CN109255303A - A kind of face identification system and method for the dynamic stream of people - Google Patents
A kind of face identification system and method for the dynamic stream of people Download PDFInfo
- Publication number
- CN109255303A CN109255303A CN201810936330.XA CN201810936330A CN109255303A CN 109255303 A CN109255303 A CN 109255303A CN 201810936330 A CN201810936330 A CN 201810936330A CN 109255303 A CN109255303 A CN 109255303A
- Authority
- CN
- China
- Prior art keywords
- face
- people
- subchannel
- video image
- safety check
- 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
- G06V40/172—Classification, e.g. identification
-
- 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)
- Image Processing (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses the face identification systems and method of a kind of dynamic stream of people, the invention belongs to technical field of face recognition, the face identification system includes: lane identification system, for tentatively being identified by channel face recognition algorithms to the people by subchannel and judging whether it is questionable person;The lane identification system includes several subchannel identifying systems being arranged in several subchannels, and several subchannel identifying systems are provided in each subchannel, and the subchannel setting is being entered the station between channel and waiting room.When channel recognition be unsuspicious people, be directly entered waiting room, questionable person then imports the information (fingerprint, voice) that safety check room is identified in detail again and can be collected using public security system and assists in identifying.Technical solution disclosed by the invention can effectively solve the technical issues of can not carrying out effective recognition of face in a dynamic state existing in the prior art, improve discrimination, and can be used in public places.
Description
Technical field
The present invention relates to technical field of face recognition, and in particular to a kind of face identification system and method for the dynamic stream of people.
Background technique
Identification is the facial video image of certain the face figure and current someone that are stored in human brain or computer to original
Identification comparison is carried out, so that judging whether is the same person.Face is three-dimensional, i.e. 3-D image.But the acquisition of 3-D image,
Presently, there are equipment volume is big, resolution ratio is not high, price, so now all with three-dimensional image projection at two dimensional image, then
It is handled with the mathematical model of advanced complexity.The dominant features of face are eye, nose, mouth and shape of face.The stream of people rolls on street
Rolling, but be identical without a face.
So with the development of technology, the appearance of face recognition technology is both provided in a lot of public regions for people
The camera of face identification acquisition, compares identification for the photo of the photo of acquisition and reference, but will be in dynamic stream of people's condition
Under compare identification, there is problems:
1, passenger often occurs crowded, blocks mutually in channel is advanced, it cannot be guaranteed that it is complete just to obtain everyone
Dough figurine face image can not just compare identification.Outstanding mathematical algorithm again also can't resolve the identification after seriously blocking and ask
Topic.
2, people are in traveling, it cannot be guaranteed that certain camera distance, having far has closely, and image varies.Some mathematics
Method can correct, but can lose accuracy rate.
3, many people advance in the channel, it cannot be guaranteed that the consistency of the radiation direction of each face and illumination, it can be serious
Influence recognition of face accuracy rate, some mathematical methods can correct but not enough thoroughly.
4, posture of the people in traveling, expression is sitting on seat with people will not be identical.
5, passenger cannot be required to remove glasses and jewelry in traveling, they also will affect the accuracy rate of identification.
6, for people in traveling, body has Plane Rotation and depth (pitching) rotation, present Plane Rotation, some algorithms
It can correct, but pitching cannot still be corrected.
So being badly in need of the new technical solution of one kind to solve the above problems.
Summary of the invention
The purpose of the present invention is to provide the face identification systems and method of a kind of dynamic stream of people, to solve the prior art
Present in can not carry out in a dynamic state the technical issues of effective recognition of face.
To achieve the above object, first aspect of the embodiment of the present invention provides the face identification system of dynamic stream of people a kind of,
The face identification system includes: lane identification system, for by channel face recognition algorithms to by the people of subchannel into
Row tentatively identifies and judges whether it is questionable person;The lane identification system includes several sons being arranged in several subchannels
Lane identification system, is provided with several subchannel identifying systems in each subchannel, the subchannel setting enter the station channel with
Between waiting room.
Preferably, the subchannel identifying system includes: camera, for acquiring the first face for passing through the people of subchannel
Static video image;First display screen, for showing the first face static state video image of camera acquisition;And first service
Device, for carrying out preliminary recognition of face according to the first face static state video image of acquisition;Wherein, it is set in the first server
Channel face recognition algorithms are equipped with, the first server is networked by the network switch and safety check identifying system, and described first
The crime that server is issued the first face static state video image of acquisition and safety check identifying system by channel face recognition algorithms
Violate photo and/or video image comparison identifies whether as questionable person.
Preferably, the subchannel identifying system further include: light compensating lamp, for acquiring the first face static state view in camera
Light filling is carried out when frequency image;The camera, the first server, first display screen and the light compensating lamp are integrated into touching
Touch all-in-one machine.
Preferably, the subchannel identifying system further include: identify seat, be equipped with an identification in front of each camera
Seat.
Preferably, the subchannel is three, and ten subchannel identifying systems are arranged in each subchannel.
Preferably, the face identification system further include: safety check identifying system, by safety check face recognition algorithms to through logical
Road face recognition algorithms are judged as that questionable person is identified in detail.
Preferably, the safety check identifying system includes: video camera, for acquiring the second face static state video image;Second
Display screen, for showing the first face static state video image or the second face static state video image;Third display screen, for showing
Criminal's photo and/or video image;And second server, for static according to the first face static state video image or the second face
Video image carries out detailed recognition of face;Wherein, it is provided with safety check face recognition algorithms in the second server, described second
Server is networked with several first servers and public security system respectively by the network switch, and each first server will be through channel
Face recognition algorithms are judged as that the first face static state video image of questionable person is sent to the second server, the public security
The criminal's photo and/or video image grasped are sent to the second server by system, and the second server passes through peace
Face recognition algorithms are examined by the first face static state video image or the second face static state video image and criminal's photo and/or video
Image comparison identifies whether as questionable person.
Preferably, the safety check identifying system further include: the 4th display screen and the 5th display screen, the 4th display screen are set
It sets on waiting room doorway, for showing the image for being identified as non-suspicious people through each subchannel identifying system, the 5th display
Screen is arranged on safety check room doorway, for showing the image for being identified as suspicious people through each subchannel identifying system.
Another aspect of embodiment disclosed by the invention, provides the face identification method of dynamic stream of people a kind of,
It is characterized in that, comprising: the dynamic stream of people enters several subchannels by the corridor diversion that enters the station;Into the dynamic stream of people in each subchannel
It is tentatively identified in batches by several subchannel identifying systems and judges whether it is questionable person;If it is questionable person,
Then it is sent into safety check room;And if not being questionable person, enter waiting room;Wherein, several subchannel identifying systems pass through
The criminal's photo and/or video that channel face recognition algorithms issue the first face static state video image of acquisition and public security system
Image comparison identifies whether as questionable person.
Preferably, the questionable person for being sent into the safety check room carries out detailed recognition of face, institute through safety check identifying system again
The second server for stating safety check identifying system passes through safety check face recognition algorithms for the first face static state video image or the second people
Face static state video image and the comparison of criminal's photo and/or video image identify whether as questionable person.
The present invention has the advantage that
Technical solution disclosed by the invention can effectively solve existing in the prior art can not carry out effective people in a dynamic state
The technical issues of face identifies, improves discrimination, and can be used in public places.
Detailed description of the invention
Fig. 1 is a kind of total block architecture diagram of the face identification system of the dynamic stream of people disclosed by the embodiments of the present invention;
Fig. 2 is a kind of subchannel layout of the face identification system of the dynamic stream of people disclosed by the embodiments of the present invention;
Fig. 3 is a kind of setting for subchannel identifying system of the face identification system of the dynamic stream of people disclosed by the embodiments of the present invention
Standby configuration diagram;
Fig. 4 is a kind of equipment of the safety check identifying system of the face identification system of the dynamic stream of people disclosed by the embodiments of the present invention
Configuration diagram;
Fig. 5 is a kind of flow chart of the face identification method of dynamic stream of people of the invention.
Specific embodiment
The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention..
It should be noted that description and claims of this specification and term " first " in above-mentioned attached drawing, "
Two " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein can in addition to illustrating herein or
Sequence other than those of description is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that cover
Cover it is non-exclusive include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to
Step or unit those of is clearly listed, but may include be not clearly listed or for these process, methods, product
Or other step or units that equipment is intrinsic.
Embodiment 1
With reference to Fig. 1, Fig. 1 is a kind of total framework frame of the face identification system of the dynamic stream of people disclosed by the embodiments of the present invention
Figure.A kind of face identification system the embodiment of the invention discloses dynamic stream of people includes: lane identification system 1, for by logical
Road face recognition algorithms tentatively identify the people by subchannel and judge whether it is questionable person;And safety check identifying system
2, it is judged as that questionable person is identified in detail to through channel face recognition algorithms by safety check face recognition algorithms.
Further, lane identification system 1 includes several subchannel identifying systems 11 being arranged in several subchannels 4,
With reference to Fig. 2, Fig. 2 is to enter the station channel layout schematic diagram, in the embodiment of the present invention, is provided between channel 5 and waiting room 6 entering the station
Three subchannels 4, waiting room 6 are located at the side of subchannel 4, the one end of waiting room 6 and each subchannel 4 far from channel 5 of entering the station
Connection, the other side of subchannel 4 are provided with safety check room 7, and safety check room 7 also connects with each subchannel 4 far from the one end in channel 5 of entering the station
It is logical, five pairs of isolated areas 8 for being symmetrically distributed in two sides in subchannel are distributed in each subchannel 4, are arranged in each subchannel 4
There are several subchannel identifying systems 11, in the embodiment of the present invention, ten sub- lane identification systems are provided in each subchannel 4
11, that is, a sub- lane identification system 11 is arranged in each isolated area 8.
With reference to Fig. 3, Fig. 3 is a kind of subchannel identification of face identification system of the dynamic stream of people disclosed by the embodiments of the present invention
The layout of equipment of system, subchannel identifying system 11 include: camera 111, for acquiring the first of the people for passing through subchannel 4
Face static state video image;First display screen 112, the first face static state video image acquired for showing camera 111;With
First server 113, for carrying out preliminary recognition of face according to the first face static state video image of acquisition;Wherein, the first clothes
Channel face recognition algorithms are provided in business device 113, first server 113 passes through the network switch 9 and safety check identifying system 2
First face static state video image of acquisition and safety check identification are by net, the first server by channel face recognition algorithms
Criminal's photo and/or the video image comparison that system 2 issues identify whether as questionable person.
Preferably, subchannel identifying system 11 further include: light compensating lamp 114, it is quiet for acquiring the first face in camera 111
Light filling is carried out when state video image, in addition, the first display screen 112 preferably has touch function, can be led to by the people of subchannel
It crosses touch function and the acquisition that camera 111 completes the first face static state video image is interacted and controlled with first server 113, the
One server 113 is connected with keyboard 115 and mouse 116, and keyboard 115 and mouse 116 are for manipulating first server 113;
Preferably, camera 111, the first display screen 112, first server 113 and light compensating lamp 114 are integrated into touch one-piece machine.It is preferred that
Ground, subchannel identifying system 11 further include: identify seat, be equipped with an identification seat in front of each camera and (do not show in figure
Out).
With reference to Fig. 4, Fig. 4 is that a kind of safety check of face identification system of the dynamic stream of people disclosed by the embodiments of the present invention identifies system
The layout of equipment of system, safety check identifying system 2 include: video camera 21, for acquiring the second face static state video image;Second is aobvious
Display screen 22, for showing the first face static state video image or the second face static state video image;Third display screen 23, for showing
Show criminal's photo and/or video image;With second server 24, it is used for according to the first face static state video image or the second face
Static video image carries out detailed recognition of face;Wherein, safety check face recognition algorithms, the second clothes are provided in second server 24
Business device 24 is networked with several first servers 113 and public security system 3 respectively by the network switch 9, each first server 113
It will be judged as that the first face static state video image of questionable person is sent to second server 24 through channel face recognition algorithms, it is public
The criminal's photo and/or video image grasped are sent to second server 24 by peace system 3, and second server 24 passes through safety check
Face recognition algorithms are by the first face static state video image or the second face static state video image and criminal's photo and/or video figure
As comparison identifies whether as questionable person.
Preferably, safety check identifying system 2 further include: the 4th display screen 25 and the 5th display screen 26, the 4th display screen 25 are set
It sets on waiting room doorway, for showing the image for being identified as non-suspicious people through each subchannel identifying system, the 5th display screen 26
It is arranged on safety check room doorway, for showing the image for being identified as suspicious people through each subchannel identifying system.In addition, of the invention
Ups power 27 is used to power in embodiment for each equipment of safety check identifying system 2.
With reference to Fig. 5, Fig. 5 is a kind of flow chart of the face identification method of the dynamic stream of people disclosed by the embodiments of the present invention.This
A kind of face identification method of the dynamic stream of people disclosed in inventive embodiments include: the dynamic stream of people entered by the corridor diversion that enters the station it is several
Subchannel;It is tentatively identified and is sentenced in batches by several subchannel identifying systems into the dynamic stream of people in each subchannel
Whether disconnected is questionable person;If it is questionable person, it is sent into safety check room;And if not being questionable person, enter waiting room;
Wherein, several subchannel identifying systems by channel face recognition algorithms by the first face static state video image of acquisition with
Criminal's photo and/or the video image comparison that public security system issues identify whether as questionable person.
Further, the questionable person for being sent into safety check room carries out detailed recognition of face through safety check identifying system again, and second
Server passes through safety check face recognition algorithms for the first face static state video image or the second face static state video image and criminal
Photo and/or video image comparison identify whether as questionable person.
According to the principle of dynamic film and television, we, which can identify Static Human Face, becomes dynamic human face identification.Assuming that one
Column train, 1500 people can be carried, it is specified that in advance half an hour enter the station and wait, if only one channel, front and back is crowded, block mutually,
It is impossible for obtaining everyone front face image.Cannot get front face image just can not be with the photo in identity card
Carry out identification comparison.To solve these problems, channel of entering the station is divided into three subchannels 4, each subchannel 4 by the embodiment of the present invention
Into 500 people, 10 identification seats are arranged in each subchannel 4, if each passenger's recognition time average out to 30 seconds, total passes through the time
It is 25 minutes.25 minutes 1500 people of identification, it should say it is a biggish dynamic recognition system, but be again one for someone
A static state identifying system.It need to only sit 30 seconds before screen, so that it may complete static identification.
The accuracy rate of present Static Human Face identification can achieve 99%, can't reach 100%, accurate in order to improve
Rate, the embodiment of the present invention are also provided with safety check identifying system 2 in safety check room, by the face recognition accuracy rate of lane identification system 1
Be relaxed to 95% (adjustable), questionable person be judged as by lane identification system 1, then is sent into safety check room, into safety check room again into
Row identification in detail.So-called detailed identification, comprising: use several different face recognition algorithms, carry out calculating matching identification in real time;
Show that the first face static state video image or the second face are static simultaneously respectively on second display screen 22 and third display screen 23
Video image and criminal's photo and/or video image carry out artificial contrast's identification by experienced public security officer;With public security system
Networking understands the present case of criminal by household register data check and wanted circular issuing unit, furthermore it is also possible to utilize public security system
The information (fingerprint, voice) that can be collected is assisted in identifying.
Above-described embodiment disclosed by the invention mainly can be set in Conventional trains station, long-distance bus station and harbour.Because
Technical solution operability disclosed by the invention is good, can be used for Conventional trains station, harbour and long-distance station, in dynamic stream of people's item
The system of recognition of face is carried out under part, it is therefore an objective to arrest runaway convict and forbid criminal Deng Che.
The embodiment of the present invention is provided with identification seat in front of the camera of lane identification system, and identification seat is stopped for passenger
Place, Conventional trains station, long-distance bus station and harbour each subchannel be arranged lane identification system for someone
For be a static identifying system because passenger only sits identification seat before screen 30 seconds, carry out static identification.This
Sample efficiently solves the technical issues of can not carrying out effective recognition of face in a dynamic state existing in the prior art.
General dynamic human face recognition system is mounted in station, at harbour entrance.Dynamic human face identification technology, it is international
Upper to be also in the starting stage at present, many problems not yet solve.Some problems such as pedestrian passage crowded the problem of blocking, with more
Advanced algorithm, also can't resolve.The present many units in China were all in installation, also runaway convict was caught in some places, just as elderly fishman
Fishing, is sitting under Xi Bianshu, fish is waited to rise to the bait, and can catch one and calculate one, this is for a set of modern equipment, and efficiency is too
It is low.Both made from now on discrimination be increased to 90% and can not use, for example, a column Conventional trains, deliver 1500 people, installed in portal
Face identification system may have in 1500 people 150 people to identify mistake, law-abiding citizen is identified as runaway convict, runaway convict is identified as
Law-abiding citizen is not only that the order of station or harbour can disorderly, and the order of entire city or country all can be in great confusion.
Design of the embodiment of the present invention for general iron, the face identification system of long-distance, harbour the dynamic stream of people a kind of and side
Method ultrahigh in efficiency.It is shunted using subchannel and most of unsuspicious personnel is excluded, a small number of a suspects are imported into safety check room, are used
Many algorithms are identified.So-called detailed identification is calculated in real time with a variety of different face recognition algorithms first
Machine matching identification.Wherein, safety check face recognition algorithms include the combination of the algorithm of following several recognitions of face: being based on geometrical characteristic
Face recognition algorithms, the face recognition algorithms based on eigenface, the face recognition algorithms based on Elastic Matching, be based on nerve net
The face recognition algorithms (currently leading method to be applied) of network, are based on branch at the face recognition algorithms based on hidden Markov model
Face recognition algorithms, the face recognition algorithms based on fractal image for holding vector machine.It is entered the station mode using subchannel, every people from channel
Number reduce, no longer occur it is crowded block, this is second innovative point, if blocked, any advanced recognizer can not all be solved
The problem of recognition accuracy.
Present face recognition algorithms can not correct head pitching, i.e. depth rotation, influence on recognition accuracy very big.
If the seat height in corridor is the same, tall person passenger will bow, and short a passenger Yao Yangtou influences correctly to identify.The present invention is real
The face identification system and method for applying a kind of dynamic stream of people disclosed in example can specify that 1.5 meters of height or less (child) without knowledge
It is not directly entered waiting room, 1.5-2 meters of height, is divided into 5 grades, into corresponding seat, them is made not have to bow or face upward head substantially, it can
Discrimination is improved, this is third innovative point.Direction of illumination and intensity influence discrimination also larger, and the embodiment of the present invention can be with
100 people are selected, the different directions of illumination and varying strength are tested repeatedly, optimum orientation and intensity are selected by light compensating lamp
Light source, to increase discrimination.This is the 4th innovative point.
Recognition of face decorates such as glasses to face, and cap etc. is more sensitive, and the embodiment of the present invention is trip by identification seat
Visitor prepares a seat, and passenger can sit down, removes glasses, cap etc. calmly, is put into the accuracy rate that identification is improved on table.This
It is the 5th innovative point.The selection of camera.There are two types of working methods for camera currently on the market, that is, monitor and capture,
Function and structure is also different.The embodiment of the present invention is used using high-definition camera and by conscientiously test selection high-definition camera
Which kind of working method seeks highest resolution.This is the 6th innovative point.The size of focal length of camera, to the ruler of facial image
Very little and recognition correct rate all has an impact.Video camera in station doorway or passageway not can guarantee pedestrian between camera
Distance also influences face identification rate that is, it cannot be guaranteed that the consistency of focal length, facial image vary.The embodiment of the present invention is logical
Crossing setting identification seat allows passenger to be sitting on seat, and camera is also contained near identification seat, when camera distance is shot using identity card
Distance, in this way, keep identity card imaging circumstances and the imaging circumstances of face identification system close as far as possible, comparative height, identification
Accuracy is also high.This is the 7th innovative point.The embodiment of the present invention solves the problems, such as primary election by channel face recognition algorithms first,
Most of passengers are freed, a small number of a suspects are imported safety check room and are identified in detail.It sets up safety check room and carries out second
Secondary identification is the 8th innovative point.In safety check room, the comprehensive computer that carries out is re-recognized, human eye control identification and with public security locality
Units concerned are linked up, and confirmation is errorless, just can determine that it is runaway convict.The embodiment of the present invention is photo (or the video on passenger's identity card
Image) and camera shooting picture, be simultaneously displayed on screen, the public security master-hand that please have wide experience carries out eye recognition,
If computer is identified as runaway convict, manual identified is considered as runaway convict, then carries out next step certification.Computer is identified and is had experience
It is our the 9th innovative point that public security manual identified, which combines,.The embodiment of the present invention is by the second server of safety check identifying system
It is networked by the network switch and public security system, public security certification may be implemented, this is the final stage of recognition of face, first to be inquired
Household register data and related case data.Railway, highway, airport and harbour have a public security garrison, safe examination system mostly by public security management,
It is more convenient to inquire household register, case data.It when these data prove, is indeed such a person, and fugitive, also wants and case responsible person's ditch
Lead to, confirm it is errorless after, first its people can be detained, after case responsible person come after arrest again.Face identification system final stage
Public security certification be the tenth innovative point.
Although above having used general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention are fallen within the scope of the claimed invention.
Claims (10)
1. a kind of face identification system of the dynamic stream of people, which is characterized in that the face identification system includes:
Lane identification system, for by channel face recognition algorithms to tentatively identified by the people of subchannel and judge be
No is questionable person;The lane identification system includes several subchannel identifying systems being arranged in several subchannels, each
Several subchannel identifying systems are provided in subchannel, the subchannel setting is being entered the station between channel and waiting room.
2. a kind of face identification system of the dynamic stream of people as described in claim 1, which is characterized in that subchannel identification system
System includes:
Camera, for acquiring the first face static state video image for passing through the people of subchannel;
First display screen, for showing the first face static state video image of camera acquisition;With
First server, for carrying out preliminary recognition of face according to the first face static state video image of acquisition;
Wherein, channel face recognition algorithms are provided in the first server, the first server passes through the network switch
It networks with safety check identifying system, the first server passes through channel face recognition algorithms for the first face static state video of acquisition
Criminal's photo and/or the video image comparison that image and safety check identifying system issue identify whether as questionable person.
3. a kind of face identification system of the dynamic stream of people as claimed in claim 2, which is characterized in that subchannel identification system
System further include: light compensating lamp, for carrying out light filling when camera acquires the first face static state video image;The camera, institute
It states first server, first display screen and the light compensating lamp and is integrated into touch one-piece machine.
4. a kind of face identification system of the dynamic stream of people as claimed in claim 3, which is characterized in that subchannel identification system
System further include: identify seat, be equipped with an identification seat in front of each camera.
5. a kind of face identification system of the dynamic stream of people as claimed in claim 3, which is characterized in that the subchannel is three
It is a, ten subchannel identifying systems are set in each subchannel.
6. the face identification system of dynamic stream of people as described in any in claim 1 to 5 a kind of, which is characterized in that the people
Face identifying system further include: safety check identifying system, for being sentenced by safety check face recognition algorithms to through channel face recognition algorithms
Break and is identified in detail for questionable person.
7. a kind of face identification system of the dynamic stream of people as claimed in claim 6, which is characterized in that the safety check identifying system
Include:
Video camera, for acquiring the second face static state video image;
Second display screen, for showing the first face static state video image or the second face static state video image;
Third display screen, for showing criminal's photo and/or video image;With
Second server, for carrying out detailed face according to the first face static state video image or the second face static state video image
Identification;
Wherein, safety check face recognition algorithms are provided in the second server, the second server passes through the network switch
It networks respectively with several first servers and public security system, each first server will be judged as through channel face recognition algorithms can
The first face static state video image of doubtful people is sent to the second server, and the public security system shines the criminal grasped
Piece and/or video image are sent to the second server, and the second server passes through safety check face recognition algorithms for first
Face static state video image or the second face static state video image and the comparison of criminal's photo and/or video image identify whether as can
Doubtful people.
8. a kind of face identification system of the dynamic stream of people as claimed in claim 7, which is characterized in that the safety check identifying system
Further include: the 4th display screen and the 5th display screen, the 4th display screen setting is on waiting room doorway, for showing through each height
Lane identification system is identified as the image of non-suspicious people, and the 5th display screen setting is on safety check room doorway, for showing through each
A sub- lane identification system is identified as the image of suspicious people.
9. a kind of face identification method of the dynamic stream of people characterized by comprising
The dynamic stream of people enters several subchannels by the corridor diversion that enters the station;
It is tentatively identified and is judged in batches by several subchannel identifying systems into the dynamic stream of people in each subchannel
It whether is questionable person;
If it is questionable person, it is sent into safety check room;And
If not being questionable person, enter waiting room;
Wherein, several subchannel identifying systems pass through channel face recognition algorithms for the first face static state video figure of acquisition
As criminal's photo and/or the video image comparison issued with public security system identifies whether as questionable person.
10. a kind of face identification method of the dynamic stream of people as claimed in claim 9, which is characterized in that be sent into the safety check room
Questionable person carry out detailed recognition of face again through safety check identifying system, the second server of the safety check identifying system passes through
Safety check face recognition algorithms are by the first face static state video image or the second face static state video image and criminal's photo and/or view
Frequency image comparison identifies whether as questionable person.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810936330.XA CN109255303A (en) | 2018-08-16 | 2018-08-16 | A kind of face identification system and method for the dynamic stream of people |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810936330.XA CN109255303A (en) | 2018-08-16 | 2018-08-16 | A kind of face identification system and method for the dynamic stream of people |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109255303A true CN109255303A (en) | 2019-01-22 |
Family
ID=65050231
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810936330.XA Pending CN109255303A (en) | 2018-08-16 | 2018-08-16 | A kind of face identification system and method for the dynamic stream of people |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109255303A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111885356A (en) * | 2020-07-21 | 2020-11-03 | 佛山市海协科技有限公司 | Security system and method based on Internet of things communication technology |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106156688A (en) * | 2015-03-10 | 2016-11-23 | 上海骏聿数码科技有限公司 | A kind of dynamic human face recognition methods and system |
CN106778556A (en) * | 2016-12-01 | 2017-05-31 | 大连因特视智能传感科技有限公司 | Intelligent identity identification system based on dynamic human face identification and Gait Recognition fusion |
CN107578521A (en) * | 2017-08-22 | 2018-01-12 | 西南交通大学 | Queuing management method and system, electronic equipment and computer-readable recording medium |
CN107945321A (en) * | 2017-11-08 | 2018-04-20 | 平安科技(深圳)有限公司 | Safety inspection method, application server and computer-readable recording medium based on recognition of face |
-
2018
- 2018-08-16 CN CN201810936330.XA patent/CN109255303A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106156688A (en) * | 2015-03-10 | 2016-11-23 | 上海骏聿数码科技有限公司 | A kind of dynamic human face recognition methods and system |
CN106778556A (en) * | 2016-12-01 | 2017-05-31 | 大连因特视智能传感科技有限公司 | Intelligent identity identification system based on dynamic human face identification and Gait Recognition fusion |
CN107578521A (en) * | 2017-08-22 | 2018-01-12 | 西南交通大学 | Queuing management method and system, electronic equipment and computer-readable recording medium |
CN107945321A (en) * | 2017-11-08 | 2018-04-20 | 平安科技(深圳)有限公司 | Safety inspection method, application server and computer-readable recording medium based on recognition of face |
Non-Patent Citations (1)
Title |
---|
都伊林: "《智能安防新发展与应用》", 31 July 2018, 华中科技大学出版社 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111885356A (en) * | 2020-07-21 | 2020-11-03 | 佛山市海协科技有限公司 | Security system and method based on Internet of things communication technology |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9875392B2 (en) | System and method for face capture and matching | |
CN107992786A (en) | A kind of people streams in public places amount statistical method and system based on face | |
CN103839346B (en) | A kind of intelligent door and window anti-intrusion device and system, intelligent access control system | |
CN111762649B (en) | Elevator article detection system and method, article detection system, elevator light curtain and elevator equipment | |
CN108470169A (en) | Face identification system and method | |
CN103440475B (en) | A kind of ATM user face visibility judge system and method | |
CN106241584A (en) | A kind of intelligent video monitoring system based on staircase safety and method | |
CN210955266U (en) | Internet of things temperature-sensing height and identification integrated machine | |
CN106600631A (en) | Multiple target tracking-based passenger flow statistics method | |
CN106156688A (en) | A kind of dynamic human face recognition methods and system | |
CN104091176A (en) | Technology for applying figure and head portrait comparison to videos | |
CN108280435A (en) | Passenger abnormal behavior recognition method based on human body posture estimation | |
CN103839373A (en) | Sudden abnormal event intelligent identification alarm device and system | |
CN104504408A (en) | Human face identification comparing method and system for realizing the method | |
CN111126219A (en) | Transformer substation personnel identity recognition system and method based on artificial intelligence | |
CN106682578A (en) | Human face recognition method based on blink detection | |
CN105022999A (en) | Man code company real-time acquisition system | |
CN105574509A (en) | Face identification system playback attack detection method and application based on illumination | |
CN114267111A (en) | All-in-one traffic control method, system and equipment | |
CN105913022A (en) | Handheld calling state determining method and handheld calling state determining system based on video analysis | |
CN208351494U (en) | Face identification system | |
CN111710077A (en) | Intelligent gate machine, management system and management method for personnel safety management and control | |
CN109657555A (en) | It is a kind of to get off the detection device and method of omission for school bus student | |
CN110097724A (en) | A kind of article based on FPGA nurses method and system automatically | |
CN108876976A (en) | Card identification device, image collecting device grab and steal card system and 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 |
Application publication date: 20190122 |
|
RJ01 | Rejection of invention patent application after publication |