CN117197876B - Face recognition security system and method based on deep learning - Google Patents
Face recognition security system and method based on deep learning Download PDFInfo
- Publication number
- CN117197876B CN117197876B CN202311464753.3A CN202311464753A CN117197876B CN 117197876 B CN117197876 B CN 117197876B CN 202311464753 A CN202311464753 A CN 202311464753A CN 117197876 B CN117197876 B CN 117197876B
- Authority
- CN
- China
- Prior art keywords
- image
- facial
- module
- face
- identity
- 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
- 238000013135 deep learning Methods 0.000 title claims abstract description 31
- 238000000034 method Methods 0.000 title claims abstract description 18
- 230000001815 facial effect Effects 0.000 claims abstract description 98
- 238000012790 confirmation Methods 0.000 claims abstract description 51
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims description 35
- 238000003708 edge detection Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 3
- 238000012544 monitoring process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention relates to the technical field of face recognition, in particular to a face recognition security system and a face recognition security method based on deep learning. The system comprises an image processing system, an identity confirmation module, a deep learning module and a safety control module; the image processing system acquires the facial image in the security guard in real time. The invention can shoot the security area and the checkpoint position, after the personnel passes the facial recognition of the checkpoint position, the personnel can successfully enter the security area, if the facial recognition is not passed at the beginning, the facial recognition is passed after the position is adjusted, or the personnel operation is matched with the identity information, the failed facial image is stored in the identity confirmation module, if the continuous facial recognition is not passed, but the facial image which is not passed is stored when the facial recognition is successful at the checkpoint position, the learning effect of the facial recognition is realized, the success rate and the accuracy of the facial recognition at the later stage are effectively improved, and the security quality in the security area is ensured.
Description
Technical Field
The invention relates to the technical field of face recognition, in particular to a face recognition security system and a face recognition security method based on deep learning.
Background
The face recognition technology is based on facial features of a person, firstly judging whether the face exists in an input face image or video stream, and if the face exists, further giving the position and the size of each face and the position information of each main facial organ. According to the information, the identity characteristics contained in each face are further extracted and compared with the known faces, so that the identity of each face is identified, in certain areas needing to be protected, in order to determine the identity of each person and the safety in security areas, the face recognition technology is widely used, and when the face recognition technology collects the information of people allowed to pass in the security areas, if the face recognition technology needs to collect the information comprehensively, a large amount of time is required, and the fact that the face recognition accuracy and success rate are lower when the face recognition technology does not collect the information comprehensively, therefore, the face recognition security system and the face recognition security method based on the deep learning are provided.
Disclosure of Invention
The invention aims to provide a face recognition security system and a face recognition security method based on deep learning, which are used for solving the problems in the background technology.
In order to achieve the above purpose, one of the purposes of the present invention is to provide a face recognition security system based on deep learning, which comprises an image processing system, an identity confirmation module, a deep learning module and a security control module;
the image processing system acquires face images in the security guard in real time, then carries out preprocessing on the acquired face images, improves the availability of the face images, the identity confirmation module extracts the face features in the image processing module after the image processing system carries out preprocessing on the face images, confirms the identity according to the face features, the deep learning module continuously acquires information transmitted by the image preprocessing module after the identity confirmation module confirms the identity, transmits the face images to the identity confirmation module for carrying out face recognition, stores the face images which are judged to be wrong by the identity confirmation module in the identity confirmation module according to the recognition result of the later stage, and the security control module controls security equipment according to the recognition result of the identity confirmation module.
As a further improvement of the technical scheme, the image processing system comprises an image acquisition module and an image preprocessing module;
the image acquisition module acquires facial images by adopting shooting equipment, and the image preprocessing module preprocesses the facial images after the image acquisition module acquires the facial images.
As a further improvement of the technical scheme, the image acquisition module adopts a camera to shoot the face of the person entering and exiting in real time, and the face image is transmitted to the image preprocessing module after being obtained.
As a further improvement of the technical scheme, after the image acquisition module acquires the facial image and converts the facial image into an electric signal, the image preprocessing module carries out preprocessing on the information transmitted by the image acquisition module, and the preprocessing comprises image size adjustment and denoising.
As a further improvement of the technical scheme, after the image preprocessing module preprocesses the face image acquired by the image acquisition module, the identity confirmation module performs edge detection on the face image, performs feature extraction on the face image, and then compares the extracted features with facial feature information in a database to confirm the identity of the facial features.
As a further improvement of the technical scheme, after the identity confirmation module confirms the identity of the personnel entering and exiting the security area, the image acquisition module continuously acquires the facial information of the personnel and transmits the facial information to the deep learning module, and the deep learning module transmits the new facial information to a database of the identity confirmation module for enriching the facial features of the personnel.
As a further improvement of the technical scheme, after the identity confirmation module performs face recognition on the personnel, the safety control module controls the entrance gateway of the security area according to the recognition result of the identity confirmation module and notifies the manager of the face recognition result.
The second object of the present invention is to provide a face recognition security method based on deep learning, including any one of the face recognition security systems based on deep learning, including the following method steps:
s1, an image acquisition module shoots the security area and the entrance and exit positions in real time, and then an image preprocessing module preprocesses the shot facial image;
s2, extracting the characteristics of the facial image in the image preprocessing module and comparing the facial image with the facial characteristics stored in the identity confirmation module, so as to determine the identity information of the personnel;
s3, the deep learning module collects face images which are not passed by face recognition, and stores facial features which are judged to be wrong by the identity confirmation module in the identity confirmation module according to a later recognition result;
and S4, the security control module controls security facilities in the security area according to the control result of the identity confirmation module.
Compared with the prior art, the invention has the beneficial effects that:
according to the face recognition security system and the face recognition security method based on deep learning, the interior of the security area and the position of the checkpoint can be shot, after face recognition of personnel passing through the position of the checkpoint, the personnel can successfully enter the interior of the security area, if the personnel does not pass through the face recognition at the beginning, the recognition passes through the position after the position is adjusted, or the personnel operation is matched with identity information, the face images which do not pass through the position are stored in the identity confirmation module, meanwhile, the face recognition is continuously carried out on each person in the interior of the security area, if continuous face recognition does not pass through the position, but the face images which do not pass through the recognition are successfully recognized at the position of the checkpoint, the learning effect of the face recognition is realized, the success rate and the accuracy of the face recognition at the later stage are effectively improved, and the security quality in the security area is ensured.
Drawings
Fig. 1 is an overall flow diagram of the present invention.
The meaning of each reference sign in the figure is:
1. an image acquisition module;
2. an image preprocessing module;
3. an identity confirmation module;
4. a deep learning module;
5. and a safety control module.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Example 1: referring to fig. 1, one of the purposes of the present embodiment is to provide a face recognition security system based on deep learning, which includes an image processing system, an identity confirmation module 3, a deep learning module 4 and a security control module 5;
the image processing system collects face images in the security guard in real time, then pre-processes the collected face images, usability of the face images is improved, the identity confirming module 3 extracts face features in the image processing module after the image processing system pre-processes the face images, identity is confirmed according to the face features, the deep learning module 4 continuously collects information transmitted by the image pre-processing module 2 after the identity confirming module 3 confirms the identity, the face images are transmitted to the identity confirming module 3 for face recognition, the face images with the identity confirming module 3 judging errors are stored in the identity confirming module 3 according to a later recognition result, and the security control module 5 controls security equipment according to the recognition result of the identity confirming module 3.
The image processing system comprises an image acquisition module 1 and an image preprocessing module 2;
the image acquisition module 1 acquires the facial image by adopting shooting equipment, and the image preprocessing module 2 preprocesses the facial image after the image acquisition module 1 acquires the facial image.
The image acquisition module 1 adopts a camera to shoot the face of an entering person in real time, the face image is transmitted to the image preprocessing module 2 after being obtained, one of the cameras is arranged at the entrance position, when the person enters and exits the security position, the camera needs to be positioned at the shooting position of the camera to shoot, and the entrance door is opened after the identity is confirmed; the other type of monitoring equipment is arranged in a security area, is used as monitoring equipment in the security area at ordinary times, collects face information of people when the people pass by, realizes identity confirmation in the security area and richness of the face information, and processes the face image by a computer after the face image is shot by a camera so as to convert the face image into an electric signal, so that the electric signal is conveniently transmitted to the image preprocessing module 2 for further processing.
After the image acquisition module 1 acquires the facial image and converts the facial image into an electric signal, the image preprocessing module 2 carries out preprocessing on the information transmitted by the image acquisition module 1, wherein the preprocessing comprises image size adjustment and denoising, and before the facial information is analyzed, the sizes of the photographed facial image are different due to different facial distances between a camera and a photographed person, so that the facial image needs to be scaled according to the actual situation, so that the size is suitable; when the camera collects the facial image, the camera is affected by various noises, such as spiced salt noise, gaussian noise and the like, which can affect the accuracy of a subsequent algorithm, but the denoising methods can affect the details of the facial image, so that a proper denoising method is adopted according to the actual situation of camera setting, and the usability of the facial image is improved.
After preprocessing the facial image acquired by the image acquisition module 1, the identity confirmation module 3 performs edge detection on the facial image, performs feature extraction on the facial image, then compares the extracted features with facial feature information in a database, confirms the identity of the facial features, and the edge detection of the image is a basic research direction and plate in image processing, and the main principle is that pixel points with obvious color changes or brightness changes in the digital image are identified, the obvious changes of the pixel points often represent important changes of the attribute of the image, wherein the significant changes comprise discontinuity in depth, discontinuity in direction, discontinuity in brightness and the like, when the edge detection algorithm detects the edge of the image, firstly roughly detects some pixel points of the image contour, then connects the pixel points through a face connection rule, finally detects and connects some boundary points which are not identified before, removes the detected false pixel points and boundary points and forms an integral edge, and adopts an LBP algorithm to extract the feature of the facial image as follows after obtaining an integral edge:
(1) determining a circle center g0 in the face image;
(2) defining R, wherein R is the number of pixel points from a circle center g0 in the face image;
(3) defining P, wherein P is the number of pixel points formed inside a circle P after a circle is drawn in the face image by taking R as a circle center;
(4) judging the pixel points according to the following judgment basis: dividing the pixel points formed in the step (3) in a multi-scale mode to form a plurality of 9 multiplied by 9 square grids, wherein the gray value of the surrounding square grids in the 9 multiplied by 9 square grids is larger than the gray value in the middle, and the square grids are marked as 1; the gray value of the surrounding grids in the 9X 9 grids is smaller than the gray value in the middle, the grids are marked as 0, a starting point is defined in the 9X 9 grids, the grids rotate according to the starting point, and binary codes are listed according to the grids passing through the rotation;
(5) converting binary codes into decimal numbers, wherein the decimal numbers are LBPs of the pixel areas, the LBPs are ranges of 0 lambda P, and the LBPs are local binary patterns;
(6) obtaining LBP of each pixel segmentation, establishing a histogram, wherein the type of each LBP is taken as a horizontal axis, the occurrence frequency of each LBP is taken as a vertical axis, and the histogram is the characteristic of the image.
Searching and comparing according to the characteristics in the facial image and the database, and if the database has the images with the same characteristics as the facial image, confirming the identity of the images; if there is no image in the database that has the same characteristics as the facial image, its identity cannot be confirmed.
After the identity confirmation module 3 confirms the identity of the personnel entering and exiting the security area, the image acquisition module 1 continuously acquires the facial information of the personnel and transmits the facial information to the deep learning module 4, the deep learning module 4 transmits the new facial information to the database of the identity confirmation module 3 for enriching the facial features of the personnel, as the facial features are generally recognized, the facial features are generally extracted, the extracted features are compared with the information stored in the database, the identity of the personnel is further distinguished, if the identity of the personnel is recognized, the shooting angle is different from the original angle recorded in the database, the identity confirmation is failed, and then the personnel is required to be confirmed, so that when the security system is just put into use, if the identity confirmation failure occurs, the personnel operation end selects the corresponding identity and allows the personnel to enter, the characteristics of the facial image failing in identity confirmation are stored in a database, when a person is in a security area, a camera in the security area monitors the security area, acquires image information of the face of the person, compares the image information with information stored in an identity confirmation module 3, sets proper comparison times, if the continuous comparison times of the person reach a set value and are not displayed, the person needs to conduct facial recognition at an entrance and a exit, if the facial recognition at the entrance and the exit is successful, the facial image characteristics failing in comparison are stored in the database, the effect of distinguishing the identity of the person in the security area is achieved, the database is continuously enriched, the success rate of the subsequent facial recognition of the person is improved, and when the person enters the security area through other entrances and exits in the security area, and the identification is performed quickly, so that the security of security is improved.
After the identity confirming module 3 carries out face recognition on the personnel, the security control module 5 controls the entrance and exit gate of the security area according to the recognition result of the identity confirming module 3, and informs the manager of the face recognition result, when the personnel is positioned at the entrance and exit gate of the security area, the gate is directly controlled by the result of the face recognition, if the face recognition is passed, the gate is directly opened, if the face recognition is not passed, the gate is locked, and the facial features are transmitted to the manual control end, at the moment, the personnel can carry out face shooting again at the original position, after the face recognition is passed, the gate is opened, the failed facial image is stored in the corresponding personnel identity information, or the manual control end is directly matched with the proper personnel identity, and the gate is opened, at the moment, the failed facial image is stored in the corresponding personnel identity information, so that the success rate of the face recognition is conveniently improved at the later stage; when the personnel is in the security area, if the continuous comparison times of the personnel reach the set value, the personnel are informed of the fact that the continuous comparison times of the personnel do not pass through the set value, and the personnel in the security area are informed of reconfirming.
The second object of the present embodiment is to provide a face recognition security method based on deep learning, including any one of the face recognition security systems based on deep learning, including the following method steps:
s1, an image acquisition module 1 shoots the security area and the entrance and exit checkpoint position in real time, and then an image preprocessing module 2 preprocesses the shot facial image;
s2, extracting features of facial images in the image preprocessing module 2 and comparing the features with facial features stored in the identity confirmation module 3, so as to determine identity information of personnel;
s3, the deep learning module 4 collects face images which are not passed by face recognition, and stores the facial features which are judged to be wrong by the identity confirmation module 3 in the identity confirmation module 3 according to a later recognition result;
and S4, the safety control module 5 controls security facilities in the security area according to the control result of the identity confirmation module 3.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the above-described embodiments, and that the above-described embodiments and descriptions are only preferred embodiments of the present invention, and are not intended to limit the invention, and that various changes and modifications may be made therein without departing from the spirit and scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (1)
1. Face identification security protection system based on deep learning, its characterized in that: comprises an image processing system, an identity confirmation module (3), a deep learning module (4) and a safety control module (5);
the image processing system acquires face images in a security guard in real time, then carries out preprocessing on the acquired face images, improves the availability of the face images, the identity confirmation module (3) extracts face features in the image processing module after the image processing system carries out preprocessing on the face images, confirms identity according to the face features, the deep learning module (4) continuously acquires information transmitted by the image preprocessing module (2), transmits the face images to the identity confirmation module (3) for carrying out face recognition, stores the face images which are judged to be wrong by the identity confirmation module (3) in the identity confirmation module (3) according to a later recognition result, and the safety control module (5) controls security equipment according to the recognition result of the identity confirmation module (3);
the image processing system comprises an image acquisition module (1) and an image preprocessing module (2);
the image acquisition module (1) acquires facial images by adopting shooting equipment, and the image preprocessing module (2) preprocesses the facial images after the image acquisition module (1) acquires the facial images;
the image acquisition module (1) adopts a camera to shoot the face of an entering person in real time, and the face image is transmitted to the image preprocessing module (2) after being obtained;
after the image acquisition module (1) acquires the facial image and converts the facial image into an electric signal, the image preprocessing module (2) preprocesses the information transmitted by the image acquisition module (1), wherein the preprocessing comprises image size adjustment and denoising;
after the image preprocessing module (2) preprocesses the facial image acquired by the image acquisition module (1), the identity confirmation module (3) performs edge detection on the facial image, performs feature extraction on the facial image, and then compares the extracted features with facial feature information in a database to confirm the identity of the facial features;
when the edge of an image is detected by an edge detection algorithm, firstly detecting some pixel points of the image outline, then connecting the pixel points by a face connection rule, finally detecting and connecting some boundary points which are not recognized before, removing false pixel points and boundary points which are detected, and forming an integral edge, and extracting the characteristics of the face image by an LBP algorithm after the integral edge is obtained, wherein the LBP algorithm comprises the following steps:
(1) determining a circle center g0 in the face image;
(2) defining R, wherein R is the number of pixel points from a circle center g0 in the face image;
(3) defining P, wherein P is the number of pixel points formed inside a circle P after a circle is drawn in the face image by taking R as a circle center;
(4) judging the pixel points according to the following judgment basis: dividing the pixel points formed in the step (3) in a multi-scale mode to form a plurality of 9 multiplied by 9 square grids, wherein the gray value of the surrounding square grids in the 9 multiplied by 9 square grids is larger than the gray value in the middle, and the square grids are marked as 1; the gray value of the surrounding grids in the 9X 9 grids is smaller than the gray value in the middle, the grids are marked as 0, a starting point is defined in the 9X 9 grids, the grids rotate according to the starting point, and binary codes are listed according to the grids passing through the rotation;
(5) converting binary codes into decimal numbers, wherein the decimal numbers are LBPs of pixel areas, the LBPs are ranges of 0 lambda P, and the LBPs are local binary patterns;
(6) obtaining LBP of each pixel segmentation, establishing a histogram, wherein the type of each LBP is taken as a horizontal axis, the occurrence frequency of each LBP is taken as a vertical axis, and the histogram is the characteristic of the image;
after the identity confirmation module (3) confirms the identity of the personnel entering and exiting the security area, the image acquisition module (1) continuously acquires the facial information of the personnel and transmits the facial information to the deep learning module (4), and the deep learning module (4) transmits the new facial information to a database of the identity confirmation module (3) for enriching the facial features of the personnel;
when the security system is just put into use, if identity confirmation fails, a manual operation end selects a corresponding identity and allows the person to enter, the characteristics of a facial image failing in identity confirmation at the time are stored in a database, when the person is in the security area, a camera in the security area monitors the security area, acquires image information of the face of the person, compares the image information with information stored in an identity confirmation module (3), sets proper comparison times, if the continuous comparison times of the person reach a set value, the person needs to perform facial recognition at an entrance and the face recognition at the entrance is successful, and stores the facial image characteristics failing in comparison in the database;
after the identity confirmation module (3) performs facial recognition on the personnel, the safety control module (5) controls the entrance gateway of the security area according to the recognition result of the identity confirmation module (3) and notifies the facial recognition result to the manager;
when a person is positioned at an entrance gateway of a security area, the gateway is directly controlled through the result of facial recognition, if the facial recognition passes, the gateway is directly opened, if the facial recognition does not pass, the gateway is locked, facial features are transmitted to a manual control end, at the moment, the person shoots the face again at the original position, after the facial recognition passes, the gateway is opened, the facial images which do not pass are stored in corresponding personnel identity information, or the manual control end is directly matched with the proper personnel identity, the gateway is opened, at the moment, the facial images which do not pass are stored in the corresponding personnel identity information, so that the success rate of the facial recognition is improved in the later period; when the personnel are in the security area, if the continuous comparison times of the personnel reach the set value, the personnel are informed of the fact that the personnel do not pass through the set value, and the personnel are informed of the fact that the personnel are reconfirmed in the security area;
the using method of the system comprises the following method steps:
s1, an image acquisition module (1) shoots the security area and the entrance and exit checkpoint position in real time, and then an image preprocessing module (2) preprocesses the shot facial image;
s2, extracting characteristics of facial images in the image preprocessing module (2) and comparing the characteristics with facial characteristics stored in the identity confirmation module (3), so as to determine identity information of personnel;
s3, the deep learning module (4) collects face images which are not passed by face recognition, and stores facial features which are judged to be wrong by the identity confirmation module (3) in the identity confirmation module (3) according to a later recognition result;
s4, the security control module (5) controls security facilities in the security area according to the control result of the identity confirmation module (3).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311464753.3A CN117197876B (en) | 2023-11-07 | 2023-11-07 | Face recognition security system and method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311464753.3A CN117197876B (en) | 2023-11-07 | 2023-11-07 | Face recognition security system and method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117197876A CN117197876A (en) | 2023-12-08 |
| CN117197876B true CN117197876B (en) | 2024-04-09 |
Family
ID=89005646
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311464753.3A Active CN117197876B (en) | 2023-11-07 | 2023-11-07 | Face recognition security system and method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117197876B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354909A (en) * | 2015-12-22 | 2016-02-24 | 重庆金瓯科技发展有限责任公司 | Intelligent safety inspection monitoring information system |
| CN207216700U (en) * | 2017-08-01 | 2018-04-10 | 深圳市永达电子信息股份有限公司 | A kind of system that speeds passage through customs based on recognition of face |
| CN109326009A (en) * | 2017-08-01 | 2019-02-12 | 深圳市永达电子信息股份有限公司 | A kind of system and method that speeds passage through customs based on recognition of face |
| CN110175572A (en) * | 2019-05-28 | 2019-08-27 | 深圳市商汤科技有限公司 | Face image processing process and device, electronic equipment and storage medium |
| CN110428522A (en) * | 2019-07-24 | 2019-11-08 | 青岛联合创智科技有限公司 | A kind of intelligent safety and defence system of wisdom new city |
| CN112132048A (en) * | 2020-09-24 | 2020-12-25 | 天津锋物科技有限公司 | Community patrol analysis method and system based on computer vision |
| CN113673450A (en) * | 2021-08-25 | 2021-11-19 | 视路信(无锡)视觉科技有限公司 | Visual system with autonomous learning function |
| CN113822907A (en) * | 2021-10-18 | 2021-12-21 | 海信集团控股股份有限公司 | Image processing method and device |
| CN115393276A (en) * | 2022-07-20 | 2022-11-25 | 华能国际电力股份有限公司河北清洁能源分公司 | Intelligent detection and analysis system for surface morphology of gear |
| CN115396705A (en) * | 2022-08-19 | 2022-11-25 | 上海哔哩哔哩科技有限公司 | Screen projection operation verification method, platform and system |
| CN115564759A (en) * | 2022-10-27 | 2023-01-03 | 深圳市好伙计科技有限公司 | Food material quality monitoring system based on big data |
| CN116050802A (en) * | 2023-02-21 | 2023-05-02 | 深圳市神州通在线科技有限公司 | Enterprise collaborative office application platform based on image analysis |
| CN116416665A (en) * | 2023-04-04 | 2023-07-11 | 江西佳信捷电子股份有限公司 | Face recognition method and device based on security system and storage medium |
| CN116778558A (en) * | 2023-07-03 | 2023-09-19 | 云南贝泰妮生物科技集团股份有限公司 | Face wrinkle detection method and system based on adaptive hybrid hessian filter |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11689526B2 (en) * | 2019-11-19 | 2023-06-27 | Paypal, Inc. | Ensemble method for face recognition deep learning models |
-
2023
- 2023-11-07 CN CN202311464753.3A patent/CN117197876B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105354909A (en) * | 2015-12-22 | 2016-02-24 | 重庆金瓯科技发展有限责任公司 | Intelligent safety inspection monitoring information system |
| CN207216700U (en) * | 2017-08-01 | 2018-04-10 | 深圳市永达电子信息股份有限公司 | A kind of system that speeds passage through customs based on recognition of face |
| CN109326009A (en) * | 2017-08-01 | 2019-02-12 | 深圳市永达电子信息股份有限公司 | A kind of system and method that speeds passage through customs based on recognition of face |
| CN110175572A (en) * | 2019-05-28 | 2019-08-27 | 深圳市商汤科技有限公司 | Face image processing process and device, electronic equipment and storage medium |
| CN110428522A (en) * | 2019-07-24 | 2019-11-08 | 青岛联合创智科技有限公司 | A kind of intelligent safety and defence system of wisdom new city |
| CN112132048A (en) * | 2020-09-24 | 2020-12-25 | 天津锋物科技有限公司 | Community patrol analysis method and system based on computer vision |
| CN113673450A (en) * | 2021-08-25 | 2021-11-19 | 视路信(无锡)视觉科技有限公司 | Visual system with autonomous learning function |
| CN113822907A (en) * | 2021-10-18 | 2021-12-21 | 海信集团控股股份有限公司 | Image processing method and device |
| CN115393276A (en) * | 2022-07-20 | 2022-11-25 | 华能国际电力股份有限公司河北清洁能源分公司 | Intelligent detection and analysis system for surface morphology of gear |
| CN115396705A (en) * | 2022-08-19 | 2022-11-25 | 上海哔哩哔哩科技有限公司 | Screen projection operation verification method, platform and system |
| CN115564759A (en) * | 2022-10-27 | 2023-01-03 | 深圳市好伙计科技有限公司 | Food material quality monitoring system based on big data |
| CN116050802A (en) * | 2023-02-21 | 2023-05-02 | 深圳市神州通在线科技有限公司 | Enterprise collaborative office application platform based on image analysis |
| CN116416665A (en) * | 2023-04-04 | 2023-07-11 | 江西佳信捷电子股份有限公司 | Face recognition method and device based on security system and storage medium |
| CN116778558A (en) * | 2023-07-03 | 2023-09-19 | 云南贝泰妮生物科技集团股份有限公司 | Face wrinkle detection method and system based on adaptive hybrid hessian filter |
Non-Patent Citations (4)
| Title |
|---|
| 基于Android平台的人脸识别系统的设计与实现;周师专;《中国优秀硕士学位论文全文数据库 (信息科技辑)》;第I138-3566页 * |
| 基于人脸识别的网络身份认证;顾勤丰;;电子技术与软件工程(第12期);第245-246页 * |
| 顾勤丰 ; .基于人脸识别的网络身份认证.电子技术与软件工程.2020,(第12期), * |
| 顾勤丰 ; .基于人脸识别的网络身份认证.电子技术与软件工程.2020,(第12期),第245-246页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117197876A (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106056079B (en) | A kind of occlusion detection method of image capture device and human face five-sense-organ | |
| CN107423690B (en) | Face recognition method and device | |
| US7215798B2 (en) | Method for forgery recognition in fingerprint recognition by using a texture classification of gray scale differential images | |
| KR100831122B1 (en) | Face authentication apparatus, face authentication method, and entrance and exit management apparatus | |
| CN109298785A (en) | A kind of man-machine joint control system and method for monitoring device | |
| CN111881726B (en) | Living body detection method and device and storage medium | |
| US20080260239A1 (en) | Object image detection method | |
| CN112001360B (en) | Face recognition monitoring system based on intelligent adjustment | |
| CN101556717A (en) | ATM intelligent security system and monitoring method | |
| Lee et al. | An automated video-based system for iris recognition | |
| CN110175553B (en) | Method and device for establishing feature library based on gait recognition and face recognition | |
| CN112183162A (en) | Face automatic registration and recognition system and method in monitoring scene | |
| KR102423934B1 (en) | Smart human search integrated solution through face recognition and multiple object tracking technology of similar clothes color | |
| CN117197876B (en) | Face recognition security system and method based on deep learning | |
| CN115984973B (en) | Human body abnormal behavior monitoring method for peeping-preventing screen | |
| CN112329597A (en) | Face recognition method and device | |
| JP2012212968A (en) | Image monitoring apparatus | |
| JP2006178651A (en) | Person recognition apparatus and method, and passage control apparatus | |
| US20120219192A1 (en) | Method of controlling a session at a self-service terminal, and a self-service terminal | |
| JP2020063659A (en) | Information processing system | |
| JP2004118359A (en) | Figure recognizing device, figure recognizing method and passing controller | |
| CN116416565A (en) | Method and system for detecting pedestrian trailing and crossing in specific area | |
| CN111062294B (en) | Passenger flow queuing time detection method, device and system | |
| CN113095289A (en) | Massive image preprocessing network method based on urban complex scene | |
| CN112215064A (en) | Face recognition method and system for public safety precaution |
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 |