300 to apply to the member management or monitoring for the attention to the unique identifier.

The photographing unit 100,

When a signal that is confirmed to be accepted in the member's membership inquiry at the moment of interacting with the employee is input, a plurality of member management photographing to photograph the visitor to be registered and transmit the visitor image to the face recognition management unit 200. Unit 110; And

It may be configured to include; visitor monitoring photographing unit 120 for transmitting the image captured by monitoring all the internal area irrespective of the interaction with the staff to the facial recognition management unit 200.

The facial recognition management unit 200,

The face DB unit includes a member table 211 storing member information based on the facial feature information of the registered visitors and a table 212 of the person who stores the facial feature information and the attribute information of the interested person. 210);

After extracting a face region from the member management target visitor image transmitted from the photographing unit 100, extract facial feature data rather than image data, generate facial feature information, and output the same, and together with the authentication request signal from the photographing unit 100. The facial feature information generation unit 220 extracts and outputs facial feature information after extracting a facial region from the transmitted visitor image, and extracts and outputs facial feature information after extracting a facial region from images of general visitors other than the membership registration request. );

The facial feature information generation unit 220 stores the facial feature information generated by the registered member ID and password to perform member registration, and the facial feature information corresponding to the authentication request signal of the facial DB unit 210. A member management unit 230 which performs member authentication in comparison with the member facial feature information of the member table and performs member management including management of compensation; And

After receiving the facial feature information of all the visitors output from the facial feature information generation unit 220, and compared with the facial feature information stored in the attention table 212, if there is matching information, registered attribute information It may be configured to include; visitor monitoring unit 240 for outputting a warning.

The facial feature information generation unit 220,

A facial data set is generated by extracting a plurality of face sheets included in a plurality of visitor images of the same visitor.

After applying facial recognition algorithm to the generated facial data sets, we learn facial features,

Output the learned facial features as facial feature information,

After the facial feature information is generated, the gray scale image may be deleted.

Member management and visitor monitoring method using the facial recognition of the present invention for achieving the above object,

Face recognition management unit 300 including a photographing unit 100 and facial DB unit 210 and facial feature information generation unit 220, member management unit 230, visitor monitoring unit 240 and attribution information management unit 250 Member management and visitor monitoring method of the member management and visitor monitoring device using the face recognition configured to include,

A facial image obtaining process (S100), wherein the facial feature information generating unit 220 acquires facial images from the visitor photographed image transmitted from the photographing unit 100;

A facial feature information extracting process (S200) by which the facial feature information generator 220 extracts facial feature information by applying a learning algorithm to the acquired facial images;

The member manager 230 stores the facial feature information generated by the facial feature information generator 220 as a registered member ID and password to perform member registration, and provides facial feature information corresponding to the authentication request signal. The member management process (S300) or the visitor monitoring unit 240 performs member authentication in comparison with the member facial feature information of the member table of the face DB unit 210 and performs member management including management of compensation. After receiving the facial feature information of all the visitors output from the facial feature information generation unit 220, and compared with the facial feature information stored in the attention table 212, if there is matching information, registered attribute information Characterized in that it comprises a; facial recognition management process including one or more of the visitor monitoring process (S400) to output a warning.

The facial feature information extraction process (S200),

A facial data set is generated by extracting a plurality of face sheets included in a plurality of visitor images of the same visitor.

After applying facial recognition algorithm to the generated facial data sets, we learn facial features,

Output the learned facial features as facial feature information,

After the facial feature information is generated, the gray scale image may be deleted.

The present invention described above generates face recognition information as face characteristic information instead of images, so that the ID and password can be replaced, so that member subscribers can register using face feature information without inputting their personal information. After signing up for membership, the convenience of membership registration and member management is remarkably improved by enabling the member management such as providing points or events by simply performing facial recognition without using a separate membership card or member number. Provide the effect of improving.

In addition, the present invention provides an effect of remarkably improving the security of personal information by allowing the entry of personal information to be excluded from membership registration, thereby preventing the leakage of personal information.

In addition, the present invention stores the facial feature information of the attention person and attribute information, such as the alerter or VIP, and then automatically identify and notify the alerter or VIP during the monitoring of visitors, thereby responding to the attention It provides an effect that can be easily performed.

1 is a functional block diagram of a member management and visitor monitoring apparatus 1 using face recognition according to an embodiment of the present invention.
2 is a block diagram of a POS system as an example of a member management and visitor monitoring system using face recognition to which the member management and visitor monitoring apparatus 1 of FIG. 1 according to an embodiment of the present invention is applied.
Figure 3 is a flow chart showing the process of membership management and visitor monitoring method using facial recognition according to an embodiment of the present invention.
4 is a view showing a face image (face image) data set applied to face learning.
5 is a subroutine diagram showing a detailed processing of the member management process (S300) of the processing of FIG.
6 is a subroutine diagram illustrating a detailed process of the visitor monitoring process of the process of FIG.

In the following description of the present invention, detailed descriptions of well-known functions or configurations will be omitted when it is deemed that they may unnecessarily obscure the subject matter of the present invention.

Since the embodiments according to the concept of the present invention can be variously modified and have various forms, specific embodiments will be illustrated in the drawings and described in detail in the present specification or application. However, this is not intended to limit the embodiments in accordance with the concept of the present invention to a particular disclosed form, it is to be understood that the present invention includes all changes, equivalents, and substitutes included in the spirit and scope of the present invention.

When a component is referred to as being "connected" or "connected" to another component, it may be directly connected to or connected to that other component, but it may be understood that other components may be present in between. Should be. On the other hand, when a component is said to be "directly connected" or "directly connected" to another component, it should be understood that there is no other component in between. Other expressions describing the relationship between the components, such as "between" and "immediately between" or "neighboring to" and "directly neighboring", should be interpreted as well.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. As used herein, the terms "comprise" or "having" are intended to indicate that there is a feature, number, step, action, component, part, or combination thereof that is described, and that one or more other features or numbers are present. It should be understood that it does not exclude in advance the possibility of the presence or addition of steps, actions, components, parts or combinations thereof.

Hereinafter, with reference to the accompanying drawings showing an embodiment of the present invention will be described in more detail the present invention.

1 is a functional block diagram of a member management and visitor monitoring apparatus 1 using face recognition according to an embodiment of the present invention, and FIG. 2 is a member management and visitor monitoring apparatus 1 of FIG. 1 according to an embodiment of the present invention. ) Is a block diagram of POS system as a member management and visitor monitoring system using facial recognition.

As shown in FIG. 1, the member management and visitor monitoring apparatus 1 includes a photographing unit 100 and a face recognition managing unit 300 transmitting a photographed image of a visitor.

The photographing unit 100 includes a plurality of member management photographing units 110 photographing customers or visitors for member management and a plurality of visitor monitoring photographing units 120 photographing customers or visitors for visitor monitoring. do.

When the member management photographing unit 110 receives a signal confirming acceptance of the employee's membership inquiry at the moment of interacting with the employee, the member management photographing unit 110 photographs a visitor who is a member to register the visitor image to the face recognition management unit 200. Is configured to transmit). For example, in the POS system of FIG. 2, the member management photographing unit 110 may be configured to include cam cameras installed in the POS terminal 310 to perform a payment.

The visitor monitoring photographing unit 120 is configured to monitor all the internal areas regardless of the interaction with the staff and transmit the captured image to the facial recognition management unit 200. For example, in the POS system of FIG. 2, the POS system of FIG. 2 may be configured with a plurality of surveillance cameras installed to monitor the inside of a store.

The facial recognition management unit 200 extracts the face region of the visitor from the photographed image or the recorded image transmitted from the photographing unit 100, and generates facial feature information by converting the face region into numeric or text data format instead of image data. And, it is configured to perform member management or monitoring for the main body using the generated facial feature information as a unique identifier.

To this end, the facial recognition management unit 200, the facial DB unit 210, facial feature information generation unit 220, member management unit 230, visitor monitoring unit 240, attribute information management unit 250, communication unit 260 ), An input unit 270, a display unit 280, an output unit 290, and the like.

The face DB unit 210 may include a member table 211 storing member information based on the facial feature information of registered visitors, and a table 212 of the owner storing facial feature information and attribute information of the advisor. It is configured to include.

The facial feature information extracts a plurality of face sheets included in a plurality of visitor images of the same visitor to generate a face data set, and repeatedly applies a face recognition algorithm to the generated face data sets. This is the result value generated after learning facial features.

The critical property information for the owners is provided with 'borders' for distinguishing those who have performed history of theft, damage, etc., 'VIPs' which are given to provide special attention services, elderly people, children, and the disabled. Or 'help' to provide.

The facial feature information generation unit 220 extracts a face region from the member management target visitor image transmitted from the photographing unit 100, extracts facial feature data rather than image data, and generates and outputs facial feature information. After extracting the facial region from the visitor image transmitted with the authentication request signal at 100, the facial feature information is extracted and extracted, and the facial feature information is extracted after extracting the facial region from the images of general visitors other than the membership registration request. Is configured to output.

In this case, as described above, the facial feature information generation unit 220 extracts a plurality of face sheets included in a plurality of visitor images photographing the same visitor to generate a face data set, and generates a face data on the generated face data sets. After repeatedly applying a recognition algorithm to learn facial features, the learned facial features are output as facial feature information and stored in the face DB unit 210. After the facial feature information is generated, the gray scale image is deleted to prevent the leakage of personal information such as the leakage of personal portrait rights.

The member manager 230 stores the facial feature information generated by the facial feature information generator 220 as a registered member ID and password to perform member registration, and provides facial feature information corresponding to the authentication request signal. The member authentication is performed in comparison with the member facial feature information of the member table of the face DB unit 210 and is configured to perform member management including management of compensation. That is, in the case of membership registration by the member management unit 230, since it does not require the input of personal information other than photographing the member subscribers to generate facial feature information, the leakage of personal information can be fundamentally prevented. Will be.

The visitor monitoring unit 240 receives the facial feature information of all visitors output from the facial feature information generation unit 220, compares the facial feature information stored in the attention table 212, and then matches the information. If so, it is configured to output the registered attribute information as a warning. As a result, it is easy to carry out vigilance for the critical person in places such as stores, government offices, large event halls, theaters, restaurants, airports, terminals, history, subways, etc., and facilitates appropriate responses to seniors, the disabled, and senior personnel. Make it work.

The attribute information management unit 250 is configured to add the attribute information of interest to the facial feature information of the subjects to be monitored, and store it in the advisor table 212 according to an input of an administrator.

The communication unit 260 is configured so that the facial recognition management unit 200 provides a communication function with the manager's smartphone 4 or manager PC 5, POS terminal 310, the store management server 500 and the like.

The member management and visitor monitoring apparatus 1 having the above-described configuration is applied to the POS system including the POS terminal 300, the store management server 500, and the product master DB 600 as shown in FIG. 2. By doing so, it is possible to perform membership management or visitor monitoring through facial recognition in the store.

3 is a flowchart illustrating a process of a member management and visitor monitoring method using face recognition according to an embodiment of the present invention.

As shown in FIG. 3, the member management and visitor monitoring method using the face recognition includes a photographing unit 100, a face DB unit 210, a facial feature information generating unit 220, a member managing unit 230, and a visitor monitoring unit ( Facial recognition management unit 300 including a 240 and the attribute information management unit 250; facial image acquisition process performed by the member management and visitor monitoring apparatus 1 using the facial recognition configured to include (S100), facial Feature information extraction process (S200), member management process (S300) or visitor monitoring process (S400) is made.

Specifically, the facial feature information generation unit 220 performs a face image acquisition process (S100) of acquiring face images from the visitor photographed image transmitted from the photographing unit 100. 4 is a diagram illustrating a face image (face image) data set applied to face learning. The facial images of FIG. 4 are facial data sets for applying a facial recognition algorithm, and include facial images generated by applying 10 frames per second.

Next, the facial feature information generation unit 220 performs a facial feature information extraction process (S200) for extracting facial feature information by applying a learning algorithm to the acquired facial images. By using OpenCV, a real-time image processing library, it is possible to provide a face recognition function with guaranteed reliability and processing speed. Therefore, OpenCV library provides Haar feature type learning algorithm for face recognition algorithm, detects a plurality of face images using Haar feature algorithm, extracts feature from Haar Cascade method from plural image files, and uses YAML (YAML Ain). Facial feature information can be generated by converting to 't Markup Language). This eliminates the need to store a separate image file for identifying an individual, eliminating the violation of the law regarding the handling of personal information.

Next, the member manager 230 performs a member management process (S300), and the visitor monitoring unit 240 performs a visitor monitoring process (S400). Here, the member management process (S300) and the visitor monitoring process (S400) are embodiments of the face recognition management process of the present invention.

In the member management process (S300), the member manager 230 stores the facial feature information generated by the facial feature information generation unit 220 as a registered member ID and password to perform member registration, and the authentication request signal. The corresponding facial feature information is compared with the member facial feature information of the member table of the face DB unit 210 to perform member authentication, and perform member management including management of compensation.

FIG. 5 is a subroutine diagram illustrating a detailed processing of the member management process S300 among the processing of FIG. 3.

As shown in Figure 5, the member management process (S300), the member facial feature information existence determination process (S310), the member registration judgment process (S320), multiple facial image acquisition process (S330), facial feature learning process (S340), Facial feature data generation can be made, including a member registration process (S350), a member judgment process (S360) and a member compensation management process (S370).

In detail, the member manager 230 receives the facial feature information extracted from the member management target photographed image from the photographing unit 100, and then searches for the member table 211 of the facial DB unit 210 to detect the corresponding facial feature. The member facial feature information existence determination process (S310) of determining whether information exists is performed.

Next, the member management unit 230 performs a member registration determination process (S320) to determine whether the member registration signal is input from the manager in the state that there is no facial feature.

If the member registration signal is not input as a result of the determination of the member registration determination process (S320), the member management process (S300) is terminated and the process is returned. On the contrary, when a member registration signal is input, the member manager 230 outputs a facial feature information generation signal to the facial feature information generator 220, and the facial feature information generator 220 controls the photographing unit 100. By performing the plurality of face image acquisition process (S300) of obtaining a plurality of face images by acquiring a plurality of images to be registered.

Next, the facial feature information generation unit 220 performs a facial feature learning process (S340) for generating learning data for a plurality of facial images by applying a facial recognition algorithm.

Thereafter, after generating the facial feature data based on the learning data, the facial feature data generation member registration process (S350) is performed to register member information using the facial feature data as an ID and a password in the member table 211.

In addition, when it is determined that facial feature information exists in the facial feature information existence determination process (S310), the member judgment process 360 for determining whether a member is performed is performed.

If it is determined that the member judgment process (S360) is not a member, the member management process (S300) is terminated and the original processing is returned.

Unlike this. As a result of the determination of the member judgment process (S360), if the member is registered or the member registration is completed, after performing the member compensation management process (S370) for performing rewards, such as points to the existing member or subscribed member returns to the processing process.

Referring back to Figure 3, the visitor monitoring process (S400), after the visitor monitoring unit 240 receives the facial feature information of all visitors output from the facial feature information generation unit 220, the attention person After comparing the facial feature information stored in the table 212, if there is a matching information, it is a process of outputting the registered attribute information as a warning.

6 is a subroutine diagram illustrating a detailed process of the visitor monitoring process of the process of FIG. 3.

6, the visitor monitoring process (S400) is a facial feature information presence determination process (S410), boundary object determination process (S420), boundary attribute output process (S430), VIP determination process (S440), VIP attribute output It includes the process (S450) and attribute notification process (S460).

Specifically, the visitor monitoring unit 240 receives the facial feature information input from the facial feature information generation unit 220 and then searches the facial DB unit 210 to determine whether the corresponding facial feature information exists. The feature information existence determination process (S410) is performed.

If the facial feature information does not exist as a result of the determination, the process returns to the process of FIG.

On the contrary, when facial feature information exists, the boundary object determination process (S420) of extracting attribute information to determine whether the boundary object is performed is performed.

As a result of the determination of the boundary target determination process (S420), a boundary attribute extraction process (S430) for extracting boundary attributes is performed.

If the determination result of the boundary determination process (S420) is not the boundary object, the VIP determination process (S440) is performed.

If the determination result of the VIP determination process (S440) is not VIP, the process returns to the process of FIG. On the contrary, if it is determined as a VIP, a VIP attribute extraction process (S450) of extracting 'VIP' is performed.

When the boundary attribute is extracted or the VIP attribute is extracted, after performing the attribute notification process (S460) of notifying the extracted administrator information to the administrator terminal according to the corresponding attribute, the administrator returns to the process of FIG. 3. do.

The present invention described above can be utilized as a means of membership registration (membership registration) and authentication by using the facial feature information identified through the camera and facial recognition algorithm as a unique identifier, simply confirming the intention to receive membership registration or benefits It is possible to omit a process such as having a physical medium such as a separate card, a tag, a stamp, a coupon, or installing or manipulating an app on a smartphone, thereby greatly simplifying member management.

In addition, the present invention is to prevent the entry of the member information is unnecessary when the member registration, it is possible to fundamentally prevent the leakage of personal information.

In addition, if the present invention records additional personal information together with the feature of the present invention, for the customers who have managed to provide the marketing information for promotional purposes among the customers managed by using the facial feature information as identifiers, It can also be used for marketing information provided through the service, gift delivery, etc., so that it can provide wider scalability in terms of customer management services.

On the other hand, it is difficult for a customer to repeatedly deal with so-called black consumer damage to a retail store by using conventional management practices or customer management systems.

However, the present invention uses the facial feature information extracted by the facial recognition algorithm as the unique identification information, and records the history of malicious complaints or other equivalent business damages, so that the customer can be immediately identified later when the customer visits again. Therefore, it can be used as a standard for preparing appropriate measures in response.

In addition, the present invention can be expected to have an effect of strengthening the countermeasures against the in-store occurrence range.

If the retailer notices that the stolen goods are damaged or the property is damaged afterwards, the cause of the damage is checked through the CCTV installed in the store, and the face of the provider is saved as an image through the recorder's capture function and printed. They are often posted in internal staff spaces for recidivism or ex post notification.

However, it is not easy to pay attention to both hospitality, store management, and criminal screening, so this approach is less effective.

In the case of business damages such as theft and vandalism recorded on CCTV, using the facial feature information of the actor as an identifier and applying the recognition function to the entrance CCTV camera, at least it will be possible to prevent crime again and respond quickly.

Although the technical spirit of the present invention described above has been described in detail in a preferred embodiment, it should be noted that the above embodiment is for the purpose of description and not of limitation. In addition, those skilled in the art will understand that various embodiments are possible within the scope of the technical idea of the present invention. Therefore, the true technical protection scope of the present invention will be defined by the technical spirit of the appended claims.

1: Member management and visitor monitoring device using facial recognition
4: manager smartphone, 5: manager PC
200: facial recognition management
110: member management shooting unit (cam camera, etc.)
120: visitor monitoring unit (CCTV camera, etc.)

Claims (6)

A photographing unit 100 transmitting a photographed image of the visitor; And
After extracting the visitor's face region from the captured image or the recorded image transmitted from the photographing unit 100, the facial feature information is generated by converting the face region into a number or text data format instead of image data, and generating the facial feature information. Face recognition management unit 300 to apply to the member management or monitoring for the main body with a unique identifier; member management and visitor monitoring apparatus using a facial recognition configured to include.
The method according to claim 1, The photographing unit 100,
When a signal is confirmed to be accepted in the member's membership inquiry at the moment of interacting with the employee, a plurality of member management shots that photograph the visitor to be registered and transmits the visitor image to the face recognition management unit 200. Unit 110; And
Visitor monitoring and photographing unit 120 for monitoring all the internal area irrespective of the interaction with the staff and transmitting the captured image to the facial recognition management unit 200; Member management and visitors using facial recognition configured to include Monitoring device.
The method of claim 1, wherein the facial recognition management unit 200,
The face DB unit includes a member table 211 storing member information based on the facial feature information of the registered visitors and a table 212 of the person who stores the facial feature information and the attribute information of the interested person. 210);
After extracting the facial region from the member management target visitor image transmitted from the photographing unit 100, the facial feature data is extracted instead of the image data to generate and display facial feature information, and together with the authentication request signal from the photographing unit 100. The facial feature information generation unit 220 extracts and outputs facial feature information after extracting a facial region from the transmitted visitor image, and extracts and outputs facial feature information after extracting a facial region from images of general visitors other than the membership registration request. );
The facial feature information generation unit 220 stores the facial feature information generated by the registered member ID and password to perform member registration, and the facial feature information corresponding to the authentication request signal of the facial DB unit 210. A member management unit 230 which performs member authentication in comparison with the member facial feature information of the member table and performs member management including management of compensation; And
After receiving the facial feature information of all the visitors output from the facial feature information generation unit 220, and compared with the facial feature information stored in the attention table 212, if there is matching information, registered attribute information Visitor monitoring unit 240 for outputting a warning; membership management and visitor monitoring apparatus using facial recognition configured to include.
The method of claim 3, wherein the facial feature information generation unit 220,
A facial data set is generated by extracting a plurality of face sheets included in a plurality of visitor images of the same visitor.
After applying facial recognition algorithm to the generated facial data sets, we learn facial features,
Output the learned facial features as facial feature information,
A member management and visitor monitoring apparatus using face recognition configured to delete gray scale images after generation of facial feature information.
Face recognition management unit 300 including a photographing unit 100 and facial DB unit 210 and facial feature information generation unit 220, member management unit 230, visitor monitoring unit 240 and attribution information management unit 250 Member management and visitor monitoring method of the member management and visitor monitoring device using the face recognition configured to include,
A face image obtaining process (S100), wherein the face feature information generating unit 220 acquires face images from the visitor photographed image transmitted from the photographing unit 100;
A facial feature information extraction process (S200), wherein the facial feature information generator 220 extracts facial feature information by applying a learning algorithm to the acquired facial images; And
The member manager 230 stores the facial feature information generated by the facial feature information generator 220 as a registered member ID and password to perform member registration, and provides facial feature information corresponding to the authentication request signal. The member management process (S300) or the visitor monitoring unit 240 performs member authentication in comparison with the member facial feature information of the member table of the face DB unit 210 and performs member management including management of compensation. After receiving the facial feature information of all the visitors output from the facial feature information generation unit 220, and compared with the facial feature information stored in the attention table 212, if there is matching information, registered attribute information Face recognition management process including one or more of the visitor monitoring process (S400) to output a warning; Member management and visitor monitoring using the face recognition is configured to include Way.
The method of claim 5, wherein the facial feature information extraction process (S200),
A facial data set is generated by extracting a plurality of face sheets included in a plurality of visitor images of the same visitor.
After applying facial recognition algorithm to the generated facial data sets, we learn facial features,
Output the learned facial features as facial feature information,
Member management and visitor monitoring method using facial transplantation configured to delete grayscale images after generation of facial feature information.

Images