CN107423673A - A kind of face identification method and system - Google Patents
A kind of face identification method and system Download PDFInfo
- Publication number
- CN107423673A CN107423673A CN201710329207.7A CN201710329207A CN107423673A CN 107423673 A CN107423673 A CN 107423673A CN 201710329207 A CN201710329207 A CN 201710329207A CN 107423673 A CN107423673 A CN 107423673A
- Authority
- CN
- China
- Prior art keywords
- face
- data
- target person
- mobile terminal
- recognition result
- 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
- G06V40/168—Feature extraction; Face representation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The embodiment of the invention discloses a kind of face identification method and system.This method includes：Target person data are obtained according to recognition of face instruction；By the target person data transfer to mobile terminal, so that the mobile terminal is handled the target person data, and face recognition result is obtained；Receive and show the face recognition result.Pass through above-mentioned technical proposal, solving the problems, such as wearable smart machine, time-consuming in face recognition process, recognition result is unstable, realize in the case where not changing wearable smart machine hardware condition, increase the stability of wearable smart machine recognition of face, shorten face recognition process in when consume, Consumer's Experience is improved, has promoted the application of wearable smart machine.
Description
Technical field
The present embodiments relate to face recognition technology, more particularly to a kind of face identification method and system.
Background technology
With the development of wearable smart machine technology, the face recognition technology based on wearable smart machine is got over
Carry out more attention, be based especially on the face recognition technology of intelligent glasses, it is in fields such as social activity, public security, security personnel and services
In demand more and more higher.
The current face recognition technology based on intelligent glasses is typically that microprocessor is carried in intelligent glasses, to intelligence
The personage's picture or video that glasses obtain carry out recognition of face, obtain preliminary face recognition result, afterwards that this is preliminary
Face recognition result carries out face matching in the face database of intelligent glasses or cloud server, obtains final face
Recognition result.
But framework, limited microprocessor performance and the wireless transmission capability excessively concentrated due to intelligent glasses so that
Above-mentioned face recognition process stability is poor, time-consuming, influences the usage experience of user.
The content of the invention
The embodiment of the present invention provides a kind of face identification method and system, more steady using wearable smart machine to realize
Determine, more efficiently carry out recognition of face.
In a first aspect, the embodiments of the invention provide a kind of face identification method, applied to wearable smart machine, including
Following steps：
Target person data are obtained according to recognition of face instruction；
By the target person data transfer to mobile terminal, so that the mobile terminal enters to the target person data
Row processing, and obtain face recognition result；
Receive and show the face recognition result.
Alternatively, it is described to include according to recognition of face instruction acquisition target person data：
Recognition of face instruction is received, obtains the photographed data in the range of camera view；
According to the photographed data, target person data are obtained；
Wherein, the photographed data includes the video stream data and/or photograph in the range of full field range and/or partial visual field
Sheet data.
Alternatively, the mobile terminal is handled the target person data, and is obtained face recognition result and included：
The mobile terminal pre-processes to the target person data, obtains intermediate objective character data, and foundation
The intermediate objective character data obtains face recognition result；Or,
The mobile terminal by the target person data transfer to setting server so that the setting server is to institute
State target person data and carry out recognition of face；
The mobile terminal receives the face recognition result of the setting server transport.
Further, the mobile terminal is handled the target person data of reception, obtains intermediate objective people
Thing data include：
The mobile terminal carries out the extraction of human face characteristic point to the target person data, and it is special to obtain target person face
Data are levied, as intermediate objective character data.
Further, it is described to include according to intermediate objective character data acquisition face recognition result：
The mobile terminal transmits the intermediate objective character data to the setting server, so that the setting clothes
Device be engaged in intermediate objective character data progress recognition of face；
The mobile terminal receives the face recognition result of the setting server transport.
Further, it is described to include according to intermediate objective character data acquisition face recognition result：
The mobile terminal determines target person information according to the intermediate objective character data；
The mobile terminal is matched the target person information with face database, obtains face recognition result.
Alternatively, the wearable smart machine is intelligent glasses.
Second aspect, the embodiment of the present invention additionally provide a kind of face identification system, and the system includes：Wearable intelligence is set
Standby and mobile terminal；
The wearable smart machine includes camera, embedded microprocessor and wireless communication apparatus；The camera
For obtaining target person data；The embedded microprocessor is used to receive external command and the processing wearable intelligence is set
Standby internal flow of information；The wireless communication apparatus is used for the target person data transfer to the mobile terminal, and
Receive the face recognition result of the mobile terminal transmission；
The mobile terminal is used for the target person data for receiving and handling the wearable smart machine transmission, obtains
The face recognition result is obtained, and the face recognition result is transmitted to the wearable smart machine；
Wherein, the wearable smart machine is intelligent glasses.
Alternatively, the mobile terminal is additionally operable to handle the target person data of reception, obtains middle mesh
Character data is marked, and the intermediate objective character data is handled, to obtain the face recognition result.
Alternatively, on the basis of said system, the system also includes setting server,
The mobile terminal is additionally operable to transmit the target person data or the intermediate objective character data to described
Server is set, so that the setting server carries out face to the target person data or the intermediate objective character data
Identification, and receive the face recognition result of the setting server transport；
The setting server is used to receive the target person data of the mobile terminal transmission or the middle mesh
Mark character data, and recognition of face is carried out according to the target person data or the intermediate objective character data, described in acquisition
Face recognition result；
The setting server is additionally operable to transmit the face recognition result to the mobile terminal.
The embodiment of the present invention according to recognition of face instruction by obtaining target person data；The target person data are passed
Mobile terminal is transported to, so that the mobile terminal is handled the target person data, and obtains face recognition result；Connect
Receive and show the face recognition result.The face recognition process stability that solves wearable smart machine is poor, time-consuming
Problem, realize in the case where not changing wearable smart machine hardware condition, overcome wearable intelligence to set to a certain extent
Standby hardware limitation, increases the stability of wearable smart machine recognition of face, shorten in face recognition process when consume, lifted
Consumer's Experience, has promoted the application of wearable smart machine.
Brief description of the drawings
Fig. 1 is a kind of flow chart of face identification method in the embodiment of the present invention one；
Fig. 2 is a kind of flow chart of face identification method in the embodiment of the present invention two；
Fig. 3 is the comparative result figure of three kinds of method transmission speeds in the embodiment of the present invention two；
Fig. 4 is the comparative result figure of three kinds of method stability in the embodiment of the present invention two；
Fig. 5 is a kind of structural representation of face identification system in the embodiment of the present invention three；
Fig. 6 is the intelligent glasses structural representation in the embodiment of the present invention three.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that in order to just
Part related to the present invention rather than entire infrastructure are illustrate only in description, accompanying drawing.
Embodiment one
Fig. 1 is a kind of flow chart for face identification method that the embodiment of the present invention one provides, and the present embodiment is applicable to profit
The situation of recognition of face is carried out with wearable smart machine, this method can be performed by face identification device, and the device can be with
Realized by the mode of software and/or hardware, the device can be integrated in wearable with image-acquisition functions and communication function
In smart machine, such as typically intelligent glasses or intelligent watch etc., illustrated in the present embodiment by taking intelligent glasses as an example.
Specifically comprise the following steps：
S110, acquisition target person data are instructed according to recognition of face.
Specifically, user use wearable smart machine, as intelligent glasses face identification functions when, intelligence can be triggered
The face identification functions of glasses, its triggering mode can be triggered by the button or touch-screen of Trackpad, can also pass through nothing
Line communication device remote control trigger, for example by modes such as the bluetooth in intelligent glasses or remote controls, phonetic order can also be passed through
Triggering, is not specifically limited here.
After intelligent glasses receive recognition of face instruction, the camera that intelligent glasses carry can be triggered, to obtain intelligence
Glasses video within the vision or photo, the image for the personage for needing to identify, i.e. target person are included in the video or photo
Data.
Exemplarily, obtaining target person data according to recognition of face instruction includes：Recognition of face instruction is received, acquisition is taken the photograph
As a photographed data within the vision；According to the photographed data, target person data are obtained；Wherein, the photographed data
Including the video stream data and/or picture data in the range of full field range and/or partial visual field.
Specifically, intelligent glasses trigger intelligent glasses camera, now on screen after recognition of face instruction is received
The video flowing in the range of camera view is shown, can now select to obtain corresponding to camera 1) that whole field ranges are (i.e. entirely
Field range) in video stream data, 2) video stream data in the range of partial visual field, 3) full shooting photo within the vision
Data and 4) any of shooting picture data in the range of partial visual field data, as photographed data.Then from the shooting
Extracting data target person data, the extraction process can be directly using photographed data as target person data, can also
It is to carry out further cutting to photographed data to handle, obtains target person data.
4 kinds of data for more than, user can select data the most suitable as shooting according to the actual demand of oneself
Data.Such as：If the user desired that obtaining, data volume is smaller, but data cover the more rich photographed data of information, then can be with
Selection is 1).If the user desired that obtaining, data volume is smaller, and only comprising personage (i.e. target person) information for oneself needing to identify
Photographed data, can select 2).Here partial visual field scope can be with sets itself.But, it is contemplated that in practical application, use
The personage for needing to identify would generally be positioned over the central area in the range of camera view by family, so the partial visual field model in 2)
Preferably central field of vision scope is enclosed, the size of the central field of vision scope, can rule of thumb be set, also according to camera and target
The distance of personage, automatically adjust for this apart from the field range size shared by lower personage.If user does not consider data volume size,
And it is desirable to obtain that resolution ratio is higher, and data cover the more rich photographed data of information, then can select 3).If user couple
The richness that data cover information content does not specially require, and is desirable to that resolution ratio is higher, and data volume is relatively small, then can select
Select 4).
In the present embodiment, preferably 2) obtained by data as photographed data.Because the resolution ratio of video flowing relative to
The resolution ratio of photo is relatively low, then the response time for obtaining video flowing can be relatively short, and response speed can be relatively fast, Neng Gou
To a certain extent reduce intelligent glasses recognition of face when consume, improve the response speed of recognition of face.In addition, the fetching portion visual field
In the range of video stream data, can realize effective utilization of resource, reduce data volume, further improve the people in the present embodiment
Face identifies stability and response speed, meanwhile, it can also simplify personage's background, so as to reduce the difficulty of follow-up recognition of face.
S120, by the target person data transfer to mobile terminal so that the mobile terminal is to the target person
Data are handled, and obtain face recognition result.
Generally, using wearable smart machine, the process as intelligent glasses carry out recognition of face is to pass through intelligent glasses
Embedded microprocessor, target person data are handled, obtain target person information, then by the target person information and people
Human face data in face database is matched, and then obtains the face recognition result of target person.Either, by said process
Cloud server is positioned over, the Wireless Fidelity net that the target person data or target person information that will be obtained pass through intelligent glasses
Network (WiFi) is transmitted to cloud server, and all or part of face recognition process is carried out by cloud server, then by acquisition
Face recognition result is sent to intelligent glasses.
But because intelligent glasses compact in architecture itself, the disposal ability of its microprocessor that can be carried generally have very much
Limit, this accuracy for allowing for carrying out recognition of face by intelligent glasses itself is difficult to ensure, and its process is time-consuming longer, produces
Heat is larger.Similarly, the concentrated type structure of intelligent glasses is limited to, the WiFi devices of intelligent glasses are difficult to ensure data at present
The stability and speed of transmission, especially out of doors in the case of long-distance transmissions, its data transmission capabilities is more limited, and this is just
So that the when consumption that face recognition process is carried out by cloud server is also longer, or even identification process is interrupted, and can not carry out face
Identification.Above-mentioned these can all cause the face recognition result that user can not obtain its needs in time, influence the usage experience of user.
Comparatively speaking, the network transmission technology of mobile terminal is highly developed, and it can utilize WiFi and mobile honeybee
The multiple networks such as nest network rapidly, continuously carry out data transmission.Moreover, user can carry movement out of doors under normal circumstances
Terminal, can guarantee that the real-time Transmission of data, and be not only restricted to the environment residing for user.In addition, user wear intelligent glasses with
Its carry mobile terminal between distance it is shorter, carry out data transmission between both be generally not in transmission slowly or in
Disconnected situation, transmission is relatively stable, can ensure effective transmission of data.
So in order that intelligent glasses recognition of face it is more quick, stably, the present embodiment is not changing intelligent glasses
On the basis of hardware device, the target person data transfer that intelligent glasses are obtained is to mobile terminal, to utilize mobile terminal pair
Target person data are handled, and obtain face recognition result.
It should be appreciated that mobile terminal is handled target person data, and obtains the process of face recognition result,
It can be the complete procedure that mobile terminal direct basis target person data perform recognition of face, finally determine target person
Identity information, that is, obtain face recognition result；Can also be mobile terminal to target person data without any processing, but
The server (setting server) that can carry out recognition of face directly is transmitted it to, complete face is carried out by setting server
Identification process, and the face recognition result of acquisition is transmitted back to mobile terminal；It can also be mobile terminal to target person data
Pre-processed, such as at least one in the operation such as cutting, compression, the extraction of character features point and determination target person information,
Then again by pretreated target person data transfer to server is set, face knowledge is further carried out by setting server
Not, and by the face recognition result of acquisition it is transmitted back to mobile terminal.
Wherein, data transfer mode can be the wired or wireless communication device of intelligent glasses.Preferably, intelligent glasses lead to
Wireless communication apparatus, such as at least one of WiFi, bluetooth and mobile cellular network are crossed, by target person data transfer to shifting
Dynamic terminal.The stability of data is transmitted to be further ensured that intelligent glasses to mobile terminal, is further preferably entered by bluetooth
The transmission of row data.Because network settings of the bluetooth independent of outside, and WiFi network is dependent on the outside net set
Network is route；In addition, being usually provided with bluetooth in current intelligent glasses, its hardware is varied without using bluetooth data transmission, no
Increase extra cost, there is universality, and mobile cellular network device only has corresponding configuration in the intelligent glasses of part at present.
Certainly, in the case of the WiFi network environment residing for user is preferable, can also be carried out data transmission using WiFi；Or use
When the intelligent glasses that family is worn are configured with mobile cellular network device, it can also be carried out data transmission using mobile cellular network.
S130, reception simultaneously show the face recognition result.
Specifically, in S120, mobile terminal obtains face recognition result and then is transmitted back to the face recognition result
Intelligent glasses, so that intelligent glasses can be to the specifying information of user display target personage.
The embodiment of the present invention according to recognition of face instruction by obtaining target person data；The target person data are passed
Mobile terminal is transported to, so that the mobile terminal is handled the target person data, and obtains face recognition result；Connect
Receive and show the face recognition result.The face recognition process stability that solves wearable smart machine is poor, time-consuming
Problem, realize in the case where not changing wearable smart machine hardware condition, overcome wearable intelligence to set to a certain extent
Standby hardware limitation, increases the stability of wearable smart machine recognition of face, shorten in face recognition process when consume, lifted
Consumer's Experience, has promoted the application of wearable smart machine.
Embodiment two
Fig. 2 is a kind of flow chart for face identification method that the embodiment of the present invention two provides, and the present embodiment is in above-mentioned implementation
On the basis of example, to " by the target person data transfer to mobile terminal, so that the mobile terminal is to the target person
Data are handled, and obtain face recognition result " the step of be optimized.Similarly, the wearable intelligence in the present embodiment
Can equipment still illustrated by taking intelligent glasses as an example, wherein the explanation of same as the previously described embodiments or corresponding term is herein not
Repeat again.
Illustrated with reference to Fig. 2 face identification methods provided the embodiment of the present invention two, the method for the present embodiment
Including：
S210, intelligent glasses obtain target person data according to recognition of face instruction, perform S220.
S220, intelligent glasses are by the target person data transfer to mobile terminal.
Specifically, intelligent glasses are by the target person data transfer to mobile terminal, mobile terminal receive it is above-mentioned
After target person data, it can select to perform S230 or S250 according to practical application request.
S230, mobile terminal by the target person data transfer to setting server so that the setting server pair
The target person data carry out recognition of face.
Specifically, mobile terminal can not be carried out after the target person data of intelligent glasses transmission are received to it
Any processing, but setting server is directly re-transmitted to, follow-up recognition of face is directly carried out by setting server
Operation, face recognition result is obtained, and the face recognition result of acquisition is transmitted back to mobile terminal, perform S240 afterwards.
So setting is advantageous in that, only by the use of mobile terminal as data transmission medium, reduces the intermediate treatment of data
Process so that face recognition process is more stable, and can ensure the security of data；Meanwhile, it is capable to save mobile terminal
Resource loss.
The technical scheme of this recognition of face, it is suitable for the high situation of face database security requirement, such as public security neck
Domain.
S240, mobile terminal receive the face recognition result of the setting server transport.
Specifically, mobile terminal receives the face recognition result of setting server transport, and performs S260.
S250, mobile terminal pre-process to the target person data, obtain intermediate objective character data, and foundation
The intermediate objective character data obtains face recognition result.
Specifically, mobile terminal can be carried out first after the target person data of intelligent glasses transmission are received to it
At least one of in pretreatment, such as the operation such as cutting, compression, the extraction of character features point and target person acquisition of information, acquisition
Intermediate objective character data, face recognition result then is obtained according to the intermediate objective character data again, and by the recognition of face
As a result transmit to intelligent glasses, perform S260 afterwards.
Exemplarily, the mobile terminal is handled the target person data of reception, obtains intermediate objective people
Thing data include：The mobile terminal carries out the extraction of human face characteristic point to the target person data, obtains target person people
Face characteristic, as intermediate objective character data.
Specifically, mobile terminal is handled the target person data of reception, extracts human face characteristic point therein, than
The position of such as face, direction, the positional information of size and each major facial organ, as intermediate objective character data.So
Setting is advantageous in that the target person data directly shot compared to traditional transmission, the data volume of human face characteristic point is smaller,
Data transmission stability can be further improved, while saves data traffic；In addition, compared to conventional transmission target person data
During the processing of feature point extraction is carried out after advanced row data compression, passed after first proposing human face characteristic point in the present embodiment
Defeated processing, more face informations can be obtained, improve the accuracy rate of follow-up recognition of face.
It should be noted that the above-mentioned process that face recognition result is obtained according to the intermediate objective character data, at least may be used
Comprising two kinds of situations, to be respectively suitable for different application scenarios.
First exemplarily, and obtaining face recognition result according to the intermediate objective character data includes：Mobile terminal will
The intermediate objective character data is transmitted to the setting server, so that the setting server is to the intermediate objective personage
Data carry out recognition of face；The mobile terminal receives the face recognition result of the setting server transport.
Specifically, mobile terminal directly passes people in the middle's data after intermediate objective character data is got again
Transport to setting server.The determination of target person information is carried out by setting server, and by the target person information of determination and people
Human face data in face database is matched, and obtains face recognition result.Then, setting server directly can know the face
Other result is transmitted back to above-mentioned mobile terminal again.Mobile terminal receives the face recognition result passed back of setting server, and by the people
Face recognition result is transmitted to intelligent glasses.
It should be appreciated that mobile terminal after intermediate objective character data is obtained, can be carried out in native internal
The determination of target person information, then again by the target person information transfer to server is set, so that setting server is carried out
The matching of human face data in face database, and finally obtain face recognition result.As long as that is, human face data library storage
In setting server.
So setting is advantageous in that, the preliminary treatment of target person data is carried out in mobile terminal, then still by setting
Determine server and carry out recognition of face, can further reduce in the case where taking mobile terminal own resources less as far as possible
With the volume of transmitted data between setting server so that the face recognition process of intelligent glasses is more stable.
The technical scheme of this recognition of face, the similarly suitable situation higher in face database security requirement, than
Such as ensure public security field or police field.
Second exemplarily, described to include according to intermediate objective character data acquisition face recognition result：The shifting
Dynamic terminal determines target person information according to the intermediate objective character data；The mobile terminal believes the target person
Breath is matched with face database, obtains face recognition result.
Specifically, after mobile terminal obtains intermediate objective character data, directly it is based on the target person in native internal
Data carry out the determination of target person information.Then, the people that mobile terminal stores up the target person information of determination and local memory
Human face data in face database is matched, and obtains face recognition result.That is, mobile terminal directly carries out face knowledge
Other processing, no longer it is processed to process and is positioned over setting server.So setting is advantageous in that, eliminate mobile terminal with
The data transmission procedure of server is set, enables to the process of recognition of face more stable.
The technical scheme of this recognition of face, because face database is directly stored in mobile terminal, so being suitable for face
Database security requires relatively low situation, such as social field or generic services field etc..
S260, intelligent glasses receive and show face recognition result.
It should also be noted that in the present embodiment, the mode of mobile terminal to setting server transmission data can root
Selected according to the network environment residing for it, for example the WiFi network of its local environment is preferable, then can be utilized with prioritizing selection
WiFi is transmitted, and can save its own traffic of mobile terminal；If the WiFi network of its local environment is poor, then can be preferential
Its cellular network data of Selection utilization is transmitted, and to ensure effective transmission of data, this mode is particularly suitable for outdoor remote
The recognition of face of Distance Transmission, such as public security open air are handled a case.
In addition, setting server directly can transmit face recognition result to intelligent glasses, still, it is contemplated that Brilliant Eyes
The hardware case of mirror may result in transmission slowly or interrupt.So preferred scheme is still to know face in the present embodiment
Other result transmission is back to mobile terminal, then is transmitted the face recognition result to intelligent glasses by mobile terminal.
It is to be further understood that the method in the present embodiment has no effect on the original recognition of face mode of intelligent glasses, i.e.,
The microprocessor and network communication device carried using intelligent glasses carries out all or part of face recognition process.
The embodiment of the present invention obtains target person data by intelligent glasses, and by the target person data transfer to movement
Terminal；Mobile terminal pre-processes to the target person data, obtains intermediate objective character data, and according to the centre
Target person data obtain face recognition result；Or mobile terminal by the target person data transfer to setting server,
So that the setting server carries out recognition of face to the target person data；The mobile terminal receives the setting service
The face recognition result of device transmission；Finally, intelligent glasses receive and show the face recognition result.Solves wearable intelligence
The problem of face recognition process stability of equipment is poor, time-consuming, realizes and is not changing wearable smart machine hardware condition
In the case of, increase the stability of wearable smart machine recognition of face, shorten in face recognition process when consume, further carry
Consumer's Experience has been risen, has promoted the application of wearable smart machine.
Embodiment three
In the present embodiment based on the various embodiments described above, using the intelligent glasses recognition of face in wearable smart machine as
Example, has carried out experimental data demonstration specific to face identification method therein.
Compared to traditional intelligent glasses face identification method (being designated as conventional method), mobile terminal is utilized in the present embodiment
As the face identification method (being designated as method 1) of data transfer intermediary, in the response stability of recognition of face and transmission speed side
Face is all greatly improved.In the case of intelligent glasses and transmission size of data identical (such as 60KB), to conventional method
4 recognitions of face test, such as table have been carried out respectively under different identification distances (such as 1m, 1.5m, 2m and 2.5m) with method 1
Shown in 1.From table 1 it follows that under different identification distances, response time of the recognition of face of method 1 in 6s or so, has
Good stability；And then there is larger fluctuation the conventional method recognition of face response time, being averaged under different identification distances
Response time is 6s~14s, less stable, especially relative to most of test 5s~7s response times, 1.5m's
Under identification distance, the response time for testing 4 is 8.26s；Under 2m identification distances, test 2, test 3 and the response time for testing 4
Respectively reach 28.84s, 13s and 9.23s；Under 2.5m identification distance, the response time of test 2 and test 3 respectively reaches
14.75s and 11.7s.As can be seen here, different degrees of response delay in conventional method be present, and response delay is with higher
Incidence and larger uncertainty, have a strong impact on the stability of its recognition of face；And above mentioned problem is then not present in method 1,
Its recognition of face has preferably repeatability.
The recognition of face test comparison result of the conventional method of table 1 and method 1
Further, using same test intelligent glasses and identification distance, it will obtain in the present embodiment and utilize movement eventually
The test result as the face identification method of data processing and data transfer intermediary (being designated as method 2) is held, as shown in table 2.Its
In, the transmission size of data of intelligent glasses to mobile terminal is identical with the transmission size of data of conventional method (or method 1), also for
60KB, and mobile terminal is then much smaller to the transmitted data amount for setting server, is 512B.From table 2 it can be seen that 1m,
Under 1.5m, 2m and 2.5m identification distance, response time of the recognition of face of method 2 in 4s or so, compared to conventional method, equally
With good stability.
The recognition of face test comparison result of the method 2 of table 2
In order to more intuitively be compared, we obtain the average time-consuming and time-consuming of conventional method, method 1 and method 2
Variance ratio is shown in Fig. 3 and Fig. 4 respectively compared with result figure.
From figure 3, it can be seen that in terms of transmission speed, the average time-consuming minimum of method 2, transmission speed is most fast, method 1 time
It, and the average time-consuming maximum of conventional method, transmission speed are most slow.Specifically, under different identification distances, compared to biography
System method, the transmission speed of method 1 improve 7.52%, 8.61%, 58.93% and 34.32% respectively, the transmission speed of method 2
Degree improves 41.25%, 36.53%, 69.81% and 51.45% respectively；Compared to method 1, the transmission speed of method 2 is then divided
36.47%, 30.56%, 26.48% and 26.08% is not improved.
From fig. 4, it can be seen that for transmission stability, under different identification distances, the time-consuming variance ratio of conventional method
The time-consuming variance of method 1 and method 2 will be big, especially under 2m and 2.5m identification distance, conventional method and method 1 and method
2 time-consuming variance difference is bigger.That is, the stability of conventional method is worst.In addition, in more tests, method 2
Stability is better than the stability of method 1 on the whole.
Although the method that the embodiment of the present invention is can be seen that by above-mentioned two comparing result adds intelligent glasses face
The transmission of identification and processing procedure, but be their ability to greatly reduce face recognition process in when consume, lifting recognition of face it is steady
It is qualitative, the practical application of intelligent glasses can be promoted well.It should be noted that the concrete numerical value of above-mentioned experimental data demonstration
As a result, can because of intelligent glasses used and its plug-in and difference, herein only as an example, to illustrate the embodiment of the present invention
The effect of middle offer method.
The technical scheme of the present embodiment, target person number is obtained according to the recognition of face instruction received by intelligent glasses
According to, and by wireless communication apparatus by target person data transfer to mobile terminal, target person number of the mobile terminal to reception
According to being handled, intermediate objective character data is obtained, the intermediate objective character data is transmitted to setting server to enter pedestrian
Face identifies, and receives the face recognition result of setting server transmission, afterwards transmits face recognition result to intelligent glasses,
Received by intelligent glasses and show the face recognition result.Solving intelligent glasses, time-consuming in face recognition process, identification
The problem of unstable result, realize in the case where not changing intelligent glasses hardware condition, increase intelligent glasses recognition of face
Stability, lifted recognition of face speed, effectively shorten face recognition process in when consume, further improve user's body
Test.
It is the embodiment of face identification system provided in an embodiment of the present invention, face identification system and above-mentioned each implementation below
The face identification method of example belongs to same inventive concept, in the embodiment of face identification system in the details of not detailed description
Hold, may be referred to the embodiment of above-mentioned face identification method.
Example IV
Fig. 5 is a kind of structural representation for face identification system that the embodiment of the present invention four provides.Referring to Fig. 5, the system
It can include：Wearable smart machine 510 and mobile terminal 520；
By taking the intelligent glasses 610 in wearable smart machine 510 as an example, referring to Fig. 6, intelligent glasses 610 include camera
611st, embedded microprocessor 612 and wireless communication apparatus 613；The camera 610 is used to obtain target person data；It is described
Embedded microprocessor 612 is used for the internal flow of information for receiving external command and the processing wearable smart machine；The nothing
Line communication device 613 is used for the target person data transfer to the mobile terminal 520, and receives the mobile terminal
The face recognition result of 520 transmission；
The mobile terminal 520 is used to receiving and handling the target person that the wearable smart machine 510 transmits
Data, the face recognition result is obtained, and the face recognition result is transmitted to the wearable smart machine 510；
Alternatively, the mobile terminal 510 is additionally operable to handle the target person data of reception, obtains middle
Target person data, and the intermediate objective character data is handled, to obtain the face recognition result.
Alternatively, on the basis of said system, the system also includes setting server 530,
The mobile terminal 520 is additionally operable to transmit the target person data or the intermediate objective character data to institute
Setting server 530 is stated, so that the setting server 530 is to the target person data or the intermediate objective character data
Recognition of face is carried out, and receives the face recognition result that the setting server 530 transmits；
The setting server 530 is used to receive the target person data or described that the mobile terminal 520 transmits
Intermediate objective character data, and recognition of face is carried out according to the target person data or the intermediate objective character data, obtain
Obtain the face recognition result；
The setting server 530 is additionally operable to transmit the face recognition result to the mobile terminal 520.
By a kind of face identification system of the embodiment of the present invention three, solves wearable smart machine in recognition of face
The problem of time-consuming in journey, recognition result is unstable, realize in the case where not changing wearable smart machine hardware condition,
The hardware limitation of wearable smart machine is overcome to a certain extent, increases the stability of wearable smart machine recognition of face, is contracted
In short face recognition process when consume, improve Consumer's Experience, promoted the application of wearable smart machine.
The face identification system that the embodiment of the present invention is provided, which can perform the face that any embodiment of the present invention is provided, to be known
Other method, possess the corresponding functional module of execution method and beneficial effect.
Pay attention to, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although being carried out by above example to the present invention
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other more equivalent embodiments can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
- A kind of 1. face identification method, applied to wearable smart machine, it is characterised in that including：Target person data are obtained according to recognition of face instruction；By the target person data transfer to mobile terminal so that the mobile terminal to the target person data at Reason, and obtain face recognition result；Receive and show the face recognition result.
- 2. according to the method for claim 1, it is characterised in that described to obtain target person data according to recognition of face instruction Including：Recognition of face instruction is received, obtains the photographed data in the range of camera view；According to the photographed data, target person data are obtained；Wherein, the photographed data includes the video stream data and/or number of pictures in the range of full field range and/or partial visual field According to.
- 3. according to the method for claim 1, it is characterised in that the mobile terminal to the target person data at Manage, and obtain face recognition result to include：The mobile terminal pre-processes to the target person data, obtains intermediate objective character data, and according to described Intermediate objective character data obtains face recognition result；Or,The mobile terminal by the target person data transfer to setting server so that the setting server is to the mesh Mark character data and carry out recognition of face；The mobile terminal receives the face recognition result of the setting server transport.
- 4. according to the method for claim 3, it is characterised in that the target person data of the mobile terminal to reception Handled, obtaining intermediate objective character data includes：The mobile terminal carries out the extraction of human face characteristic point to the target person data, obtains target person face characteristic number According to as intermediate objective character data.
- 5. according to the method for claim 3, it is characterised in that described to obtain face according to the intermediate objective character data Recognition result includes：The mobile terminal transmits the intermediate objective character data to the setting server, so that the setting server Recognition of face is carried out to the intermediate objective character data；The mobile terminal receives the face recognition result of the setting server transport.
- 6. according to the method for claim 3, it is characterised in that described to obtain face according to the intermediate objective character data Recognition result includes：The mobile terminal determines target person information according to the intermediate objective character data；The mobile terminal is matched the target person information with face database, obtains face recognition result.
- 7. according to the method described in claim any one of 1-6, it is characterised in that the wearable smart machine is Brilliant Eyes Mirror.
- A kind of 8. face identification system, it is characterised in that including：Wearable smart machine and mobile terminal；The wearable smart machine includes camera, embedded microprocessor and wireless communication apparatus；The camera is used for Obtain target person data；The embedded microprocessor is used to receive external command and handles the wearable smart machine Internal flow of information；The wireless communication apparatus is used for the target person data transfer to the mobile terminal, and receives The face recognition result of the mobile terminal transmission；The mobile terminal is used for the target person data for receiving and handling the wearable smart machine transmission, obtains institute Face recognition result is stated, and the face recognition result is transmitted to the wearable smart machine；Wherein, the wearable smart machine is intelligent glasses.
- 9. system according to claim 8, it is characterised in that the mobile terminal is additionally operable to the target person to reception Thing data are handled, and obtain intermediate objective character data, and the intermediate objective character data is handled, to obtain State face recognition result.
- 10. system according to claim 9, it is characterised in that also include setting server,The mobile terminal is additionally operable to transmit the target person data or the intermediate objective character data to the setting Server, so that the setting server carries out face knowledge to the target person data or the intermediate objective character data Not, and receive it is described setting server transport face recognition result；The setting server is used to receive the target person data of the mobile terminal transmission or the intermediate objective people Thing data, and recognition of face is carried out according to the target person data or the intermediate objective character data, obtain the face Recognition result；The setting server is additionally operable to transmit the face recognition result to the mobile terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710329207.7A CN107423673A (en) | 2017-05-11 | 2017-05-11 | A kind of face identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710329207.7A CN107423673A (en) | 2017-05-11 | 2017-05-11 | A kind of face identification method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107423673A true CN107423673A (en) | 2017-12-01 |
Family
ID=60425164
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710329207.7A Pending CN107423673A (en) | 2017-05-11 | 2017-05-11 | A kind of face identification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107423673A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108805091A (en) * | 2018-06-15 | 2018-11-13 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating model |
| CN109886199A (en) * | 2019-02-21 | 2019-06-14 | 百度在线网络技术（北京）有限公司 | Information processing method, device, vehicle and mobile terminal |
| CN110135222A (en) * | 2018-02-08 | 2019-08-16 | 上海荟宸信息科技有限公司 | A method of realizing the transmission of wisdom commercial circle face characteristic information acquisition terminal data |
| CN112450559A (en) * | 2020-12-09 | 2021-03-09 | 深圳可为创想智能科技有限公司 | Intelligent wearable device for detecting and recognizing human body posture and motion |
| CN113361332A (en) * | 2021-05-17 | 2021-09-07 | 北京中海前沿材料技术有限公司 | Video data acquisition processing method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8519909B2 (en) * | 2011-06-09 | 2013-08-27 | Luis Ricardo Prada Gomez | Multimode input field for a head-mounted display |
| CN106127411A (en) * | 2016-07-04 | 2016-11-16 | 中国科学院深圳先进技术研究院 | A kind of electric operating householder method based on intelligent glasses and system |
| CN106156788A (en) * | 2015-04-24 | 2016-11-23 | 小米科技有限责任公司 | Face identification method, device and intelligent glasses |
| CN106295554A (en) * | 2016-08-08 | 2017-01-04 | 北京文安智能技术股份有限公司 | A kind of face identification method, system and mobile terminal and recognition of face auxiliary device |
-
2017
- 2017-05-11 CN CN201710329207.7A patent/CN107423673A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8519909B2 (en) * | 2011-06-09 | 2013-08-27 | Luis Ricardo Prada Gomez | Multimode input field for a head-mounted display |
| CN106156788A (en) * | 2015-04-24 | 2016-11-23 | 小米科技有限责任公司 | Face identification method, device and intelligent glasses |
| CN106127411A (en) * | 2016-07-04 | 2016-11-16 | 中国科学院深圳先进技术研究院 | A kind of electric operating householder method based on intelligent glasses and system |
| CN106295554A (en) * | 2016-08-08 | 2017-01-04 | 北京文安智能技术股份有限公司 | A kind of face identification method, system and mobile terminal and recognition of face auxiliary device |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135222A (en) * | 2018-02-08 | 2019-08-16 | 上海荟宸信息科技有限公司 | A method of realizing the transmission of wisdom commercial circle face characteristic information acquisition terminal data |
| CN108805091A (en) * | 2018-06-15 | 2018-11-13 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating model |
| CN109886199A (en) * | 2019-02-21 | 2019-06-14 | 百度在线网络技术（北京）有限公司 | Information processing method, device, vehicle and mobile terminal |
| CN109886199B (en) * | 2019-02-21 | 2022-04-12 | 阿波罗智联(北京)科技有限公司 | Information processing method and device, vehicle and mobile terminal |
| CN112450559A (en) * | 2020-12-09 | 2021-03-09 | 深圳可为创想智能科技有限公司 | Intelligent wearable device for detecting and recognizing human body posture and motion |
| CN112450559B (en) * | 2020-12-09 | 2022-03-29 | 深圳可为创想智能科技有限公司 | Intelligent wearable device for detecting and recognizing human body posture and motion |
| CN113361332A (en) * | 2021-05-17 | 2021-09-07 | 北京中海前沿材料技术有限公司 | Video data acquisition processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107423673A (en) | A kind of face identification method and system | |
| US20230276430A1 (en) | Resource scheduling method and apparatus, communication device and storage medium | |
| EP3876140B1 (en) | Method and apparatus for recognizing postures of multiple persons, electronic device, and storage medium | |
| CN108171152A (en) | Deep learning human eye sight estimation method, equipment, system and readable storage medium storing program for executing | |
| CN103945105A (en) | Intelligent photographing and picture sharing method and device | |
| US20160286153A1 (en) | Content capture and transmission | |
| CN107959795B (en) | Information acquisition method, information acquisition equipment and computer readable storage medium | |
| US10038834B2 (en) | Video call method and device | |
| CN108419008B (en) | Shooting method, terminal and computer readable storage medium | |
| CN115835010A (en) | Shooting method and terminal | |
| CN103945001A (en) | Picture sharing method and device | |
| WO2017084295A1 (en) | Automatic focusing method and device | |
| WO2021115483A1 (en) | Image processing method and related apparatus | |
| CN106331569A (en) | Method and system for transforming figure face in instant video picture | |
| CN108898171A (en) | Recognition processing method, system and computer readable storage medium | |
| CN108595600B (en) | Photo classification method, mobile terminal and readable storage medium | |
| CN114168235A (en) | Function switching entry determining method and electronic equipment | |
| US20240147378A1 (en) | Scheduling requests for location data | |
| CN105933023A (en) | Intelligent terminal-based information interaction system and intelligent head-mounted device | |
| CN108513738A (en) | Data transmission method, equipment, machine readable storage medium and system | |
| CN107743198B (en) | Photographing method, terminal and storage medium | |
| CN103186590A (en) | Method for acquiring identity information of wanted criminal on run through mobile phone | |
| CN112437172A (en) | Photographing method, terminal and computer readable storage medium | |
| CN113626710B (en) | Push information generation method, related device and computer program product | |
| CN113286106B (en) | Video recording method, mobile terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| 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: 20171201 |