Smart park access control system based on face recognition technology

By introducing feature matching and authenticity detection of multiple facial images taken in the access control system, combined with access frequency and level management, the problems of insufficient security and unfriendly access in the existing technology are solved, and higher security is achieved Sex and convenience.

CN118942178AInactive Publication Date: 2024-11-12HANGZHOU SHANGQI TECHNOLOGY CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202411411740.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-11
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing access control system based on facial recognition is not very secure and is not friendly to frequent visitors, so it requires manual confirmation of each entry.

Method used

设计了一种包括人脸图像摄录单元、数据处理单元、人脸信息存储单元、对比匹配单元、检测单元、访客单元、消息通知单元和门禁解除单元的智慧园区门禁系统,通过多次拍摄面部图像进行特征匹配和真实性检测,并根据访问频率和等级管理访客权限。

Benefits of technology

It improves the security of the access control system, allows frequent visitors to directly enter the park through facial recognition, reduces the number of manual confirmations, and regulates visitor behavior through access levels and regulations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118942178A_ABST
    Figure CN118942178A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of access control, and discloses a smart park access control system based on a face recognition technology, which comprises a face image recording unit, a data processing unit, a face information storage unit, a comparison matching unit, a detection unit, a visitor unit and the like. The face image recording unit is used for recording the face image of the person; and the data processing unit is used for processing the face image of the person recorded by the face image recording unit so as to obtain face feature data of the face. After the face recognition of the person is passed, a graphic mark pops up at a random position on a screen of the access control system, at the moment, the attention of the person under the curiosity psychology can be transferred to the popped-up graphic mark, meanwhile, eyeballs can pay attention to the popped-up graphic mark, so that the eyeballs are deflected, at the moment, the face graph of the person can be shot again, and the face of the person can be identified. And the initially shot face image is compared with the re-shot face image, so that the authenticity of the person can be judged.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of access control technology, and more specifically discloses a smart park access control system based on face recognition technology. Background Art

[0002] Park refers to a standard building or group of buildings that are generally planned and constructed by the government, with complete water supply, power supply, gas supply, communication, roads, storage and other supporting facilities, reasonable layout and can meet the needs of production and scientific experiments in a specific industry, including industrial parks, industrial parks, logistics parks, urban industrial parks, science and technology parks, creative parks, etc. In order to standardize the management of the park, an access control system is usually installed at the entrance and exit of the park. The commonly used access control system is an access control system using face recognition technology. By entering the face information of the park personnel in advance, when entering the park later, the face information of the personnel is recorded and compared with the stored face information. After the comparison is passed, the access control can be released to enter the park. However, the current access control system based on face recognition uses 2D face recognition technology, and the access control can be released by the facial photo of the person whose face information has been recorded, which is not very safe. In addition, the current access control system is not friendly to visitors who frequently visit the park. Each time you enter the park, you need to be manually confirmed before you can enter the park. Summary of the invention

[0003] The present invention mainly provides a smart park access control system based on face recognition technology, which can solve the problem that the current access control system is not very secure and is not friendly to frequent visitors.

[0004] To solve the above technical problems, according to one aspect of the present invention, more specifically, a smart park access control system based on face recognition technology includes: a face image recording unit, a data processing unit, a face information storage unit, a comparison and matching unit, a detection unit, a visitor unit, a message notification unit and an access control release unit; wherein:

[0005] Face image recording unit: used to record the face image of a person;

[0006] Data processing unit: used for processing the facial image of the person photographed by the facial image recording unit, so as to obtain facial feature data of the person;

[0007] Face information storage unit: used to store the face feature data obtained by the data processing unit;

[0008] Comparison and matching unit: used to compare and match the feature data of the subsequently recorded facial image of the person with the facial feature data in the facial information storage unit. If the facial feature data matches the facial feature data in the facial information storage unit, a successful match result is returned; otherwise, a failed match result is returned;

[0009] Detection unit: used to record the facial image of the person again after the comparison and matching unit returns a successful matching result, and compare the facial image recorded initially with the facial image recorded again, so as to detect the authenticity of the person;

[0010] Visitor unit: used to define a person as a visitor and record visitor information after the comparison and matching unit returns a match failure result. At the same time, it determines whether the visitor is a frequent visitor based on the recorded visitor information. If so, the visitor's facial information is temporarily stored;

[0011] Message notification unit: used to notify security personnel in the form of push messages after the visitor information is recorded in the visitor unit;

[0012] Access release unit: used to release the access control after the detection unit passes the detection, and the visitor information is confirmed by the security personnel and the access control is manually released by the security personnel.

[0013] Furthermore, the data processing unit is composed of an image data receiving module, an image data processing module and a feature analysis and extraction module;

[0014] Image data receiving module: used for receiving facial image information data captured by the facial image capturing unit;

[0015] Image data processing module: used for graying the facial image information data received by the image data receiving module;

[0016] Feature analysis and extraction module: used to extract facial feature data from the facial image processed by the image data processing module.

[0017] Furthermore, the comparison and matching unit includes: a data retrieval module, a comparison and analysis module, and a result determination module;

[0018] Data retrieval module: used to retrieve stored facial feature data from the face information storage unit;

[0019] Comparison and analysis module: used to compare and analyze the feature data of the facial image of the person recorded subsequently with the facial feature data retrieved by the data retrieval module, so as to analyze whether the feature data of the facial image of the person recorded subsequently can match the facial feature data stored in the face information storage unit;

[0020] Result determination module: used to return a successful match result if the feature data of the facial image of the person subsequently recorded in the comparison and analysis module can match the facial feature data stored in the face information storage unit, otherwise, return a failed match result.

[0021] Furthermore, the detection unit includes: a logo pop-up module, a recording and calling module, and an analysis and judgment module;

[0022] Logo pop-up module: used to pop up graphic logos at random locations on the access control display screen;

[0023] The recording calling module is used to call the facial image recording unit to record the facial image of the person again 1 second after the logo pop-up module pops up the graphic logo at a random position on the access control display screen;

[0024] Analysis and judgment module: used to compare the facial image of the person recorded initially with the facial image of the person recorded again, so as to judge the authenticity of the person.

[0025] Furthermore, in the process of determining the authenticity of a person, the analysis and judgment respectively uses the center points of the initially recorded facial image of the person and the re-recorded facial image of the person as the origin and establishes a two-dimensional coordinate system with a single pixel as the unit, and then determines the coordinates of each pixel of the iris of the eye according to the RGB value of the iris of the eye, if:

[0026] ;

[0027] If the person is true, then he is judged as a real person, otherwise he is judged as a fake person; , are the x and y values ​​of each pixel point p in the iris of the eye of the re-recorded facial image of the person, , are the x and y values ​​of each pixel point p in the iris of the person’s face image recorded initially, S is the total number of iris pixels, , They are the preset threshold parameters in the x-axis and y-axis directions respectively.

[0028] Furthermore, the visitor unit includes: a visitor information recording module, a visit interval recording module, a visit level assigning module and a face information temporary storage module;

[0029] Visitor information recording module: used to record visitor information, including but not limited to name and contact information;

[0030] Visit interval recording module: used to determine the time interval information of the same visitor's visit based on the visitor information recorded by the visitor information recording module;

[0031] Access level assignment module: used to automatically assign a visitor access level after approving visitor access. The default access level is 1.

[0032] Facial information temporary storage module: It is used to determine whether the same visitor has visited the park multiple times within a period of time based on the visitor information recorded by the visitor information recording module, and temporarily store the visitor's facial information for a period of time so that the visitor can enter the park directly through facial recognition during subsequent visits.

[0033] Furthermore, after the access level assigning module assigns a default access level to a visitor, when the visitor performs an operation that violates the park regulations in the park and is identified, the corresponding access level is deducted according to the severity of the violation.

[0034] Furthermore, the facial information temporary storage module temporarily stores the visitor's facial information for a period of time:

[0035] ;

[0036] in, The duration (min) for temporarily storing visitor facial information. is the time interval between each visit of the same visitor i (days), is the visit level of the same visitor each time he visits i, and Z is the total number of visits by the same visitor.

[0037] The beneficial effects of the smart park access control system based on face recognition technology of the present invention are:

[0038] (1) When performing face recognition on a person, when the face recognition is passed, the face image of the person will be photographed again, and then the initial facial image will be compared with the facial image photographed again, so as to determine the authenticity of the person. When it is determined that the person is a real person, the access control will be released. Otherwise, even if the face recognition is passed, the access control will not be released. Compared with the current access control system, the access control system disclosed in the present invention can greatly improve the security of the access control system.

[0039] (2) For frequent visitors, their facial images will be temporarily stored for a period of time. When the visitor visits the park again within the facial image storage time, the facial recognition will be directly passed, so that the frequent visitors can enter the park directly after verification, which makes it convenient for frequent visitors to enter the park without having to manually confirm the door opening every time they visit. It is more friendly to frequent visitors.

[0040] (3) Since the access level is linked to the park regulations, the behavior of visitors can be regulated. When they violate the park regulations seriously and their access level is deducted a lot, it will directly affect the temporary storage time of their facial information, and even directly make the temporary storage time of their facial information zero, thus not allowing them to enter the park directly through facial recognition in the future. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0042] Figure 1 This is a schematic diagram of the system principle. DETAILED DESCRIPTION

[0043] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0044] According to one aspect of the present invention, a smart park access control system based on face recognition technology is provided. The system first uses a high-definition camera through a face image recording unit to capture facial images of staff members in the park, and then sends the captured facial images to a data processing unit for processing. The data processing unit receives the facial images of the personnel through an image data receiving module.

[0045] Then, the image data processing module performs grayscale and binarization processing on the received facial image, and then the feature analysis and extraction module extracts the facial feature data of the person, such as face width, face length, eyebrow distance, eye distance, etc. The extracted facial feature data is stored in the face information storage unit.

[0046] In the subsequent use process, the facial image recording unit uses a high-definition camera to capture the facial image of the person who wants to enter the park, and the captured facial image is handed over to the data processing unit for processing. The data processing unit receives the facial image of the person through the image data receiving module, and then performs grayscale, binarization and other processing on the received facial image through the image data processing module.

[0047] The facial feature data of the person is then extracted through the feature analysis and extraction module, and then the facial feature data is sent to the comparison and matching unit. At the same time, the comparison and matching unit will retrieve the stored facial feature data from the face information storage unit through the data retrieval module, and then the facial feature data of the person who wants to enter the park will be compared and analyzed with the facial feature data stored in the retrieved face information storage unit through the comparison and analysis module, so as to analyze whether the feature data of the facial image of the person recorded subsequently can match the facial feature data stored in the face information storage unit. If they can match, the result determination module returns a successful match result, otherwise, the match failure result is returned.

[0048] When the result determination module returns a successful match result, the logo pop-up module in the detection unit pops up a graphic logo, such as a colored arrow, circle, etc., at a random position on the display screen of the access control system. One second after the graphic logo pops up, the facial image recording unit is called by the recording calling module to record the facial image of the person again, and the initial recorded facial image and the re-recorded facial image are sent to the analysis and judgment module. The analysis and judgment module uses the center points of the initial recorded facial image and the re-recorded facial image as the origin and establishes a two-dimensional coordinate system in units of a single pixel, and then determines the coordinates of each pixel of the eye iris according to the RGB value of the eye iris. If:

[0049] ;

[0050] If the person is judged to be a real person, the access control release unit will release the access control and allow the person to enter the park. Otherwise, it is judged to be a fake person. Even if the face recognition passes, the access control release unit will still not release the access control. , are the x and y values ​​of each pixel point p in the iris of the eye of the re-recorded facial image of the person, , are the x and y values ​​of each pixel point p in the iris of the person’s face image recorded initially, S is the total number of iris pixels, , They are the preset threshold parameters in the x-axis and y-axis directions respectively.

[0051] When the result determination module returns a match failure result, the visitor unit will define the person as a visitor. At the beginning, the security personnel will be notified through the message notification unit for confirmation. After confirmation, the access control will be manually opened. At the same time, the visitor's name, contact information and other information will be recorded through the visitor information recording module. The visitor information recorded by the visitor information recording module is used to determine whether the visitor frequently visits for a period of time, for example, every 1-3 days. If the visitor is determined to be a frequent visitor, the visit interval recording module will record the time interval information of the visitor's visit, and the access level assignment module will assign the default access level of 1 to the visitor each time it visits.

[0052] When a visitor enters the park, the camera in the park is used with a recognition algorithm trained by a neural network to detect whether the visitor has violated any rules. If a violation occurs, the corresponding access level will be deducted. For example, if a visitor is detected to have violated the rules by damaging the park facilities, the visitor's access level will be deducted by 0.3, leaving a remaining access level of 0.7. After the visitor ends the visit, the access level will be recorded. Then, the face information temporary storage module temporarily stores the visitor's face information for a period of time:

[0053] ;

[0054] in, The duration (min) for temporarily storing visitor facial information. is the time interval between each visit of the same visitor i (days), is the access level of the same visitor each time he visits i, and Z is the total number of visits by the same visitor. In the subsequent visits, you can directly enter the park through face recognition, without having to manually confirm the access control before entering the park every time you visit. It is more friendly to visitors who frequently visit the park.

[0055] In addition, when the access level of the visitor is low, the temporary storage time of his facial information is correspondingly short, or even directly 0. At this time, after the shorter facial information storage time, he is not allowed to enter the park directly through facial recognition or he is not allowed to enter the park directly through facial recognition next time.

[0056] Specifically, in this embodiment, the system is set to determine that the person who wants to enter the park is a real person when the pixel points on the iris of the eye are offset by 1 pixel unit (i.e. 1px) in the horizontal or vertical direction. When person A wants to enter the park, the facial image of person A is recorded by a high-definition camera and compared with the facial image of the person in the system database. After the comparison, taking the center point of the right eye iris as an example, the coordinates of the center point of the eyeball in the initial facial image of the person are (20px, 25px). Then a graphic logo pops up in the upper left corner of the display screen of the access control system, and the facial image of the person is recorded again after 1s, and the coordinates of the center point of the eye iris are obtained as (10px, 30px). At this time: |10px-20px|>1px or |30px-25px|>1px, it is determined that person A is a real person, the system releases the access control through control, and allows person A to pass through and enter the park.

[0057] In this embodiment, after the facial image matching of person B fails, the access record of person B is retrieved from the data storage area of ​​the system, as shown in the following table:

[0058] Table 1: Visit record of visitor B

[0059] From Table 1, we can see that from each access record of visitor B, we can know that person B visited the park multiple times within 8 days, so person B is judged as a frequent visitor. At this time, from each access record, the access time interval of person B (since the access record on May 2, 2024 is the first visit, the access time interval is recorded as 0), the access level of each visit to the park, and the total number of visits are extracted, so that the facial image of the person can be stored in the system. min, when the person visits the park within this time, he can enter the park for access after directly passing the same verification as in the first embodiment.

[0060] The electrical components that appear in this article are all electrical components that exist in reality.

[0061] The smart campus access control system based on face recognition technology of the present invention includes the following beneficial effects.

[0062] (1) When performing facial recognition on a person, if the person's facial recognition is passed, a graphic logo will pop up at a random position on the screen of the access control system. At this time, the person's attention will be shifted to the pop-up graphic logo out of curiosity, and the eyeballs will be focused on the pop-up graphic logo, causing the eyeballs to deflect. At this time, the person's facial image will be photographed again, and then the initial facial image will be compared with the facial image photographed again, so that the authenticity of the person can be determined. If it is determined to be a real person, the access control will be released. Otherwise, even if the facial recognition is passed, the access control will not be released. Compared with the current access control system, the access control system disclosed in the present invention can greatly improve the security of the access control system.

[0063] (2) For frequent visitors, the facial image storage time of the visitor can be obtained based on the visit interval and visit level of each visit, and then the facial image is temporarily stored for a period of time. When the visitor visits the park again within the facial image storage time, the facial recognition is directly passed, so that the frequent visitors can enter the park directly after verification, which makes it convenient for frequent visitors to enter the park without having to manually determine to open the access control every time they visit. It is more friendly to frequent visitors.

[0064] (3) Since the access level is linked to the park regulations, the behavior of visitors can be regulated. When they violate the park regulations seriously and their access level is deducted a lot, it will directly affect the temporary storage time of their facial information, and even directly make the temporary storage time of their facial information zero, thus not allowing them to enter the park directly through facial recognition in the future.

[0065] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention also fall within the protection scope of the present invention.

Claims

1. A smart park access control system based on face recognition technology. It is characterized in that include: A face image recording unit, a data processing unit, a face information storage unit, a comparison and matching unit, a detection unit, a visitor unit, a message notification unit, and a door access release unit; wherein: Face image recording unit: used to record the face image of a person; Data processing unit: used for processing the facial image of the person photographed by the facial image recording unit, so as to obtain facial feature data of the person; Face information storage unit: used to store the face feature data obtained by the data processing unit; Comparison and matching unit: used to compare and match the feature data of the subsequently recorded facial image of the person with the facial feature data in the facial information storage unit. If the facial feature data matches the facial feature data in the facial information storage unit, a successful match result is returned; otherwise, a failed match result is returned; Detection unit: used to record the facial image of the person again after the comparison and matching unit returns a successful matching result, and compare the facial image recorded initially with the facial image recorded again, so as to detect the authenticity of the person; Visitor unit: used to define a person as a visitor and record visitor information after the comparison and matching unit returns a match failure result. At the same time, it determines whether the visitor is a frequent visitor based on the recorded visitor information. If so, the visitor's facial information is temporarily stored so that the visitor can directly access the park after passing facial authentication during subsequent visits; Message notification unit: used to notify security personnel in the form of push messages after the visitor information is recorded in the visitor unit; Access release unit: used to release the access control after the detection unit passes the detection, and the visitor information is confirmed by the security personnel and the access control is manually released by the security personnel.

2. According to claim 1, the smart park access control system based on face recognition technology, Features: The data processing unit is composed of an image data receiving module, an image data processing module and a feature analysis and extraction module; Image data receiving module: used for receiving facial image information data captured by the facial image capturing unit; Image data processing module: used for graying the facial image information data received by the image data receiving module; Feature analysis and extraction module: used to extract facial feature data from the facial image processed by the image data processing module.

3. According to claim 1, the smart park access control system based on face recognition technology, Features: The comparison and matching unit includes: a data retrieval module, a comparison and analysis module and a result determination module; Data retrieval module: used to retrieve stored facial feature data from the face information storage unit; Comparison and analysis module: used to compare and analyze the feature data of the facial image of the person recorded subsequently with the facial feature data retrieved by the data retrieval module, so as to analyze whether the feature data of the facial image of the person recorded subsequently can match the facial feature data stored in the face information storage unit; Result determination module: used to return a successful match result if the feature data of the facial image of the person subsequently recorded in the comparison and analysis module can match the facial feature data stored in the face information storage unit, otherwise, return a failed match result.

4. According to claim 1, the smart park access control system based on face recognition technology, Features: The detection unit includes: a logo pop-up module, a recording and calling module, and an analysis and judgment module; Logo pop-up module: used to pop up graphic logos at random locations on the access control display screen; The recording calling module is used to call the facial image recording unit to record the facial image of the person again 1 second after the logo pop-up module pops up the graphic logo at a random position on the access control display screen; Analysis and judgment module: used to compare the facial image of the person recorded initially with the facial image of the person recorded again, so as to judge the authenticity of the person.

5. According to claim 4, the smart park access control system based on face recognition technology, Features: In the process of judging the authenticity of a person, the analysis and judgment respectively takes the center point of the initially recorded facial image of the person and the re-recorded facial image of the person as the origin and establishes a two-dimensional coordinate system with a single pixel as the unit, and then determines the coordinates of each pixel point of the iris of the eye according to the RGB value of the iris of the eye, if: ; If the person is true, then he is judged as a real person, otherwise he is judged as a fake person; , are the x and y values ​​of each pixel point p in the iris of the eye of the re-recorded facial image of the person, , are the x and y values ​​of each pixel point p in the iris of the person’s face image recorded initially, S is the total number of iris pixels, , They are the preset threshold parameters in the x-axis and y-axis directions respectively.

6. The smart park access control system based on face recognition technology according to claim 1, Features: The visitor unit includes: a visitor information recording module, a visit interval recording module, a visit level assigning module and a face information temporary storage module; Visitor information recording module: used to record visitor information, including name and contact information; Visit interval recording module: used to determine the time interval information of the same visitor's visit based on the visitor information recorded by the visitor information recording module; Access level assignment module: used to automatically assign a visitor access level after approving visitor access. The default access level is 1. Facial information temporary storage module: It is used to determine whether the same visitor has visited the park multiple times within a period of time based on the visitor information recorded by the visitor information recording module, and temporarily store the visitor's facial information for a period of time so that the visitor can enter the park directly through facial recognition during subsequent visits.

7. According to claim 6, the smart park access control system based on face recognition technology, Features: After the access level assigning module assigns a default access level to a visitor, if the visitor performs an operation that violates the park regulations and is identified in the park, the corresponding access level will be deducted according to the severity of the violation.

8. According to claim 7, the smart park access control system based on face recognition technology, Features: The face information temporary storage module temporarily stores the visitor's face information for a period of time: ; in, The duration for which visitor facial information is temporarily stored. is the time interval between each visit of the same visitor i, is the visit level of the same visitor each time he visits i, and Z is the total number of visits by the same visitor.

Citation Information

Patent Citations

  • Living body human face recognition method and device

    CN105184277A

  • Method for controlling intelligent door lock and terminal

    CN106296894A

  • Reservation access method and system, computer device and storage medium

    CN109447297A

  • Entrance guard management system and management method thereof

    CN109636952A

  • Face recognition-based school safety management early warning method for K12 education stage

    CN111145414A