CN109544754B - Access control method and system based on face recognition and computer equipment - Google Patents
Access control method and system based on face recognition and computer equipment Download PDFInfo
- Publication number
- CN109544754B CN109544754B CN201811422542.2A CN201811422542A CN109544754B CN 109544754 B CN109544754 B CN 109544754B CN 201811422542 A CN201811422542 A CN 201811422542A CN 109544754 B CN109544754 B CN 109544754B
- Authority
- CN
- China
- Prior art keywords
- personnel
- advertisement
- current
- face
- stay time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/00174—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys
- G07C9/00563—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys using personal physical data of the operator, e.g. finger prints, retinal images, voicepatterns
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an entrance guard method and system based on face recognition and computer equipment, wherein the method comprises the following steps: when a person is in front of a camera of the access control, acquiring a person image; after the personnel image is obtained, analyzing personnel characteristics in the personnel image, and matching advertisements according to the personnel characteristics; extracting face feature data in the personnel image, and performing face recognition unlocking; and when the stay time of the personnel is greater than a threshold value, reporting the current personnel, the stay time of the personnel, the advertisement playing information and the personnel characteristics to a server. The invention has the advantages that the high-precision face recognition is carried out on the front-end equipment, and the real-time performance is strong; the face recognition module and the advertisement matching module are executed in parallel, so that the timeliness of accurate advertisement pushing is improved, and the effective playing time of the advertisements for personnel can be prolonged. The invention can accelerate the popularization and the use of the face recognition access control system in communities.
Description
Technical Field
The invention relates to the field of face recognition and intelligent access control, in particular to an access control method and system based on face recognition and computer equipment.
Background
In recent years, with the development of smart city construction, the demand of the community for smart access control shows explosive growth. The traditional entrance guard generally adopts the mode of punching the card to pass in and out residential districts, must hand-carry entrance guard card, brings inconvenience for the life, also causes the loss easily, if picked up by illegal personnel, also is very big hidden danger to the district safety.
With the rapid development of the AI technology, especially the wide application of the deep learning technology in the field of image recognition, the face recognition technology is promoted to mature continuously. The application of the face recognition technology in business scenes such as security, high-speed rail stations, airports, banks, payment, intelligent businesses, intelligent communities and the like is increased explosively, the current face recognition precision can exceed the level that human eyes can recognize, and the requirement of a plurality of business scenes on face recognition can be met by greatly improving the face recognition precision. The application of face identification technique community entrance guard is also more and more extensive, just can pass in and out the community through brushing the face, has solved inconvenience and the potential safety hazard problem that traditional access control card carried, has brought the convenience for the resident's life of community, has also strengthened the security of community simultaneously greatly, can effectually prevent the emergence of criminal incident, promotes the efficiency of chasing down criminal suspect by a wide margin, and the personnel of any business turn over community can both find at face identification system.
The existing face access control system mostly adopts a cloud platform to perform face recognition and advertisement calculation, and advertisement pushing is performed after the face recognition, so that the problems of long face recognition delay time, untimely advertisement pushing, long waiting time for opening a door of a person and the like exist, and the advertisement pushing module is arranged behind the face recognition module, so that the time for watching the advertisement by the person is short, and the person is difficult to attract to stand and watch the advertisement. In addition, most of the existing face access control advertisement pushing is based on gender and age characteristics obtained by face analysis, or based on characteristics of background personnel data analysts, and the problem that the pushed advertisements are not accurate enough exists.
Disclosure of Invention
The invention provides an access control method and system based on face recognition and computer equipment, which are used for solving the technical problems of long time delay of face recognition, untimely advertisement pushing and long waiting time for opening a door for personnel.
In order to solve the technical problems, the technical scheme provided by the invention is as follows:
a door control method based on face recognition comprises the following steps:
when a person is in front of a camera of the access control, acquiring a person image;
after the personnel image is obtained, analyzing personnel characteristics in the personnel image, and matching advertisements according to the personnel characteristics; extracting face feature data in the personnel image, and performing face recognition unlocking;
and when the stay time of the personnel is greater than a set threshold value, reporting the current personnel, the stay time of the personnel, the advertisement playing information and the personnel characteristics to a server.
As a further improvement of the process of the invention:
preferably, the personnel features include: age, gender, body type, facial classification of five sense organs, dressing, and associated articles; the advertisement matching according to the personnel characteristics is classified and matched by utilizing a trained personnel characteristic classification model, and the personnel characteristic classification model adopts the following steps: deep learning, reinforcement learning, generation of confrontation network and pattern recognition method, or any combination thereof, and designing and training.
Preferably, after extracting the face feature data in the personnel image, comparing the current face feature data with the face feature of the latest passing object in a ratio of 1: 1;
if the current face feature data is the same as the latest passing object, judging that the current personnel is not changed, and accumulating the personnel stay time of the current personnel; the latest passing object is a person who passes the latest face recognition verification;
and if the current face feature data is different from the latest passing object, performing 1: N face comparison on the current face feature data and an authorized white list library, identifying the current face feature data as the latest passing object when the comparison is successful, recalculating the stay time of the personnel, reporting the identification record to a server, and opening the access control.
Preferably, when the stay time of the people is longer than a set threshold value, the currently played advertisement information is set as the advertisement which is interested by the current people, and the advertisements or the advertisements in the same category as the advertisements are pushed to the people when the next people pass through the entrance guard or use the mobile phone APP.
Preferably, the face feature data in the extracted personnel image is extracted by adopting one or a combination of any more of deep learning, reinforcement learning, generation of an antagonistic network and pattern recognition methods, the current face feature data is compared with an authorized white list library by a 1: N face, and the retrieval comparison is carried out by adopting one or a combination of more of linear query comparison, Hash-based indexing methods and support point measurement spatial indexing methods.
The invention also provides an access control system based on face recognition, which comprises:
the face detection module is used for acquiring a person image when a person is in front of a camera of the access control;
the personnel feature analysis module is used for analyzing personnel features based on the personnel images after the personnel images are obtained;
the face recognition module is used for extracting face features in the personnel image and carrying out face recognition unlocking;
the advertisement matching unit is used for matching advertisements according to the personnel characteristics and playing the advertisements matched with the personnel characteristics on a screen;
the stay time calculation unit is used for calculating stay time of the personnel;
the reporting unit is used for acquiring the current stay time of the personnel and the current stay time of the personnel from the stay time calculating unit when the stay time of the personnel is greater than a threshold value; and the broadcast advertisement information and the personnel characteristics acquired from the advertisement matching unit are reported to the server.
As a further improvement of the access control system of the invention:
preferably, the system further comprises an intelligent access control unit, wherein the intelligent access control unit is used for comparing the current face feature data with the face feature of the latest passing object in a ratio of 1:1 after extracting the face feature data in the personnel image;
if the current face feature data is the same as the latest passing object, judging that the current personnel is not changed, and informing a stay time calculation unit to accumulate the stay time of the current personnel; the latest passing object is a person who passes the latest face recognition verification;
and if the current face characteristic data is different from the latest passing object, performing 1: N face comparison on the current face characteristic data and an authorized white list library, identifying the current face characteristic data as the latest passing object when the comparison is successful, informing the stay time calculation unit to recalculate the stay time of the personnel, reporting the identification record to a server through a reporting unit, and opening the access control. And when the identification fails, the identification record is also reported to the server for background inquiry.
Preferably, the person characteristics include, but are not limited to, age, gender, size, facial classification of five sense organs, dress, and associated items; the advertisement matching unit carries out advertisement matching according to personnel characteristics, and carries out classification matching by utilizing a trained personnel characteristic classification model, wherein the personnel characteristic classification model adopts the following steps: deep learning, reinforcement learning, generation of confrontation network and pattern recognition method, or any combination thereof, and designing and training.
Preferably, when the stay time of the people is longer than a set threshold, the reporting unit sets the currently played advertisement information as the advertisement which is interested by the current people and reports the advertisement to the server; the server is used for pushing advertisements or advertisements of the same category with the advertisements to the personnel when the personnel pass through the entrance guard or use the mobile phone APP next time.
The present invention also provides a computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the methods described above when executing the computer program.
The invention has the following beneficial effects:
the access control method and the access control system based on the face recognition have the advantages that: high-precision face recognition is carried out on front-end equipment, and the real-time performance is strong; personnel feature analysis is carried out on front-end equipment on the basis of personnel images by using a machine learning method, so that the granularity of feature extraction can be improved, more dimensional features can be extracted, and the advertisement matching precision is improved; the face recognition and advertisement matching module is executed in parallel, so that the timeliness of accurate advertisement pushing is improved, and the effective playing time of the advertisement to a person can be prolonged; the system supports the operation in the equipment based on the ARM processor, and the operation power consumption of the equipment can be greatly reduced. Considerable benefits are brought to the operation of the equipment, and the popularization and the use of the face recognition access control system in the community can be accelerated.
In addition to the objects, features and advantages described above, other objects, features and advantages of the present invention are also provided. The present invention will be described in further detail below with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this application, illustrate embodiments of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart of an access control method based on face recognition according to a preferred embodiment of the present invention;
fig. 2 is a schematic flow chart of a face recognition-based access control method according to preferred embodiment 2 of the present invention.
Detailed Description
The embodiments of the invention will be described in detail below with reference to the drawings, but the invention can be implemented in many different ways as defined and covered by the claims.
Referring to fig. 1, the access control method based on face recognition of the present invention includes the following steps:
when a person is in front of a camera of the access control, acquiring a person image;
after the personnel image is obtained, analyzing personnel characteristics in the personnel image, and matching advertisements according to the personnel characteristics; extracting face feature data in the personnel image, and performing face recognition unlocking;
and broadcasting the advertisement matched with the personnel characteristics on a screen, updating the stay time of the personnel, and reporting the current personnel, the stay time of the personnel, the advertisement broadcasting information and the personnel characteristics to a server when the stay time of the personnel is greater than a set threshold (in the embodiment, the threshold range is 5-10s and can be configured as required, and the larger the threshold is, the more interested the user with the stay time exceeding the threshold is in the current advertisement, but the acquired users are reduced).
Through the steps, the installation and operation cost of the face access control equipment can be greatly reduced, considerable benefits can be brought to the operation of the equipment through real-time accurate advertisement pushing, and popularization and use of the face recognition access control system in the community can be accelerated.
In practice, the above method can be expanded or applied as follows, all the technical features in the following embodiments can be combined with each other, and the embodiments are only used as examples and are not limited to the normal combination of the technical features.
Example 1:
referring to fig. 1, the access control method based on face recognition of the embodiment includes the following steps:
when a person is in front of a camera of the access control, acquiring a person image;
after obtaining personnel's image, personnel's characteristics in the analysis personnel's image, personnel's characteristics include: classifying age, gender, body type, facial type of five sense organs, dressing and related articles, and carrying out advertisement matching according to the characteristics of people; the personnel features are classified by using trained personnel feature classification models, the models can be designed and trained by adopting machine learning methods including but not limited to deep learning, reinforcement learning, generation of confrontation networks, traditional pattern recognition methods and the like, and the models can also be combined by machine learning models of a plurality of different methods to carry out personnel feature analysis. Personnel feature analysis is carried out aiming at personnel images, and personnel features in the images can be analyzed in real time and used for carrying out advertisement matching. For example, the current person is analyzed to be a young person wearing sports wear through an algorithm model, and then gym advertisements can be pushed; if the person is young women with high face value, wearing short skirt, carrying small bag, it can push advertisement such as cosmetic, weight-reducing medicine, LV bag or travel; if the man is a middle-aged man wearing formal clothes, tying a tie and pulling a leather case, the man can push advertisements such as house property, luxury cars, hotels, clubs and the like. The advertisement is matched with the advertisement to be pushed better and accurately through combination of multiple characteristics such as age, gender, dressing, facial type five sense organs classification (one or combination of any more of characteristics such as fat, thin, round or long facial type, bright or dark skin color, rough or smooth skin, eye eyebrow shape, nose length and chin shape), body type and related objects, and the accuracy of advertisement pushing is improved.
And simultaneously extracting the face characteristic data in the personnel image to carry out face recognition unlocking. The method comprises the steps of comparing the current face feature data with the latest passing object by the face feature ratio of 1: 1;
if the current face feature data is the same as the latest passing object, judging that the current personnel is not changed, and accumulating the stay time of the personnel; the latest passing object is a person who passes the latest face recognition verification;
and if the current face feature data is different from the latest passing object, performing 1: N face comparison on the current face feature data and an authorized white list library, identifying the current face feature data as the latest passing object when the comparison is successful, recalculating the stay time of the personnel, reporting the identification record to a server, and opening the access control.
And simultaneously carrying out advertisement matching: and when the stay time of the personnel is greater than a threshold value, reporting the current personnel, the stay time of the personnel, the advertisement playing information and the personnel characteristics to a server.
When the stay time of the person is longer than the threshold value, the person can be determined to be interested in playing the advertisement, the stay time is in direct proportion to the interest degree, the current playing advertisement information is further set as the advertisement which is interested by the current person, and when the next person passes through the entrance guard or uses the mobile phone APP, the advertisement or the advertisement of the same category as the advertisement is pushed to the person.
Extracting the face feature data in the personnel image by adopting a neural network algorithm model based on deep learning; and performing 1: N face comparison on the current face feature data and an authorized white list library, and performing retrieval comparison by adopting one or a combination of a linear query comparison method, a Hash-based indexing method and a support point measurement spatial indexing-based method.
This embodiment still provides an access control system based on face identification, includes:
the face detection module is positioned in the front-end camera and used for acquiring a person image when a person is in front of the camera of the entrance guard;
the personnel feature analysis module is positioned in the front-end entrance guard machine and used for analyzing personnel features in the personnel images after the personnel images are acquired; person characteristics include, but are not limited to, age, gender, size, facial classification of five sense organs, dress, and associated items; the advertisement matching unit carries out advertisement matching according to personnel characteristics, and carries out classification matching by utilizing a trained personnel characteristic classification model, wherein the personnel characteristic classification model adopts the following steps: deep learning, reinforcement learning, generation of confrontation network and pattern recognition method, or any combination thereof, and designing and training.
The face recognition module is positioned in the front-end access control machine, and is used for extracting face characteristic data in the personnel image and carrying out face recognition unlocking;
the advertisement matching unit is used for matching advertisements according to the personnel characteristics and playing the advertisements matched with the personnel characteristics on a screen;
the stay time calculation unit is used for calculating stay time of the personnel;
the reporting unit is used for acquiring the current stay time of the personnel and the current stay time of the personnel from the stay time calculating unit when the stay time of the personnel is greater than a threshold value; and the broadcast advertisement information and the personnel characteristics acquired from the advertisement matching unit are reported to the server; the system is also used for setting the currently played advertisement information as the advertisement which is interested by the current personnel and reporting the advertisement to the server when the stay time of the personnel is longer than the set threshold; the server is used for pushing advertisements or advertisements of the same category with the advertisements to the personnel when the personnel pass through the entrance guard or use the mobile phone APP next time.
The intelligent access control unit is used for comparing the current human face feature data with the latest passing object by the human face feature ratio of 1:1 after extracting the human face feature data in the personnel image; if the current face feature data is the same as the latest passing object, judging that the current personnel is not changed, and informing a stay time calculation unit to accumulate the stay time of the current personnel; the latest passing object is a person who passes the latest face recognition verification; and if the current face characteristic data is different from the latest passing object, performing 1: N face comparison on the current face characteristic data and an authorized white list library, identifying the current face characteristic data as the latest passing object when the comparison is successful, informing the stay time calculation unit to recalculate the stay time of the personnel, reporting the identification record to a server through a reporting unit, and opening the access control.
The face detection module is positioned in the front-end camera, and the personnel feature analysis module, the face recognition module, the advertisement matching unit, the stay time calculation unit and the reporting unit are all positioned in the front-end access control equipment.
Example 2:
the embodiment is the application example of embodiment 1, the hardware of this embodiment adopts the ARM board of RK3399, Cortex-a72+ Cortex-a53 six-core 64-bit 2.0GHz processor, the performance is strong, the operating system is ubuntu, adopt RK3399 to ensure that the application obtains powerful performance support, reduce the cost that the product was used, the low-power consumption of ARM can reduce the power consumption that access control system long-time operation brought, the operation cost is reduced, the use threshold of product has also been reduced simultaneously, be convenient for the quick popularization and application of product.
Referring to fig. 2, the access control method based on face recognition of the embodiment specifically includes the following steps:
1) the face detection module is used for detecting the face at the front end of the camera, and if the face is detected, the follow-up steps are executed;
2) extracting human face features, namely acquiring a human face image from the step 1), and extracting human face feature data by using a deep neural network-based human face recognition algorithm model;
3) judging whether the current person is the latest passing object, comparing the current face features with the latest passing object in a ratio of 1:1 to judge whether the current person changes, and if not, executing the step 10); if the change occurs, executing step 4);
4) comparing the face, namely comparing the face features captured by the current camera with the face features captured by the authorized white list library in a ratio of 1: N if the face is in the white list library, and executing the step 5) if the face features are in the white list library; if the face is not in the white list, the face recognition process is ended;
5) and identifying the latest passing object, and reporting the record to a server. After the comparison and verification in the step 4) are completed, reporting the identified record information to a management center server;
6) opening the electromagnetic lock, and after the step 5) is finished, opening the door control and releasing the personnel;
7) analyzing the person image (age, sex, body type, dressing and the like), acquiring the shot person picture from the step 1), and performing person image characteristic analysis by using a characteristic analysis algorithm;
8) the advertisement module is used for accurately matching advertisements, and the advertisements are accurately matched based on personnel characteristics obtained by personnel portrait analysis in the step 7);
9) playing the advertisement, and putting the matched advertisement in the step 8) through a screen;
10) calculating the stay time, matching the current personnel characteristics obtained in the step 3) with the latest passing object characteristics, and updating the stay time of the personnel;
11) dwell time > threshold, a threshold is set for determining whether a person is interested in placing an advertisement, and if dwell time exceeds the set threshold, step 12) is performed;
12) and reporting the current personnel, the stay time, the advertisement playing information and the personnel characteristic analysis data to a server.
Example 3:
the present embodiment provides a computer system, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above embodiments when executing the computer program.
In conclusion, the ARM-based front-end equipment can greatly reduce the equipment installation and operation cost, and considerable operation income is brought by real-time accurate advertisement pushing through personnel portrait analysis. The access control device based on ARM face recognition has the advantages that in ARM operation, hardware purchasing cost is low, operation power consumption is low, installation and operation cost can be remarkably reduced, and popularization of products is facilitated. Personnel portrait analysis utilizes machine learning algorithm to carry out the analysis to the personnel's photo that the camera was shot to obtain characteristic information such as personnel's age, sex, dress, accurate analysis personnel affiliated crowd, characteristic, interest preference utilize the advertisement module to carry out real-time accurate advertisement propelling movement, thereby bring observable income for equipment operation. Accurate propelling movement of advertisement and face identification parallel execution have promoted the real-time of advertisement propelling movement, can increase the advertisement and to long to personnel's broadcast, promote advertisement propelling movement effect. Accurate propelling movement of advertisement and face identification parallel execution have promoted the real-time of advertisement propelling movement, can increase the advertisement and to long to personnel's broadcast, promote advertisement propelling movement effect.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. A door control method based on face recognition is characterized by comprising the following steps:
when a person is in front of a camera of the access control, acquiring a person image;
after a personnel image is acquired, personnel characteristics in the personnel image are analyzed, and advertisement matching is carried out according to the personnel characteristics; and extracting the face characteristic data in the personnel image, and carrying out face recognition unlocking: the method comprises the steps of comparing the current face feature data with the latest passing object by the face feature ratio of 1: 1;
if the current face feature data is the same as the latest passing object, judging that the current personnel is not changed, and accumulating the personnel stay time of the current personnel; the latest passing object is a person who passes the latest face recognition verification;
if the current face feature data is different from the latest passing object, performing 1: N face comparison on the current face feature data and an authorized white list library, identifying the current face feature data as the latest passing object when the comparison is successful, recalculating the stay time of personnel, reporting the identification record to a server, and opening the access control;
broadcasting the advertisement matched with the personnel characteristics on a screen, updating the stay time of personnel, and reporting the current personnel, the stay time of personnel, the broadcasting advertisement information and the personnel characteristics to a server when the stay time of personnel is greater than a set threshold value; the range of the set threshold is 5-10 s.
2. The access control method based on the face recognition, according to claim 1, wherein the personnel features comprise: age, gender, body type, facial classification of five sense organs, dressing, and associated articles; and according to the personnel characteristics, carrying out advertisement matching by using a trained personnel characteristic classification model, wherein the personnel characteristic classification model adopts the following steps: deep learning, reinforcement learning, generation of confrontation network and pattern recognition method, or any combination thereof, and designing and training.
3. The access control method based on the face recognition is characterized in that when the stay time of the personnel is longer than a set threshold value, currently played advertisement information is set as an advertisement which is interested by the current personnel, and the advertisement or the advertisement of the same category as the advertisement is pushed to the personnel next time the personnel passes through the access control or uses a mobile phone APP.
4. The access control method based on the face recognition is characterized in that the face feature data in the extracted personnel image is extracted by adopting one or a combination of any more of deep learning, reinforcement learning, generation of an antagonistic network and a pattern recognition method, the current face feature data is compared with an authorized white list library by a 1: N face, and the retrieval comparison is carried out by adopting one or a combination of more of linear query comparison, a Hash-based index method and a support point measurement space index-based method.
5. The utility model provides an access control system based on face identification which characterized in that includes:
the face detection module is used for acquiring a person image when a person is in front of a camera of the access control;
the personnel feature analysis module is used for analyzing personnel features based on the personnel images after the personnel images are obtained;
the face recognition module is used for extracting face features in the personnel image and carrying out face recognition unlocking;
the advertisement matching unit is used for matching advertisements according to the personnel characteristics and playing the advertisements matched with the personnel characteristics on a screen;
the intelligent access control unit is used for comparing the current human face feature data with the latest passing object by the human face feature ratio of 1:1 after extracting the human face feature data in the personnel image;
if the current face feature data is the same as the latest passing object, judging that the current personnel is not changed, and informing a stay time calculation unit to accumulate the stay time of the current personnel; the latest passing object is a person who passes the latest face recognition verification;
if the current face feature data is different from the latest passing object, performing 1: N face comparison on the current face feature data and an authorized white list library, identifying the current face feature data as the latest passing object when the comparison is successful, informing a stay time calculation unit to recalculate the stay time of personnel, reporting the identification record to a server through a reporting unit, and opening the access control;
the stay time calculation unit is used for calculating stay time of the personnel;
the reporting unit is used for acquiring the current stay time of the personnel and the current stay time of the personnel from the stay time calculating unit when the stay time of the personnel is greater than a threshold value; and the broadcast advertisement information and the personnel characteristics acquired from the advertisement matching unit are reported to the server; the range of the set threshold is 5-10 s.
6. The face recognition based access control system of claim 5, wherein the person characteristics include, but are not limited to, age, gender, size, facial classification of five sense organs, dressing, and associated items; the advertisement matching unit performs advertisement matching according to the personnel characteristics by using a trained personnel characteristic classification model for classification matching, wherein the personnel characteristic classification model adopts the following steps: deep learning, reinforcement learning, generation of confrontation network and pattern recognition method, or any combination thereof, and designing and training.
7. The access control system based on the face recognition is characterized in that when the stay time of the personnel is longer than a set threshold value, the reporting unit sets the currently played advertisement information as an advertisement which is interested by the current personnel and reports the advertisement to the server;
the server is used for pushing the advertisements or the advertisements of the same category with the advertisements to the personnel next time when the personnel pass through an entrance guard or use a mobile phone APP.
8. A computer system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any one of claims 1 to 4 are performed when the computer program is executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811422542.2A CN109544754B (en) | 2018-11-27 | 2018-11-27 | Access control method and system based on face recognition and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811422542.2A CN109544754B (en) | 2018-11-27 | 2018-11-27 | Access control method and system based on face recognition and computer equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109544754A CN109544754A (en) | 2019-03-29 |
| CN109544754B true CN109544754B (en) | 2021-09-28 |
Family
ID=65850353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811422542.2A Active CN109544754B (en) | 2018-11-27 | 2018-11-27 | Access control method and system based on face recognition and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109544754B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110185333A (en) * | 2019-04-30 | 2019-08-30 | 南京邮电大学 | A kind of recognition of face intelligent lock |
| CN110276320A (en) * | 2019-06-26 | 2019-09-24 | 杭州创匠信息科技有限公司 | Guard method, device, equipment and storage medium based on recognition of face |
| CN110610379A (en) * | 2019-08-16 | 2019-12-24 | 北京网聘咨询有限公司 | Advertisement marketing system convenient to judge commodity colour |
| CN111932754B (en) * | 2019-08-19 | 2021-12-28 | 北京戴纳实验科技有限公司 | Laboratory entrance guard verification system and verification method |
| CN112735011A (en) * | 2019-10-15 | 2021-04-30 | 四川省数字商企智能科技有限公司 | Identification system and method for oil and gas station |
| CN111951452A (en) * | 2020-08-24 | 2020-11-17 | 合肥优恩物联网科技有限公司 | Thousand people thousand faces information push cloud platform based on entrance guard video information gathers |
| CN113191725A (en) * | 2021-04-20 | 2021-07-30 | 吕世福 | Method and system for managing and controlling access of hospital personnel, storage medium, equipment and terminal |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003122872A (en) * | 2001-10-17 | 2003-04-25 | Ricoh Co Ltd | Visitor reception system, program and recording medium |
| CN104573619A (en) * | 2014-07-25 | 2015-04-29 | 北京智膜科技有限公司 | Method and system for analyzing big data of intelligent advertisements based on face identification |
| CN106548545A (en) * | 2016-10-25 | 2017-03-29 | 深圳市万睿智能科技有限公司 | Gate inhibition and the System and method for of advertising linkage |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2711143C (en) * | 2007-12-31 | 2015-12-08 | Ray Ganong | Method, system, and computer program for identification and sharing of digital images with face signatures |
| CN104112209A (en) * | 2013-04-16 | 2014-10-22 | 苏州和积信息科技有限公司 | Audience statistical method of display terminal, and audience statistical system of display terminal |
| CN104464056A (en) * | 2014-12-12 | 2015-03-25 | 北京中研众力科技有限公司 | Facial recognition access control equipment with advertising information media function |
| WO2016187692A1 (en) * | 2015-05-27 | 2016-12-01 | Idk Interactive Inc. | Display systems using facial recognition for viewership monitoring purposes |
| CN106296264A (en) * | 2016-07-28 | 2017-01-04 | 河海大学常州校区 | A kind of pushing intelligent advertisements system based on recognition of face |
| CN106846066A (en) * | 2017-02-07 | 2017-06-13 | 张帆 | Intelligent advertisement put-on method and system |
| CN106887062A (en) * | 2017-02-13 | 2017-06-23 | 合肥美信息科技有限公司 | Recognition of face advertisement gate inhibition's main frame based on high in the clouds |
| CN108492431B (en) * | 2018-05-30 | 2024-03-29 | 中海云智慧（北京）物联网科技有限公司 | Intelligent access control system |
-
2018
- 2018-11-27 CN CN201811422542.2A patent/CN109544754B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003122872A (en) * | 2001-10-17 | 2003-04-25 | Ricoh Co Ltd | Visitor reception system, program and recording medium |
| CN104573619A (en) * | 2014-07-25 | 2015-04-29 | 北京智膜科技有限公司 | Method and system for analyzing big data of intelligent advertisements based on face identification |
| CN106548545A (en) * | 2016-10-25 | 2017-03-29 | 深圳市万睿智能科技有限公司 | Gate inhibition and the System and method for of advertising linkage |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109544754A (en) | 2019-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109544754B (en) | Access control method and system based on face recognition and computer equipment | |
| CN107330396B (en) | Pedestrian re-identification method based on multi-attribute and multi-strategy fusion learning | |
| Ryoo et al. | An overview of contest on semantic description of human activities (sdha) 2010 | |
| Zheng et al. | Partial person re-identification | |
| CN104881637B (en) | Multimodal information system and its fusion method based on heat transfer agent and target tracking | |
| CN110414441B (en) | Pedestrian track analysis method and system | |
| CN105787472B (en) | A kind of anomaly detection method based on the study of space-time laplacian eigenmaps | |
| CN109784274B (en) | Method for identifying trailing and related product | |
| CN110598543B (en) | Model training method based on attribute mining and reasoning and pedestrian re-identification method | |
| CN106845373A (en) | Towards pedestrian's attribute forecast method of monitor video | |
| CN105389562B (en) | A kind of double optimization method of the monitor video pedestrian weight recognition result of space-time restriction | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| CN103049459A (en) | Feature recognition based quick video retrieval method | |
| CN100589117C (en) | Gender recognition method based on gait | |
| CN107315795B (en) | The instance of video search method and system of joint particular persons and scene | |
| CN104091176A (en) | Technology for applying figure and head portrait comparison to videos | |
| CN111985348B (en) | Face recognition method and system | |
| CN104504408A (en) | Human face identification comparing method and system for realizing the method | |
| CN111047621A (en) | Target object tracking method, system, equipment and readable medium | |
| CN112464730B (en) | Pedestrian re-identification method based on domain-independent foreground feature learning | |
| CN113269091A (en) | Personnel trajectory analysis method, equipment and medium for intelligent park | |
| CN109902644A (en) | Face identification method, device, equipment and computer-readable medium | |
| CN115841683B (en) | Lightweight pedestrian re-identification method combining multi-level features | |
| CN111353338A (en) | Energy efficiency improvement method based on business hall video monitoring | |
| CN108681928A (en) | A kind of intelligent advertisement put-on method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |