CN107480658B - Face recognition device and method based on multi-angle video - Google Patents
Face recognition device and method based on multi-angle video Download PDFInfo
- Publication number
- CN107480658B CN107480658B CN201710851859.7A CN201710851859A CN107480658B CN 107480658 B CN107480658 B CN 107480658B CN 201710851859 A CN201710851859 A CN 201710851859A CN 107480658 B CN107480658 B CN 107480658B
- Authority
- CN
- China
- Prior art keywords
- face
- main thread
- recognition
- threads
- camera
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
Landscapes
- Engineering & Computer Science (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a face recognition device and a method based on a multi-angle video, the face recognition device based on the multi-angle video comprises a face recognition processing control unit, a recognition channel and at least three cameras which are arranged on the recognition channel and are in signal connection with the face recognition processing control unit, wherein the at least three cameras comprise a first camera which is arranged right above the recognition channel and is used for collecting the face characteristics of a user, at least one second camera which is arranged at the left side part of the recognition channel and is used for collecting the left side face characteristics of the user and at least one third camera which is arranged at the right side part of the recognition channel and is used for collecting the right side face characteristics of the user. The face recognition method using the face recognition device based on the multi-angle video comprises face registration, face detection and recognition.
Description
Technical Field
The invention relates to a face recognition device and method based on a multi-angle video.
Background
With the development of the biometric technology, the face recognition technology is more and more going into people's lives. The face recognition attendance checking technology has the advantages of non-contact, intuition, friendliness, strong applicability and the like. At present, the face recognition technical layer breaks through the influence of day night light, and can achieve rapid recognition in a natural state. With the continuous maturity of the technology and the reduction of the cost, the face attendance gradually shows the trend of replacing the fingerprint attendance. In the aspect of biological identification, the application of the face identification technology is gradually beyond the technical field of security protection.
The face recognition system mainly comprises four modules of face detection, face alignment, feature extraction and identity recognition. The face detection comprises the steps of firstly judging whether a face exists on a given image, and if the face exists, outputting position coordinates of the face in the image and the size of the face. The human face alignment is to automatically position key feature points of the face, such as eyes, nose tips, corner points of the mouth, eyebrows, contour points of each part of the human face and the like according to an input human face image. The accuracy of face recognition can be improved through face alignment. The face feature extraction is to express a face image into a high-dimensional feature vector, so that the features of the face can be quantized, and the face similarity can be calculated. The identity recognition is to compare the face to be recognized with the existing face and find out the face label with the highest similarity, so as to determine the identity of the person.
At present, the classification is carried out according to a method for extracting human face features, and the human face recognition technology can be roughly divided into two types, namely a traditional feature extraction method based on artificial structural features and a feature extraction method based on deep learning. The face recognition method based on deep learning becomes a mainstream method of the face recognition technology, and is widely applied in academic and industrial fields. A deep learning-based feature extraction method generally includes the steps of firstly collecting human faces, then extracting human face features, sending the extracted human face features to an existing classifier for training, generating a training model, when the identity of a certain user needs to be determined, firstly carrying out human face detection, then carrying out feature extraction on the human faces by using the same feature extraction method, then sending the extracted features to the trained classifier, and enabling the classifier to output a human face recognition result.
However, the existing face recognition system usually adopts a single image to recognize the face identity, which makes the collected image easily interfered by factors such as illumination, recognition environment, face pose and shooting angle, so that the accuracy can be influenced to a certain extent when the face identity is determined by using a face recognition algorithm from a single angle.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a face recognition device and method based on multi-angle video.
In order to achieve the above purpose, the present invention provides a technical solution: face identification device based on multi-angle video, it includes face identification processing control unit, the discernment passageway, install on the discernment passageway and with the at least three camera of face identification processing control unit phase sign connection, at least three camera is including setting up the first camera that is used for gathering the positive face characteristic of user directly over the discernment passageway, set up at the left side of discernment passageway and be used for gathering at least one second camera of user's left side face characteristic, set up at the right side of discernment passageway and be used for gathering at least one third camera of user's right side face characteristic.
Furthermore, a light supplement lamp used for providing fixed illumination for the camera is installed on the identification channel.
Further, there are two second cameras and two third cameras.
The invention also provides another technical scheme: the face recognition method using the face recognition device based on the multi-angle video comprises face registration, face detection and recognition.
The face registration comprises the following steps:
a1. and designating the control thread of one camera as a main thread, wherein the main thread is in an active state, reading image data collected by the corresponding camera, and the control threads of other cameras are in a blocking state. When the thread is in a blocking state, the thread does not read image data from the corresponding camera;
a2. when the main thread detects a face in the acquired image, other threads are awakened, and the image data is read from the corresponding cameras after the other threads are awakened. The thread is awakened, namely, the thread enters an active state from a blocking state;
a3. when the main thread cannot detect the human face within a preset time interval, the main thread informs other threads of entering a blocking state, and meanwhile, the other threads send the image data acquired by the main thread to the main thread;
a4. after receiving image data sent by other threads, the main thread counts the number of faces detected by each thread, if the number is larger than or equal to a preset value, the main thread extracts face features of all the faces detected by each thread, and then stores the face features and corresponding identity information into a face database; and if the number is smaller than the preset value, the system sends a prompt for re-acquisition.
The face detection and recognition comprises the following steps:
b1. and appointing a control thread of one camera as a main thread. The main thread is in an active state, image data collected by the corresponding camera is read, and the control threads of the other cameras are in a blocking state;
b2. when the main thread detects a face in the acquired image, other threads are awakened, and the other threads are awakened and then read image data from the corresponding cameras;
b3. when the main thread cannot detect the human face within a preset time interval, the main thread informs other threads of entering a blocking state, and meanwhile, the other threads send the image data acquired by the main thread to the main thread;
b4. the main thread extracts the face features of all faces detected by all threads, then similarity calculation is carried out on the extracted face features and the face features stored in a database, and a face recognition processing control unit selects a plurality of candidate sets with similarity meeting a preset rule according to the angle of the collected faces;
b5. and the face recognition processing control unit judges the identity of the user walking from the recognition channel in the candidate sets according to a preset recognition strategy.
Further, in step a1 and step b1, the control thread of the lower left camera is designated as the main thread.
Further, in step a2 and step b2, the main thread performs face detection on the acquired image at regular intervals, and when the main thread detects a face in the acquired image, other threads are awakened. Preferably, the fixed time is 200 milliseconds.
Further, in step a4, the main thread performs face feature extraction on all faces detected by each thread, and the method includes: the method comprises the steps of storing three human face features of different angles according to the installation positions of different cameras, wherein the human face features are respectively used for recognizing a front face, a left side face and a right side face.
Further, in step b4, when similarity calculation is performed between the extracted facial features and the facial features stored in the database, the front facial features stored in the database are used for comparison with the face collected by the first camera, the left facial features stored in the database are used for comparison with the face collected by the second camera, and the right facial features stored in the database are used for comparison with the face collected by the third camera.
Further, in step b5, the policy is identified as: accumulating the number of the same candidate objects in the multiple candidate sets, and if the proportion of the number of the same candidate objects in all the candidate objects is greater than a preset threshold, taking the candidate object corresponding to the current threshold as a final identification result; otherwise, the user is judged to be a stranger, namely not belonging to the registered user.
Further, in step b5, the policy is identified as: merging the candidate objects with the similarity greater than a preset threshold value in each candidate set into a new candidate set, accumulating the number of the same candidate objects in the obtained new candidate set, and taking the candidate object corresponding to the current threshold value as a final identification result if the proportion of the number of the same candidate objects in all the candidate objects is greater than a preset threshold value; otherwise, the user is judged to be a stranger, namely not belonging to the registered user.
Further, in step b5, the policy is identified as: firstly, selecting candidate objects with the maximum similarity in each candidate set, then forming a new candidate set by the candidate objects, counting the number of the same candidate objects in the new candidate set, and if the maximum value in the obtained values is greater than or equal to 3, taking the candidate object corresponding to the value as a final recognition result; if the maximum value of the obtained values is less than 3 and greater than or equal to 2 and the second largest value is not 2, taking the candidate object corresponding to the value of 2 as the final recognition result; if the maximum value in the obtained values is less than 3 and greater than or equal to 2 and the second largest value is also 2, comparing the similarity sums of the obtained values, and taking the candidate object corresponding to the maximum value of the similarity sum as a final recognition result; and if the maximum value in the obtained values is less than 2, judging that the user is a stranger, namely judging that the user does not belong to the registered user.
By adopting the technical scheme, the face recognition device and method based on the multi-angle video, disclosed by the invention, have the advantages that the plurality of cameras ensure that a considerable number of face images are collected, the face image information is collected from multiple angles, the recognition channel limits the moving range of a user, the user can actively cooperate to complete the face recognition process, and the light supplement lamp ensures that the face recognition system completes the face recognition process under the uniform illumination condition. Compared with the similar technology for collecting a single image, the method greatly weakens the interference of factors such as illumination, recognition environment, human face posture, shooting angle and the like when the image is collected, and effectively improves the robustness and recognition rate of the human face recognition system.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood and to implement them in accordance with the contents of the description, the following detailed description is given with reference to the preferred embodiments of the present invention and the accompanying drawings.
Drawings
FIG. 1 is a schematic structural diagram of a face recognition device based on multi-angle video according to a first embodiment of the present invention;
FIG. 2 is a flowchart of the steps of face registration by a face recognition method using a face recognition apparatus based on a multi-angle video according to a first embodiment of the present invention;
FIG. 3 is a flow chart of the steps of face detection and recognition by a face recognition method using a face recognition apparatus based on a multi-angle video according to a first embodiment of the present invention;
FIG. 4 is a flowchart of the steps of face detection and recognition by a face recognition method using a face recognition apparatus based on a multi-angle video according to a second embodiment of the present invention;
fig. 5 is a flow chart of steps of face detection and recognition by a face recognition method using a face recognition apparatus based on a multi-angle video in the third embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Example one
Referring to fig. 1, the face recognition apparatus based on multi-angle video in this embodiment includes a face recognition processing control unit, a recognition channel (not shown in the drawing), at least three cameras installed on the recognition channel and connected with the face recognition processing control unit in a signal manner, wherein the at least three cameras include a first camera 1 disposed right above the recognition channel and used for collecting front face features of a user, at least one second camera 2 disposed at a left side of the recognition channel and used for collecting left side face features of the user, and at least one third camera 3 disposed at a right side of the recognition channel and used for collecting right side face features of the user.
In a more preferred embodiment, a fill-in light (not shown in the drawings) for providing fixed illumination for the camera is installed on the identification channel.
In a more preferred embodiment, there are two second cameras 2 and two third cameras 3.
The embodiment also provides another technical scheme: the face recognition method using the face recognition device based on the multi-angle video comprises face registration, face detection and recognition. (the present invention uses one computer to control multiple cameras, requiring the use of a multi-threaded mechanism.)
Referring to fig. 2, the face registration includes the following steps:
a1. the control thread of one camera is designated as the main thread, and preferably, the control thread of the camera on the lower left side is designated as the main thread. The main thread is in an active state, image data collected by the corresponding camera is read, and control threads of other cameras are in a blocking state. When the thread is in a blocking state, the thread does not read image data from the corresponding camera;
a2. the main thread performs face detection on the acquired image once at regular intervals, in this embodiment, the regular time is 200 milliseconds. When the main thread detects a face in the acquired image, other threads are awakened, and the image data is read from the corresponding cameras after the other threads are awakened. The thread is awakened, namely, the thread enters an active state from a blocking state;
a3. when the main thread cannot detect the human face within a preset time interval, the main thread informs other threads of entering a blocking state, and meanwhile, the other threads send the image data acquired by the main thread to the main thread;
a4. after receiving image data sent by other threads, the main thread counts the number of human faces detected by each thread, if the number is larger than or equal to a preset value, the main thread extracts human face features of all human faces detected by each thread, then stores the human face features and corresponding identity information into a human face database, and stores three human face features of different angles according to the installation positions of different cameras, wherein the human face features are respectively used for identifying a front face, a left face and a right face; and if the number is smaller than the preset value, the system sends a prompt for re-acquisition.
Referring to fig. 3, the face detection and recognition includes the following steps:
b1. and appointing a control thread of one camera as a main thread. The camera controlled by the thread can shoot the whole body of the user. Preferably, the control thread of the lower left camera is designated as the main thread. The main thread is in an active state, image data collected by the corresponding camera is read, and the control threads of the other cameras are in a blocking state;
b2. the main thread performs face detection on the acquired image once at regular intervals, in this embodiment, the regular time is 200 milliseconds. When the main thread detects a face in the acquired image, other threads are awakened, and the other threads are awakened and then read image data from the corresponding cameras;
b3. when the main thread cannot detect the human face within a preset time interval, the main thread informs other threads of entering a blocking state, and meanwhile, the other threads send the image data acquired by the main thread to the main thread;
b4. the main thread extracts the face features of all the faces detected by the threads, similarity calculation is carried out on the extracted face features and the face features stored in the database, for the face collected by the first camera, the front face features stored in the database are used for comparison, for the face collected by the second camera, the left face features stored in the database are used for comparison, and for the face collected by the third camera, the right face features stored in the database are used for comparison. The face recognition processing control unit selects a plurality of candidate sets with similarity meeting a preset rule according to the angle of the collected face;
b5. and the face recognition processing control unit judges the identity of the user walking from the recognition channel in the candidate sets according to a preset recognition strategy. The identification strategy is: accumulating the number of the same candidate objects in the multiple candidate sets, and if the proportion of the number of the same candidate objects in all the candidate objects is greater than a preset threshold, taking the candidate object corresponding to the current threshold as a final identification result; otherwise, the user is judged to be a stranger, namely not belonging to the registered user.
Example two
Referring to fig. 4, the present embodiment is different from the first embodiment only in that: in step b5 of the face recognition method using the face recognition apparatus based on multi-angle video, the recognition policy is: merging the candidate objects with the similarity greater than a preset threshold value in each candidate set into a new candidate set, accumulating the number of the same candidate objects in the obtained new candidate set, and taking the candidate object corresponding to the current threshold value as a final identification result if the proportion of the number of the same candidate objects in all the candidate objects is greater than a preset threshold value; otherwise, the user is judged to be a stranger, namely not belonging to the registered user.
EXAMPLE III
Referring to fig. 5, the present embodiment differs from the first embodiment only in that: in step b5 of the face recognition method using the face recognition apparatus based on multi-angle video, the recognition policy is: firstly, selecting candidate objects with the maximum similarity in each candidate set, then forming a new candidate set by the candidate objects, counting the number of the same candidate objects in the new candidate set, and if the maximum value in the obtained values is greater than or equal to 3, taking the candidate object corresponding to the value as a final recognition result; if the maximum value of the obtained values is less than 3 and greater than or equal to 2 and the second largest value is not 2, taking the candidate object corresponding to the value of 2 as the final recognition result; if the maximum value in the obtained values is less than 3 and greater than or equal to 2 and the second largest value is also 2, comparing the similarity sums of the obtained values, and taking the candidate object corresponding to the maximum value of the similarity sum as a final recognition result; and if the maximum value in the obtained values is less than 2, judging that the user is a stranger, namely judging that the user does not belong to the registered user.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, it should be noted that, for those skilled in the art, many modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (7)
1. A face recognition method of a face recognition device based on multi-angle video is characterized in that: the multi-angle video-based face recognition device comprises a face recognition processing control unit, a recognition channel and at least three cameras which are arranged on the recognition channel and are in signal connection with the face recognition processing control unit, wherein the at least three cameras comprise a first camera which is arranged right above the recognition channel and is used for collecting front face features of a user, at least one second camera which is arranged at the left side part of the recognition channel and is used for collecting left side face features of the user, and at least one third camera which is arranged at the right side part of the recognition channel and is used for collecting right side face features of the user;
the face recognition method comprises face registration, face detection and recognition, wherein the face registration comprises the following steps:
a1. appointing a control thread of one camera as a main thread, wherein the main thread is in an active state, reading image data collected by the corresponding camera, and the control threads of the other cameras are in a blocking state;
a2. when the main thread detects a face in the acquired image, other threads are awakened, and the other threads are awakened and then read image data from the corresponding cameras;
a3. when the main thread cannot detect the human face within a preset time interval, the main thread informs other threads of entering a blocking state, and meanwhile, the other threads send the image data acquired by the main thread to the main thread;
a4. after receiving image data sent by other threads, the main thread counts the number of faces detected by each thread, if the number is larger than or equal to a preset value, the main thread extracts face features of all the faces detected by each thread, and then stores the face features and corresponding identity information into a face database; if the number is smaller than a preset value, the system sends a prompt for re-acquisition;
the face detection and recognition comprises the following steps:
b1. appointing a control thread of one camera as a main thread, wherein the main thread is in an active state, reading image data collected by the corresponding camera, and the control threads of the other cameras are in a blocking state;
b2. when the main thread detects a face in the acquired image, other threads are awakened, and the other threads are awakened and then read image data from the corresponding cameras;
b3. when the main thread cannot detect the human face within a preset time interval, the main thread informs other threads of entering a blocking state, and meanwhile, the other threads send the image data acquired by the main thread to the main thread;
b4. the main thread extracts the face features of all faces detected by all threads, then similarity calculation is carried out on the extracted face features and the face features stored in a database, and a face recognition processing control unit selects a plurality of candidate sets with similarity meeting a preset rule according to the angle of the collected faces;
b5. and the face recognition processing control unit judges the identity of the user walking through the recognition channel in the candidate sets according to a preset recognition strategy.
2. The face recognition method of the face recognition device based on multi-angle video according to claim 1, characterized in that: in the step a2 and the step b2, the main thread performs face detection on the acquired image once at regular intervals, and when the main thread detects a face in the acquired image, other threads are awakened.
3. The face recognition method of the face recognition device based on multi-angle video according to claim 1, characterized in that: in the step a4, the main thread performs face feature extraction on all faces detected by each thread, and the method includes: the method comprises the steps of storing three human face features of different angles according to the installation positions of different cameras, wherein the human face features are respectively used for recognizing a front face, a left side face and a right side face.
4. The face recognition method of the face recognition device based on multi-angle video according to claim 1, characterized in that: step b4 in, when carrying out the similarity calculation to the face characteristic of extraction and the face characteristic of leaving in the database, for the face that first camera was gathered uses the positive face characteristic of saving in the database to compare with it, for the face that the second camera was gathered uses the left side face characteristic of saving in the database to compare with it, for the face that the third camera was gathered uses the right side face characteristic of saving in the database to compare with it.
5. The face recognition method of the face recognition device based on multi-angle video according to claim 1, characterized in that: in the step b5, the identification policy is: accumulating the number of the same candidate objects in the multiple candidate sets, and if the proportion of the number of the same candidate objects in all the candidate objects is greater than a preset threshold, taking the candidate object corresponding to the current threshold as a final identification result; otherwise, the user is judged to be a stranger, namely not belonging to the registered user.
6. The face recognition method of the face recognition device based on multi-angle video according to claim 1, characterized in that: in the step b5, the identification policy is: merging the candidate objects with the similarity greater than a preset threshold value in each candidate set into a new candidate set, accumulating the number of the same candidate objects in the obtained new candidate set, and taking the candidate object corresponding to the current threshold value as a final identification result if the proportion of the number of the same candidate objects in all the candidate objects is greater than a preset threshold value; otherwise, the user is judged to be a stranger, namely not belonging to the registered user.
7. The face recognition method of the face recognition device based on multi-angle video according to claim 1, characterized in that: in the step b5, the identification policy is: firstly, selecting candidate objects with the maximum similarity in each candidate set, then forming a new candidate set by the candidate objects, counting the number of the same candidate objects in the new candidate set, and if the maximum value in the obtained values is greater than or equal to 3, taking the candidate object corresponding to the value as a final recognition result; if the maximum value of the obtained values is less than 3 and greater than or equal to 2 and the second largest value is not 2, taking the candidate object corresponding to the value of 2 as the final recognition result; if the maximum value in the obtained values is less than 3 and greater than or equal to 2 and the second largest value is also 2, comparing the similarity sums of the obtained values, and taking the candidate object corresponding to the maximum value of the similarity sum as a final recognition result; and if the maximum value in the obtained values is less than 2, judging that the user is a stranger, namely judging that the user does not belong to the registered user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710851859.7A CN107480658B (en) | 2017-09-19 | 2017-09-19 | Face recognition device and method based on multi-angle video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710851859.7A CN107480658B (en) | 2017-09-19 | 2017-09-19 | Face recognition device and method based on multi-angle video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107480658A CN107480658A (en) | 2017-12-15 |
| CN107480658B true CN107480658B (en) | 2020-11-06 |
Family
ID=60585978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710851859.7A Active CN107480658B (en) | 2017-09-19 | 2017-09-19 | Face recognition device and method based on multi-angle video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107480658B (en) |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993028A (en) * | 2017-12-29 | 2019-07-09 | 技嘉科技股份有限公司 | Human face recognition device and method, the method for promoting image identification rate |
| CN107967609A (en) * | 2017-12-29 | 2018-04-27 | 深圳正品创想科技有限公司 | A kind of settlement channel and unmanned shop |
| CN108460332A (en) * | 2018-01-18 | 2018-08-28 | 新开普电子股份有限公司 | A kind of carrier-borne recognition of face safety protection device of naval's warship |
| CN108564053A (en) * | 2018-04-24 | 2018-09-21 | 南京邮电大学 | Multi-cam dynamic human face recognition system based on FaceNet and method |
| CN108564052A (en) * | 2018-04-24 | 2018-09-21 | 南京邮电大学 | Multi-cam dynamic human face recognition system based on MTCNN and method |
| CN108540817B (en) * | 2018-05-08 | 2021-04-20 | 成都市喜爱科技有限公司 | Video data processing method, device, server and computer readable storage medium |
| CN108960841A (en) | 2018-07-16 | 2018-12-07 | 阿里巴巴集团控股有限公司 | Method of payment, apparatus and system |
| CN109345253A (en) * | 2018-09-04 | 2019-02-15 | 阿里巴巴集团控股有限公司 | Resource transfers method, apparatus and system |
| CN109087429B (en) * | 2018-09-19 | 2020-12-04 | 重庆第二师范学院 | Method for checking consistency of library book-borrowing testimony of witness based on face recognition technology |
| CN109635635A (en) * | 2018-10-29 | 2019-04-16 | 安徽智传科技有限公司 | A kind of the face recognition accuracy rate improvement method and system of multi-angle |
| CN109376016A (en) * | 2018-10-29 | 2019-02-22 | 安徽智传科技有限公司 | A kind of the recognition of face efficiency improvement method and system of multithreading |
| CN109376686A (en) * | 2018-11-14 | 2019-02-22 | 睿云联(厦门)网络通讯技术有限公司 | A kind of various dimensions human face data acquisition scheme, acquisition system and acquisition method |
| CN109800643B (en) * | 2018-12-14 | 2023-03-31 | 天津大学 | Identity recognition method for living human face in multiple angles |
| CN109658572B (en) | 2018-12-21 | 2020-09-15 | 上海商汤智能科技有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN109740501A (en) * | 2018-12-28 | 2019-05-10 | 广东亿迅科技有限公司 | A kind of Work attendance method and device of recognition of face |
| CN110032955B (en) * | 2019-03-27 | 2020-12-25 | 深圳职业技术学院 | Novel face recognition method based on deep learning |
| CN109948594A (en) * | 2019-04-08 | 2019-06-28 | 银河水滴科技（北京）有限公司 | A kind of method and device of physical characteristics collecting |
| CN110443110B (en) * | 2019-06-11 | 2023-08-25 | 平安科技（深圳）有限公司 | Face recognition method, device, terminal and storage medium based on multipath camera shooting |
| CN110309805A (en) * | 2019-07-08 | 2019-10-08 | 业成科技（成都）有限公司 | Face recognition device |
| CN110675433A (en) | 2019-10-31 | 2020-01-10 | 北京达佳互联信息技术有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN113095116B (en) * | 2019-12-23 | 2024-03-22 | 深圳云天励飞技术有限公司 | Identity recognition method and related product |
| CN111144326B (en) * | 2019-12-28 | 2023-10-27 | 神思电子技术股份有限公司 | Human face anti-re-recognition method for man-machine interaction |
| CN113126506A (en) * | 2019-12-30 | 2021-07-16 | 佛山市云米电器科技有限公司 | Household appliance control method, control device and computer readable storage medium |
| CN112016508B (en) * | 2020-09-07 | 2023-08-29 | 杭州海康威视数字技术股份有限公司 | Face recognition method, device, system, computing device and storage medium |
| CN112613421B (en) * | 2020-12-26 | 2023-04-14 | 数源科技股份有限公司 | Dimension reduction feature analysis and comparison method for face picture |
| CN115273264A (en) * | 2022-08-09 | 2022-11-01 | 平安付科技服务有限公司 | Attendance system, method, storage medium and computer equipment |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529494A (en) * | 2016-11-24 | 2017-03-22 | 深圳市永达电子信息股份有限公司 | Human face recognition method based on multi-camera model |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100886557B1 (en) * | 2007-05-03 | 2009-03-02 | 삼성전자주식회사 | System and method for face recognition based on adaptive learning |
| CN101639891B (en) * | 2008-07-28 | 2012-05-02 | 汉王科技股份有限公司 | Double-camera face identification device and method |
| CN102332185A (en) * | 2011-08-17 | 2012-01-25 | 中国铁道科学研究院电子计算技术研究所 | L-type passage used for security inspection area face recognition |
| CN104301669A (en) * | 2014-09-12 | 2015-01-21 | 重庆大学 | Suspicious target detection tracking and recognition method based on dual-camera cooperation |
| CN104539848A (en) * | 2014-12-31 | 2015-04-22 | 深圳泰山在线科技有限公司 | Human face multi-pose collecting system |
-
2017
- 2017-09-19 CN CN201710851859.7A patent/CN107480658B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529494A (en) * | 2016-11-24 | 2017-03-22 | 深圳市永达电子信息股份有限公司 | Human face recognition method based on multi-camera model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107480658A (en) | 2017-12-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107480658B (en) | Face recognition device and method based on multi-angle video | |
| Mantoro et al. | Multi-faces recognition process using Haar cascades and eigenface methods | |
| CN102194131B (en) | Fast human face recognition method based on geometric proportion characteristic of five sense organs | |
| CN109800643B (en) | Identity recognition method for living human face in multiple angles | |
| Zhang et al. | Fast and robust occluded face detection in ATM surveillance | |
| CN109558810B (en) | Target person identification method based on part segmentation and fusion | |
| Tsao et al. | A data mining approach to face detection | |
| CN103678984A (en) | Method for achieving user authentication by utilizing camera | |
| CN109145742A (en) | A kind of pedestrian recognition method and system | |
| CN105825176A (en) | Identification method based on multi-mode non-contact identity characteristics | |
| US9449217B1 (en) | Image authentication | |
| CN101344914A (en) | Human face recognition method based on characteristic point | |
| Kim et al. | Eye detection in a facial image under pose variation based on multi-scale iris shape feature | |
| CN106874877A (en) | A kind of combination is local and global characteristics without constraint face verification method | |
| TW201917636A (en) | A method of face recognition based on online learning | |
| CN102880864A (en) | Method for snap-shooting human face from streaming media file | |
| Wiliem et al. | Detecting uncommon trajectories | |
| Moallem et al. | Fuzzy inference system optimized by genetic algorithm for robust face and pose detection | |
| CN107862298B (en) | Winking living body detection method based on infrared camera device | |
| Shanmugavadivu et al. | Rapid face detection and annotation with loosely face geometry | |
| CN113591692A (en) | Multi-view identity recognition method | |
| Li et al. | Disguised face detection and recognition under the complex background | |
| Paul et al. | Face recognition using eyes, nostrils and mouth features | |
| Yi et al. | Face detection method based on skin color segmentation and facial component localization | |
| Sankaran et al. | Pose angle determination by face, eyes and nose localization |
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 |