CN109961031A - Face fusion identifies identification, target person information display method, early warning supervision method and system - Google Patents
Face fusion identifies identification, target person information display method, early warning supervision method and system Download PDFInfo
- Publication number
- CN109961031A CN109961031A CN201910203860.8A CN201910203860A CN109961031A CN 109961031 A CN109961031 A CN 109961031A CN 201910203860 A CN201910203860 A CN 201910203860A CN 109961031 A CN109961031 A CN 109961031A
- Authority
- CN
- China
- Prior art keywords
- facial image
- identification
- identified
- face
- target person
- 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.)
- Pending
Links
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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Alarm Systems (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a kind of face fusions to identify identification, and step includes: that facial image to be identified is sent to a wherein algorithm service quotient, first time identification is carried out, if facial image to be identified carries out next step by identification；Facial image to be identified is sent at least other algorithm services quotient and carries out second of identification, if carrying out next step if facial image to be identified is identified by any family other algorithm services quotient；If any family other algorithm services quotient in the algorithm service quotient in identification and second of identification for the first time, if facial image to be identified is identified as same target person, then regards as corresponding target person for facial image to be identified.The face fusion recognition identification not only merges the algorithm advantage for having each algorithm service quotient, improves the precision of recognition of face, reduces rate of false alarm, advantage also few with operand.
Description
Technical field
The present invention relates to field of face identification more particularly to a kind of face fusion to identify identification, target person information
Display methods, early warning supervision method and system.
Background technique
Public security organ carry out early warning deploy to ensure effective monitoring and control of illegal activities when, need to shoot the stream of people by monitoring camera, with obtain to
It identifies facial image, recognition of face then is carried out to the target person in facial image to be identified and personnel library, it finally will identification
Successful target person is pushed to police on duty, is carried out live confirmation by police on duty or is arrested, but adopts
It often will appear more wrong report with the face recognition algorithms of single algorithm service quotient, and the face of algorithms of different service provider is known
Other algorithm has respectively different algorithm advantages.
A kind of suspect's recognition of face identification side is disclosed in the patent document of a Publication No. CN109002749A
Method, by the way that face picture push is distributed to each family " face alignment identification engine ", by least two, " face alignment identification is drawn
Hold up " independent matching identification is carried out to same face picture, the then matching identification according to each family " face alignment identification engine " again
As a result hit rate is counted, suspect is ranked up from high to low according to hit rate, exports suspicion list List.This side
Although method can make full use of the advantage of each family " face alignment identification engine ", the precision of recognition of face is improved, reduces wrong report
Rate, but operand is too big, each family's " face alignment identification engine " needs all to carry out every face picture individually to compare to know
Not, too many resource is occupied, especially China personnel are numerous, and there are thousands of flow of the people in some public places daily.
Summary of the invention
In order to solve above-mentioned the deficiencies in the prior art, the present invention provides a kind of face fusion identification identification, not only
Fusion has the algorithm advantage of each algorithm service quotient, improves the precision of recognition of face, reduces rate of false alarm, also has operand few
Advantage.
The present invention also provides a kind of target person information display method and early warning supervision methods.
The technical problems to be solved by the invention are achieved by the following technical programs:
A kind of face fusion identification identification, step include:
Facial image to be identified is sent to a wherein algorithm service quotient, first time identification is carried out, if facial image to be identified
By identification, then next step is carried out；
Facial image to be identified is sent at least other algorithm services quotient and carries out second of identification, if face figure to be identified
If identifying by any family other algorithm services quotient, then next step is carried out；
If any family other algorithm services quotient in the algorithm service quotient in identification and second of identification for the first time, will be to be identified
If facial image is identified as same target person, then facial image to be identified is regarded as into corresponding target person.
Further, the step of recognition of face includes:
The facial image of the target person stored in facial image to be identified and face database is successively compared, face is calculated
Similarity；
If calculated human face similarity degree is greater than preset similarity threshold, by identification, and the target by identification is exported
Personnel.
Further, before carrying out first time identification, further includes: preset to the similarity threshold of recognition of face.
Further, recognition of face is carried out using different face recognition algorithms between each algorithm service quotient.
Further, after by identifying for the first time, by facial image to be identified and pass through the target person of first time identification
The facial image of member is sent at least other algorithm services quotient together；When carrying out second and identifying, at least one other
Only facial image to be identified is successively compared with the facial image by the target person of identification for the first time by algorithm service quotient.
A kind of target person information display method, step include:
Using above-mentioned face fusion recognizer, by same facial image to be identified respectively with the face figure of pre-recording in personnel library
Picture and the facial image of identification captured in library are individually assert；
The identity information of the target person by identification is transferred in the personnel library, and, it transfers in capturing library by recognizing
The location information of fixed target person；
It will be display together by the identity information and location information for the target person assert.
Further, the personnel Ku Nei stores the facial image of pre-recording of the target person of preparatory typing, face of pre-recording
Image is corresponding with its identity information；It is stored in the candid photograph library in the personnel library through the knowledge for the target person assert
Others' face image, has identified that facial image is corresponding with its location information.
A kind of early warning supervision method, step include:
Creation and starting are deployed to ensure effective monitoring and control of illegal activities task；
According to the task of deploying to ensure effective monitoring and control of illegal activities of starting, dispatches corresponding monitoring camera and shot, to obtain facial image to be identified；
Using above-mentioned face fusion recognizer, facial image to be identified is carried out to face fusion identification in personnel library and is recognized
It is fixed；
According to the task of deploying to ensure effective monitoring and control of illegal activities of starting, early warning will be pushed to corresponding police on duty by the target person assert.
Further, using above-mentioned target person information display method to the identity information of the target person by assert
It is display together with location information.
A kind of system, comprising:
The front-end monitoring module, for obtaining facial image to be identified, including at least one monitoring camera；
Merge Identification of Images platform, for carrying out recognition of face, including at least two algorithm service quotient, each algorithm service quotient it
Between using different face recognition algorithms carry out recognition of face；
Rear end administrative center, for being accessed for the front-end monitoring module and fusion Identification of Images platform, to dispatch each monitoring
Camera and each algorithm service quotient carry out above-mentioned face fusion recognition methods, above-mentioned target person information display method or
Above-mentioned early warning supervision method.
The invention has the following beneficial effects: the face fusion recognition identification by more algorithm service quotient fusions one
It rises and forms the fusion Identification of Images platform, not only merged the algorithm advantage of each algorithm service quotient, by will face twice
Recognition result is collided, and facial image to be identified is only identified as same mesh in first time identification and second of identification
Mark personnel just calculate the precision assert and successfully improve recognition of face, and precision can reach 99.99% or more, reduce wrong report
Rate, and only after facial image to be identified by identifying for the first time, second just is carried out to facial image to be identified and is known
Not, reduce a large amount of unnecessary identification operations, with the few advantage of operand.
Detailed description of the invention
Fig. 1 is the functional block diagram of system provided by the invention；
Fig. 2 is the step block diagram that face fusion provided by the invention identifies identification；
Fig. 3 is the step block diagram of target person information display method provided by the invention；
Fig. 4 is the step block diagram of early warning supervision method provided by the invention.
Specific embodiment
The present invention will be described in detail with reference to the accompanying drawings and examples.
Embodiment one
As shown in Figure 1, a kind of system, comprising:
The front-end monitoring module, for obtaining facial image to be identified, including at least one monitoring camera；
Merge Identification of Images platform, for carrying out recognition of face, including at least two algorithm service quotient, each algorithm service quotient it
Between using different face recognition algorithms carry out recognition of face；
Face memory module, for storing the facial image of target person, for being compared with facial image to be identified, including
At least one face database；
Rear end administrative center, for being accessed for the front-end monitoring module, fusion Identification of Images platform, personnel library and candid photograph library,
To dispatch each monitoring camera, each algorithm service quotient and each face database.
Wherein, each face database in the face memory module is divided by the source of facial image:
An at least personnel library, for storing the facial image of pre-recording of preparatory typing, facial image of pre-recording is corresponding with its (target person
Member) identity information；
One captures library, for being stored in the personnel library through the identification facial image assert, has identified facial image pair
Ying Youqi (shooting) location information.
The facial image of pre-recording stored in the personnel library is the earliest source of the facial image of target person, by public security machine
Put the preparatory typing of row into.
Embodiment two
A kind of face fusion identifies identification, applied in system described in embodiment one, as shown in Fig. 2, step includes:
Facial image to be identified is sent to a wherein algorithm service quotient, first time identification is carried out, if facial image to be identified
By identification, then next step is carried out；
Facial image to be identified is sent at least other algorithm services quotient and carries out second of identification, if face figure to be identified
If identifying by any family other algorithm services quotient, then next step is carried out；
If any family other algorithm services quotient in the algorithm service quotient in identification and second of identification for the first time, will be to be identified
If facial image is identified as same target person, then facial image to be identified is regarded as into corresponding target person.
More algorithm service quotient are fused together to form the fusion Identification of Images by the face fusion recognition identification
Platform has not only merged the algorithm advantage of each algorithm service quotient, by that face recognition result will collide twice, only exists
It facial image to be identified is identified as same target person in identification for the first time and second identification just calculates to assert successfully, raising
The precision of recognition of face, precision can reach 99.99% or more, reduce rate of false alarm, and only in face figure to be identified
After by identifying for the first time, second just is carried out to facial image to be identified and is identified, reduces a large amount of unnecessary identifications
Operation, with the few advantage of operand.
When in use, at least one monitoring camera in the front-end monitoring module described in rear end management central dispatching into
Row shooting, to obtain facial image to be identified；When identify for the first time to facial image to be identified, the step of recognition of face
Include:
The facial image of the target person stored in facial image to be identified and the face database (is pre-recorded and facial image or known
Others' face image) successively it is compared, calculate human face similarity degree；
If calculated human face similarity degree is greater than preset similarity threshold, by identification, and the target by identification is exported
Personnel.
Therefore, before carrying out first time identification, further includes: preset to the similarity threshold of recognition of face；When pre-
If similarity threshold it is larger when, then identification accuracy is higher, identifies that the quantity of the target person of output is fewer, when preset phase
When smaller like degree threshold value, then identification accuracy is lower, identifies that the quantity of the target person of output is more.
When carrying out second of identification, other algorithm services quotient can be as when identifying, by face to be identified first time
The facial image of the target person stored in image and the face database is successively compared, it is preferable that at least one other calculation
Method service provider only can also successively carry out facial image to be identified with the facial image by the target person of identification for the first time
It compares, then after by identifying for the first time, the rear end administrative center is by facial image to be identified and passes through first time identification
The facial image of target person is sent at least other algorithm services quotient together.
Output when the rear end administrative center identifies according to the target person list exported when identifying for the first time and second
Target person list counted, by hit-count n be higher than preset value (n >=2) target person by hit-count by up to
It is low to be ranked up.
It preferably, will be corresponding by the target person assert after the facial image for the target person for clicking through identification
Hit-count and the algorithm service quotient that recognizes check, so that the recognizer to each algorithm service quotient carries out laterally
Compare.
The face fusion recognition identification further include: after assert to facial image to be identified, if people to be identified
Face image be in the personnel library by assert if, then by facial image to be identified it is corresponding with its (shooting) location information after
Formation has identified facial image, and is stored into the candid photograph library.
Embodiment three
A kind of target person information display method, applied in system described in embodiment one, as shown in figure 3, step includes:
Using face fusion recognizer described in embodiment two, by same facial image to be identified respectively with the personnel library
In pre-record facial image and it is described capture library in the facial image of identification individually assert；
The identity information of the target person by identification is transferred in the personnel library, and, it transfers in capturing library by recognizing
The location information of fixed target person；
It will be display together by the identity information and location information for the target person assert.
The target person information display method in the personnel library and captures same band identification facial image in library simultaneously
Individually assert, then will be display together by the identity information and location information for the target person assert, to obtain simultaneously
The identity information and location information of target person.
Preferably, after the facial image for the target person for clicking through identification, the every of the target person of identification will be passed through
A location information forms mark point, is marked to form motion track on map, and in the mark point of each location information
When identifying the frequency of occurrence of target person, and distinguishing using the mark point of the different colours depth appearance of different location information
Between.
Embodiment three
A kind of early warning supervision method, applied in system described in embodiment one, as shown in figure 4, step includes:
Creation and starting are deployed to ensure effective monitoring and control of illegal activities task；
In this step, administrative staff are created by the rear end administrative center and starting is deployed to ensure effective monitoring and control of illegal activities task, include in task of deploying to ensure effective monitoring and control of illegal activities
But be not limited to creation deploy to ensure effective monitoring and control of illegal activities task when selected territorial scope, Ren Yuanku, the time cycle, recognition of face similarity threshold
The information of deploying to ensure effective monitoring and control of illegal activities such as the police on duty of value, task grade and selection push.
According to the task of deploying to ensure effective monitoring and control of illegal activities of starting, dispatches corresponding monitoring camera and shot, to obtain facial image to be identified；
In this step, the rear end administrative center can according to the territorial scope selected in the task of deploying to ensure effective monitoring and control of illegal activities, to cover by
Multiple monitoring cameras of the territorial scope of selection are scheduled, the multiple prisons selected in the task that can also be deployed to ensure effective monitoring and control of illegal activities with direct basis
Control camera is scheduled.
Using face fusion recognizer described in embodiment two, by facial image to be identified in the personnel library into
Pedestrian's face fusion recognition identification；
In this step, at least two algorithm service quotient that the rear end administrative center is selected according to scheduling in task of deploying to ensure effective monitoring and control of illegal activities incite somebody to action
Facial image to be identified and the facial image of pre-recording of the target person in the personnel library selected are assert.
According to the task of deploying to ensure effective monitoring and control of illegal activities of starting, early warning will be pushed to corresponding police on duty by the target person assert；
In this step, the rear end administrative center will be sent to by the target person assert by network according to task of deploying to ensure effective monitoring and control of illegal activities
On the law-enforcing recorder of corresponding police on duty.
The early warning supervision method further include: using information display method described in embodiment three to the target by assert
The identity information and location information of personnel is display together.
Embodiments of the present invention above described embodiment only expresses, the description thereof is more specific and detailed, but can not
Therefore limitations on the scope of the patent of the present invention are interpreted as, as long as skill obtained in the form of equivalent substitutions or equivalent transformations
Art scheme should all be fallen within the scope and spirit of the invention.
Claims (10)
1. a kind of face fusion identifies identification, which is characterized in that step includes:
Facial image to be identified is sent to a wherein algorithm service quotient, first time identification is carried out, if facial image to be identified
By identification, then next step is carried out；
Facial image to be identified is sent at least other algorithm services quotient and carries out second of identification, if face figure to be identified
If identifying by any family other algorithm services quotient, then next step is carried out；
If any family other algorithm services quotient in the algorithm service quotient in identification and second of identification for the first time, will be to be identified
If facial image is identified as same target person, then facial image to be identified is regarded as into corresponding target person.
2. face fusion according to claim 1 identifies identification, which is characterized in that the step of recognition of face includes:
The facial image of the target person stored in facial image to be identified and face database is successively compared, face is calculated
Similarity；
If calculated human face similarity degree is greater than preset similarity threshold, by identification, and the target by identification is exported
Personnel.
3. face fusion according to claim 2 identifies identification, which is characterized in that carrying out identifying it for the first time
Before, further includes: the similarity threshold of recognition of face is preset.
4. face fusion recognition methods according to claim 1 to 3, which is characterized in that each algorithm service quotient it
Between using different face recognition algorithms carry out recognition of face.
5. face fusion recognition methods according to claim 1 to 3, which is characterized in that by identifying for the first time
Afterwards, by facial image to be identified and by first time identification target person facial image be sent to together at least one other
Algorithm service quotient；When carrying out second and identifying, at least other algorithm services quotient only by facial image to be identified with pass through
The facial image of the target person of identification is successively compared for the first time.
6. a kind of target person information display method, which is characterized in that step includes:
Using face fusion recognizer as claimed in any one of claims 1 to 5, by same facial image to be identified respectively with people
The facial image of identification in pre-record facial image and candid photograph library in member library is individually assert；
The identity information of the target person by identification is transferred in the personnel library, and, it transfers in capturing library by recognizing
The location information of fixed target person；
It will be display together by the identity information and location information for the target person assert.
7. target person information display method according to claim 6, which is characterized in that the personnel Ku Nei stores pre-
The facial image of pre-recording of the target person of first typing, facial image of pre-recording are corresponding with its identity information；It is stored in the candid photograph library
Have in the personnel library through the identification facial image for the target person assert, has identified that facial image is corresponding with its position
Information.
8. a kind of early warning supervision method, which is characterized in that step includes:
Creation and starting are deployed to ensure effective monitoring and control of illegal activities task；
According to the task of deploying to ensure effective monitoring and control of illegal activities of starting, dispatches corresponding monitoring camera and shot, to obtain facial image to be identified；
Using face fusion recognizer as claimed in any one of claims 1 to 5, by facial image to be identified in personnel library into
Pedestrian's face fusion recognition identification；
According to the task of deploying to ensure effective monitoring and control of illegal activities of starting, early warning will be pushed to corresponding police on duty by the target person assert.
9. early warning supervision method according to claim 8, which is characterized in that using target described in claim 6 or 7
Personal information display methods is display together to by the identity information and location information for the target person assert.
10. a kind of system characterized by comprising
The front-end monitoring module, for obtaining facial image to be identified, including at least one monitoring camera；
Merge Identification of Images platform, for carrying out recognition of face, including at least two algorithm service quotient, each algorithm service quotient it
Between using different face recognition algorithms carry out recognition of face；
Rear end administrative center, for being accessed for the front-end monitoring module and fusion Identification of Images platform, to dispatch each monitoring
Camera and each algorithm service quotient carry out face fusion recognition methods as claimed in any one of claims 1 to 5, claim 6 or
Early warning supervision method described in target person information display method described in 7 or claim 8 or 9.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2019100754088 | 2019-01-25 | ||
| CN201910075408 | 2019-01-25 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109961031A true CN109961031A (en) | 2019-07-02 |
Family
ID=67024324
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910203860.8A Pending CN109961031A (en) | 2019-01-25 | 2019-03-18 | Face fusion identifies identification, target person information display method, early warning supervision method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109961031A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110519556A (en) * | 2019-07-09 | 2019-11-29 | 杭州亿圣信息技术有限公司 | A kind of method that electricity encloses base station Yu video detection target association |
| CN110688892A (en) * | 2019-08-20 | 2020-01-14 | 武汉烽火众智数字技术有限责任公司 | Portrait identification alarm method and system based on data fusion technology |
| CN111968152A (en) * | 2020-07-15 | 2020-11-20 | 桂林远望智能通信科技有限公司 | Dynamic identity recognition method and device |
| CN112699752A (en) * | 2020-12-22 | 2021-04-23 | 安徽配隆天环保科技有限公司 | Automatic tracking evidence obtaining system for unmanned aerial vehicle feature marker |
| CN113435295A (en) * | 2021-06-23 | 2021-09-24 | 平安养老保险股份有限公司 | Face recognition aggregation method, device, equipment and medium |
| CN114049659A (en) * | 2021-09-24 | 2022-02-15 | 公安部第一研究所 | 1-to-N face recognition engine multi-algorithm fusion method |
| CN116524569A (en) * | 2023-05-10 | 2023-08-01 | 深圳大器时代科技有限公司 | Multi-concurrency face recognition system and method based on classification algorithm |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101901340A (en) * | 2010-08-04 | 2010-12-01 | 惠州市华阳多媒体电子有限公司 | Suspect tracking method and system |
| CN101931789A (en) * | 2009-06-26 | 2010-12-29 | 上海宝康电子控制工程有限公司 | High-resolution human figure automatic recording and comparing system and method in key region |
| CN105184267A (en) * | 2015-09-15 | 2015-12-23 | 重庆智韬信息技术中心 | Face-identification-based secondary-deformation auxiliary authorization method |
| CN108875466A (en) * | 2017-06-01 | 2018-11-23 | 北京旷视科技有限公司 | Monitoring method, monitoring system and storage medium based on recognition of face |
| CN109002749A (en) * | 2017-12-11 | 2018-12-14 | 罗普特（厦门）科技集团有限公司 | Suspect's recognition of face identification |
| CN109117808A (en) * | 2018-08-24 | 2019-01-01 | 深圳前海达闼云端智能科技有限公司 | Face recognition method and device, electronic equipment and computer readable medium |
-
2019
- 2019-03-18 CN CN201910203860.8A patent/CN109961031A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101931789A (en) * | 2009-06-26 | 2010-12-29 | 上海宝康电子控制工程有限公司 | High-resolution human figure automatic recording and comparing system and method in key region |
| CN101901340A (en) * | 2010-08-04 | 2010-12-01 | 惠州市华阳多媒体电子有限公司 | Suspect tracking method and system |
| CN105184267A (en) * | 2015-09-15 | 2015-12-23 | 重庆智韬信息技术中心 | Face-identification-based secondary-deformation auxiliary authorization method |
| CN108875466A (en) * | 2017-06-01 | 2018-11-23 | 北京旷视科技有限公司 | Monitoring method, monitoring system and storage medium based on recognition of face |
| CN109002749A (en) * | 2017-12-11 | 2018-12-14 | 罗普特（厦门）科技集团有限公司 | Suspect's recognition of face identification |
| CN109117808A (en) * | 2018-08-24 | 2019-01-01 | 深圳前海达闼云端智能科技有限公司 | Face recognition method and device, electronic equipment and computer readable medium |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110519556A (en) * | 2019-07-09 | 2019-11-29 | 杭州亿圣信息技术有限公司 | A kind of method that electricity encloses base station Yu video detection target association |
| CN110688892A (en) * | 2019-08-20 | 2020-01-14 | 武汉烽火众智数字技术有限责任公司 | Portrait identification alarm method and system based on data fusion technology |
| CN111968152A (en) * | 2020-07-15 | 2020-11-20 | 桂林远望智能通信科技有限公司 | Dynamic identity recognition method and device |
| CN111968152B (en) * | 2020-07-15 | 2023-10-17 | 桂林远望智能通信科技有限公司 | Dynamic identity recognition method and device |
| CN112699752A (en) * | 2020-12-22 | 2021-04-23 | 安徽配隆天环保科技有限公司 | Automatic tracking evidence obtaining system for unmanned aerial vehicle feature marker |
| CN113435295A (en) * | 2021-06-23 | 2021-09-24 | 平安养老保险股份有限公司 | Face recognition aggregation method, device, equipment and medium |
| CN113435295B (en) * | 2021-06-23 | 2024-05-10 | 平安养老保险股份有限公司 | Face recognition aggregation method, device, equipment and medium |
| CN114049659A (en) * | 2021-09-24 | 2022-02-15 | 公安部第一研究所 | 1-to-N face recognition engine multi-algorithm fusion method |
| CN114049659B (en) * | 2021-09-24 | 2024-04-19 | 公安部第一研究所 | Multi-algorithm fusion method for 1-to-N face recognition engine |
| CN116524569A (en) * | 2023-05-10 | 2023-08-01 | 深圳大器时代科技有限公司 | Multi-concurrency face recognition system and method based on classification algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109961031A (en) | Face fusion identifies identification, target person information display method, early warning supervision method and system | |
| US11704936B2 (en) | Object tracking and best shot detection system | |
| CN110222640B (en) | Method, device and method for identifying suspect in monitoring site and storage medium | |
| US8379917B2 (en) | Face recognition performance using additional image features | |
| CN110969118B (en) | Track monitoring system and method | |
| CN105631430A (en) | Matching method and apparatus for face image | |
| CN110263613A (en) | Monitor video processing method and processing device | |
| CN107657232B (en) | Pedestrian intelligent identification method and system | |
| CN109522853A (en) | Face datection and searching method towards monitor video | |
| CN106204948A (en) | Locker management method and locker managing device | |
| CN106951846A (en) | A kind of face 3D models typing and recognition methods and device | |
| CN110543868A (en) | Monitoring method and system based on face recognition and head and shoulder detection | |
| CN106326835A (en) | Human face data collection statistical system and method for gas station convenience store | |
| WO2022134916A1 (en) | Identity feature generation method and device, and storage medium | |
| CN110599129A (en) | Campus attendance checking method, device, identification terminal and system based on image tracking | |
| CN111191506A (en) | Personnel flow statistical method and device, computer equipment and storage medium | |
| CN112699328A (en) | Network point service data processing method, device, system, equipment and storage medium | |
| CN107967743A (en) | A kind of personal identification method being applied in e-bidding and system | |
| WO2021129491A1 (en) | Pedestrian search method, server, and storage medium | |
| CN110866418A (en) | Image base generation method, device, equipment, system and storage medium | |
| CN110443134B (en) | Face recognition tracking system based on video stream and working method | |
| JP2010067008A (en) | Imaging management system, imaging management method, authentication system, and authentication method | |
| CN116886874A (en) | Ecological garden security monitoring and early warning data acquisition method and system | |
| CN111832451A (en) | Airworthiness monitoring process supervision system and method based on video data processing | |
| KR100911121B1 (en) | Image-based human blob tracking 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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190702 |