CN109726680A - Face recognition method, device, system and electronic equipment - Google Patents
Face recognition method, device, system and electronic equipment Download PDFInfo
- Publication number
- CN109726680A CN109726680A CN201811632600.4A CN201811632600A CN109726680A CN 109726680 A CN109726680 A CN 109726680A CN 201811632600 A CN201811632600 A CN 201811632600A CN 109726680 A CN109726680 A CN 109726680A
- Authority
- CN
- China
- Prior art keywords
- face alignment
- comparison
- face
- request
- image
- 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
Abstract
The invention discloses a kind of face recognition method, device, system and electronic equipments, which comprises obtains human face recognition request, human face recognition request includes facial image；And human face recognition request is parsed, obtain multiple face alignment requests；Based on the mapping parameters of multiple face alignment modules, the request of multiple face alignments is distributed to multiple face alignment modules respectively, and obtains each face alignment module for multiple comparison results of each face alignment request feedback；According to the time of multiple multiple comparison results of face alignment module feedback and every similarity size compared between image and facial image, asynchronous recommendation comparison result.Solve technical problem high to human face recognition personnel requirement, human face recognition comparison result accuracy is low existing in the prior art, reached improve human face recognition comparison result accuracy, for ordinary people can Fast Identification face technical effect.
Description
Technical field
Electronic information process field of the present invention, in particular to a kind of face recognition method, device, system and electronics
Equipment.
Background technique
In practical business, people's police generally require to log in different systems, to face to determine the identity of a face
Carry out discrimination comparison, for each system provide as a result, it is artificial to these results carry out comprehensive analysis, to obtain final ratio
To result.Artificial compare can be because of the influence of subjectivity, and the comparison result objectivity of acquisition is low, accuracy is poor, also, artificial ratio
It is high to the requirement to the personnel of comparison, it needs to compare personnel with experience abundant and outstanding judgement and has foot
Enough face alignment technical foundation, could obtain accurate comparison result.For comparison personnel, logs in and different be
System, troublesome in poeration, heavy workload reduce the comparison efficiency of comparison personnel.Therefore, it is quasi- to be badly in need of a kind of human face recognition comparison result
The human face recognition mode that true property is high, is all suitable for for ordinary people.
Summary of the invention
The purpose of the present invention is to provide a kind of face recognition method, device, system and electronic equipment, it is intended to improve existing
There is technical problem low to human face recognition personnel requirement height, human face recognition comparison result accuracy present in technology.
In a first aspect, the embodiment of the invention provides a kind of face recognition methods, comprising:
Human face recognition request is obtained, the human face recognition request includes facial image；
The human face recognition request is parsed, multiple face alignment requests are obtained；
Based on the mapping parameters of multiple face alignment modules, the request of the multiple face alignment is distributed to respectively described more
A face comparison module, and multiple comparison results that each face alignment module is fed back for the request of each face alignment are obtained,
Wherein, each comparison result includes one or multiple compare image；
According to the time of the multiple multiple comparison results of face alignment module feedback and every comparison image with it is described
Similarity size between facial image, it is asynchronous to recommend the comparison result.
Optionally, each face comparison module has a mapping parameters；In the reflecting based on multiple face alignment modules
Parameter is penetrated, the request of the multiple face alignment is distributed to the multiple face alignment module respectively, and obtain each face ratio
Before the multiple comparison results fed back to module for the request of each face alignment, the method also includes:
Establish the corresponding relationship between each face alignment request and each mapping parameters.
Optionally, described to be compared according to the time of the multiple multiple comparison results of face alignment module feedback and every
Similarity size between image and the facial image, it is asynchronous to recommend the comparison result, comprising:
Obtain multiple comparison results of the multiple face alignment module feedback within the scope of setting time；
If each comparison result includes that multiple compare image, for one of comparison knot in the multiple comparison result
The wherein comparison image that fruit includes obtains in multiple comparison images of the multiple comparison result and compares image phase with this
The comparison image of symbol obtains the similarity for comparing image and face alignment request；
Calculate the average value of the similarity；
It is asynchronous to recommend the comparison image according to the size of the average value.
Optionally, described to be compared according to the time of the multiple multiple comparison results of face alignment module feedback and every
Similarity size between image and the facial image, it is asynchronous to recommend the comparison result, further includes:
It obtains in preset time range, multiple comparison results of the multiple face alignment module feedback, it is described each
Comparison result further include with it is described multiple compare that image is corresponding to compare score value；
It is asynchronous to recommend the corresponding comparison image of the comparison score value according to the size for comparing score value.
Optionally, the method also includes: based on the multiple face alignment request and the multiple face alignment mould
The feedback that block requests the multiple face alignment constructs face alignment log.
Optionally, the method also includes: the performance parameter of multiple face alignment modules is obtained according to face alignment log.
Second aspect, the embodiment of the invention provides a kind of human face recognition devices characterized by comprising
Module is obtained, for obtaining human face recognition request, the human face recognition request includes facial image；
Processing module obtains multiple face alignment requests, is based on multiple face ratios for parsing the human face recognition request
To the mapping parameters of module, the request of the multiple face alignment is distributed to the multiple face alignment module respectively, and obtain
Multiple comparison results of each face alignment module for each face alignment request feedback, wherein each comparison result includes
One or multiple comparison images；Compared according to the time of the multiple multiple comparison results of face alignment module feedback and every
It is asynchronous to recommend the comparison result to the similarity size between image and the facial image.
The third aspect, the embodiment of the invention provides a kind of face recognition systems characterized by comprising
First interface, for obtaining human face recognition request, the human face recognition request includes facial image；
Processor obtains multiple face alignment requests for parsing the human face recognition request；Based on multiple face alignments
The mapping parameters of module request the multiple face alignment to be distributed to the multiple face ratio respectively by multiple second interfaces
To module, and each face alignment module is obtained for multiple comparison results of each face alignment request feedback, wherein each
Comparison result includes one or multiple compare image；According to the time of multiple multiple comparison results of face alignment module feedback with
And every similarity size compared between image and the facial image, it is asynchronous to recommend the comparison result；
Multiple face alignment modules, for feeding back multiple comparison results for the request of each face alignment.
Fourth aspect, the embodiment of the invention provides a kind of computer readable storage mediums, are stored thereon with computer journey
Sequence, when which is executed by processor the step of realization any of the above-described the method.
5th aspect, the embodiment of the invention provides a kind of electronic equipment, which is characterized in that including memory, processor
And the computer program that can be run on a memory and on a processor is stored, the processor is realized when executing described program
The step of stating any one the method.
Compared with the prior art, the invention has the following advantages:
The embodiment of the invention provides a kind of face recognition method, device, system and electronic equipments, which comprises
Human face recognition request is obtained, human face recognition request includes facial image；And human face recognition request is parsed, obtain multiple face alignments
Request；Based on the mapping parameters of multiple face alignment modules, the request of multiple face alignments is distributed to multiple face alignments respectively
Module, and each face alignment module is obtained for multiple comparison results of each face alignment request feedback, wherein each ratio
To result include one or multiple compare image；According to the time of multiple multiple comparison results of face alignment module feedback and
Every similarity size compared between image and facial image, asynchronous recommendation comparison result.By the way that multiple face alignments are asked
It asks and is distributed to multiple face alignment modules respectively, obtain each face alignment module for the more of each face alignment request feedback
A comparison result, multiple comparison results can increase the accuracy of human face recognition, multiple according to multiple face alignment module feedbacks
The time of comparison result and every similarity size compared between image and facial image, asynchronous recommendation comparison result, root
It is big according to the similarity between the time and comparison image and facial image of multiple multiple comparison results of face alignment module feedback
Comparison result that is small, recommending human face recognition accuracy high, improves the reliability of the comparison result of recommendation, opens up in a recommended manner
Existing comparison result, can be according to the comparison result Fast Identification face of recommendation, using asynchronous recommendation for ordinary person
Mode can quickly respond human face recognition request.It solves existing in the prior art to face
The technical problem that personnel requirement is high, human face recognition comparison result accuracy is low is recognized, raising human face recognition comparison result has been reached
Accuracy, for ordinary people can Fast Identification face technical effect.
Other feature and advantage of the embodiment of the present invention will illustrate in subsequent specification, also, partly from specification
In become apparent, or by implement understanding of the embodiment of the present invention.The objectives and other advantages of the invention can be by institute
Specifically noted structure is achieved and obtained in specification, claims and the attached drawing write.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows a kind of frame structure schematic diagram of face recognition system 100 provided in an embodiment of the present invention.
Fig. 2 shows a kind of flow charts of face recognition method provided in an embodiment of the present invention.
Fig. 3 shows a kind of frame structure schematic diagram of human face recognition device 200 provided in an embodiment of the present invention.
Fig. 4 shows the frame structure schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
The embodiment of the invention provides the embodiment of the invention provides a kind of face recognition method, device, system and electronics
Equipment is existing in the prior art low to human face recognition personnel requirement height, human face recognition comparison result accuracy to solve
The technical issues of.
Embodiment
The embodiment of the invention provides a kind of face recognition systems 100, including first interface 110 as shown in Figure 1, processing
Device 120, second interface 130 and multiple face alignment modules 140.First interface 110 and second interface 130 and processor 120 are logical
Connection is crossed, multiple face alignment modules 140 are connect with second interface 130.As an alternative embodiment, human face recognition system
System 100 includes multiple second interfaces 130, and multiple second interfaces 130 connect with processor 120, each second interface 130 respectively with
Multiple face alignment modules 140 connect.Specifically, first interface 110, for obtaining human face recognition request, human face recognition request
Including facial image.Processor 120 obtains multiple face alignment requests for parsing human face recognition request；Based on multiple faces
Multiple face alignments are requested to be distributed to respectively by multiple second interfaces 130130 the multiple by the mapping parameters of comparison module
Face alignment module 140, and multiple comparison results that each face alignment module is fed back for the request of each face alignment are obtained,
Wherein, each comparison result includes one or multiple compare image；According to the multiple comparison knots of multiple face alignment module feedbacks
The time of fruit and every similarity size compared between image and facial image, asynchronous recommendation comparison result；Multiple faces
Comparison module, for feeding back multiple comparison results for the request of each face alignment.
In embodiments of the present invention, each face comparison module 140 calls the format of face alignment request different, therefore,
Face alignment request is converted respectively by multiple second interfaces 130, so that multiple face alignment modules 140 can be directed to face ratio
Multiple comparison results are fed back to request.The format for the comparison result that each face comparison module 140 is fed back may be different, therefore need
Multiple second interfaces 130 are wanted to convert the format of these comparison results, so that this can be uniformly processed in the processor 120
A little comparison results.By the way that the request of multiple face alignments is distributed to multiple face alignment modules respectively, each face alignment is obtained
Multiple comparison results that module is fed back for the request of each face alignment, multiple comparison results can increase the accurate of human face recognition
Property, according between the time of multiple multiple comparison results of face alignment module feedback and every comparison image and facial image
Similarity size, asynchronous recommendation comparison result, according to the time of multiple multiple comparison results of face alignment module feedback and ratio
To the similarity size between image and facial image, the comparison result for recommending human face recognition accuracy high improves recommendation
The reliability of comparison result, shows comparison result in a recommended manner, can be according to the comparison of recommendation for ordinary person
As a result Fast Identification face can quickly respond human face recognition request using the asynchronous way of recommendation.Solves the prior art
Present in technology high to human face recognition personnel requirement, that human face recognition comparison result accuracy is low existing in the prior art ask
Topic, reached improve human face recognition comparison result accuracy, for ordinary people can Fast Identification face technical effect.
For a kind of face recognition system provided by the above embodiment, the embodiment of the present application is also corresponding to be provided a kind of face and distinguishes
Knowledge method is applied to above system, including S100~S400 as shown in Figure 2, below in conjunction with Fig. 2 it is right~S400 is illustrated.
S100: obtaining human face recognition request, and human face recognition request includes facial image.
S200: parsing human face recognition request obtains multiple face alignment requests.
S300: the request of multiple face alignments is distributed to multiple by the mapping parameters based on multiple face alignment modules respectively
Face alignment module, and multiple comparison results that each face alignment module is fed back for the request of each face alignment are obtained,
In, each comparison result includes one or multiple compare image.
S400: according to the time of multiple multiple comparison results of face alignment module feedback and every comparison image and face
Similarity size between image, asynchronous recommendation comparison result.
It is obtained by using above scheme by the way that the request of multiple face alignments is distributed to multiple face alignment modules respectively
Each face alignment module is obtained for multiple comparison results of each face alignment request feedback, multiple comparison results can increase
The accuracy of human face recognition, according to the time of multiple multiple comparison results of face alignment module feedback and every comparison image with
Similarity size between facial image, asynchronous recommendation comparison result are tied according to the multiple comparisons of multiple face alignment module feedbacks
Similarity size between the time and comparison image and facial image of fruit, the comparison knot for recommending human face recognition accuracy high
Fruit improves the reliability of the comparison result of recommendation, shows comparison result in a recommended manner, can be with for ordinary person
Human face recognition request can be quickly responded using the asynchronous way of recommendation according to the comparison result Fast Identification face recommended.
Solve, human face recognition comparison result standard high to human face recognition personnel requirement existing in the prior art
The true low technical problem of property, has reached raising human face recognition comparison result accuracy, can Fast Identification people for ordinary people
The technical effect of face.
The format for the human face recognition request that first interface 110 obtains in embodiments of the present invention is unsatisfactory for setting call format
When, the method also includes standardization first interfaces 110 to obtain the human face recognition request that format meets setting call format.People
Face identification request may include facial image and preset solicited message.
In embodiments of the present invention, face alignment module 140 calls the format of face alignment request different, it is therefore desirable to will
Human face recognition request is parsed, multiple face alignment requests are obtained, each face comparison module 140 there are a mapping parameters, each
Face alignment request and each face comparison module 140 corresponding relationship can be it is preconfigured, be also possible to S300 it
Before, the corresponding relationship between each face alignment request and each mapping parameters is established, and then obtain each face alignment request
With the corresponding relationship of each face comparison module 140, it is only necessary to according between the request of each face alignment and each mapping parameters
Corresponding relationship distribution face alignment request, can by each face alignment request be distributed to corresponding face alignment module 140
In.Before face alignment request is distributed to corresponding face alignment module 140, need to distinguish by multiple second interfaces 130
Face alignment request is converted, so that multiple face alignment modules 140 can feed back multiple comparison results for face alignment request.
Specifically, the corresponding plugin interface of each face comparison module 140, processor 120 pass through the request of multiple face alignments
When multiple second interfaces 130 are distributed to multiple face alignment modules 140 respectively, the specific can be that, call each face alignment
The corresponding plugin interface of module 140, and then the request of multiple face alignments is distributed to each face comparison module 140.
As a kind of optional embodiment of S400, specifically: obtaining multiple face alignments within the scope of setting time
Multiple comparison results of module feedback；If each comparison result includes that multiple compare image, for its in multiple comparison results
In a comparing result wherein comparison image that includes, multiple in multiple comparison results compare in images and obtain and the ratio
To the comparison image that image is consistent, the similarity for comparing image and face alignment request is obtained；Calculate the average value of similarity；Root
According to the size of average value, asynchronous recommendation compares image.Wherein according to the size of average value, asynchronous recommendation compares image, specifically:
According to the size of average value, descending sequence is carried out to average value, selectivity recommends to compare image, specifically, preferential recommendation
The comparison image that average value sorts forward.
For example, there is 3 face comparison modules, 3 face comparison modules are face alignment modules A, face alignment mould B respectively
A human face recognition request is obtained, is obtained after the human face recognition request analysis within the scope of setting time with face alignment module C
The multiple face alignments request obtained is face alignment request 1, face alignment request 2 and face alignment request 3 respectively, based on face
The mapping parameters of comparison module A, face alignment mould B and face alignment module C, face alignment request 1, face alignment request 2 and
Face alignment requests 3 corresponding with face alignment modules A, face alignment mould B and face alignment module C respectively, face alignment modules
A, face alignment mould B and face alignment module C is directed to face alignment request 1, face alignment request 2 and face alignment request respectively
3 feedback comparison results 1, comparison result 2 and comparison result 3.Wherein, comparison result 1 includes 3 records, and 3 records are note respectively
A, record b and record c are recorded, recording a, record b and recording the value of c is 80%, 70% and 69% respectively, and comparison result 2 includes
3 record respectively be record d, record e and record f, record d, record e and record f value be 90%, 88% and respectively
70%.Wherein, which indicates the similarity of comparison image and facial image in the record.For example, 3 of comparison result 3
Record is record g, record h and record i respectively, and record g, record h and record i value are 92%, 88% and 69% respectively.Every
Record includes between comparison image, the comparison image and facial image for the facial image for including in human face recognition request
Similarity, i.e. the comparison score value of the comparison image, and the identity information of the face in image is compared, identity information includes surname
The information such as name, gender, age, identification card number.For comparison result 1, it is opposite with the comparison score value recorded in g to record a, record d
It answers, i.e. record d and the comparison score value recorded in g are corresponding with the comparison image recorded in a.Likewise, record b, record e and note
The comparison score value recorded in h is corresponding, i.e. record e and the comparison score value recorded in d are corresponding with the comparison image recorded in b, note
The comparison score value recorded in c, record f and record i is corresponding, i.e. the comparison compared in score value and record c in record f and record f
Image is corresponding.
For the record a of comparison result 1, the average value of the comparison image in record a is calculated are as follows: (80%+90%+
92%)/3=87.33%.Similarly, the average value of the comparison image in record b is calculated are as follows: (70%+88%+88%)/3=
82%.Similarly, the average value of the comparison image in record c is calculated are as follows: (69%+70%+69%)/3=69.33%.To note
Record a, record b and the average value for recording c are ranked up according to descending sequence, obtain 87.33%, 82% and 69.33%
Order, recommend record in the top, i.e. recommendation record a.If obtaining multiple human face recognitions within the scope of setting time and asking
It asks, then according to the request time of face identification request, the first human face recognition of preferential recommendation request time requests corresponding comparison
As a result.It is logical to use above scheme, the comparison result preferential recommendation of human face recognition high reliablity can be improved raising face and be distinguished
Know comparison result accuracy, layman can be according to the comparison result Fast Identification face of recommendation for ordinary people.Together
When, using the asynchronous way of recommendation, override requests, which preferentially obtain, recommends comparison result, shortens the waiting time of user, improves
User experience.
As the optional embodiment of another kind of S400, specifically: obtaining in preset time range, multiple face ratios
To multiple comparison results of module feedback, each comparison result further includes comparing that image is corresponding to compare score value with multiple；Root
According to the size for comparing score value, asynchronous recommendation compares the corresponding comparison image of score value.Specifically, according to every in each comparison result
The value of item record, i.e. comparison score value recommend whole every records in each comparison result.Specifically, to all comparison knots
Daily record in fruit is ranked up according to the size of value, recommends the forward record that sorts.Such as comparison result 1 includes 3
Record, 3 records are to record a, record b and record c respectively, and the value of record a, record b and record c are 80%, 70% respectively
With 69%, 3 records that comparison result 2 includes are record d, record e and record f respectively, record d, record e and record taking for f
Value is 90%, 88% and 70% respectively.Wherein, which indicates the similarity of comparison image and facial image in the record.
For example, 3 records of comparison result 3 are record g, record h and record i respectively, record g, record h and record i value are respectively
92%, 88% and 69%.If setting recommend 3 probabilities, the comparison result finally recommended include record g, record a, record e or
Person records h.
When the similarity of the comparison result of multiple face alignment module feedbacks is all to reach setting value, according to similarity pair
The comparison result that part face comparison module in multiple face alignment modules returns is recommended.
By using above scheme, the accuracy of human face recognition comparison result is improved.By multiple face alignment module collection
Cheng Yi system, by system set face alignment module to the comparison result for human face recognition request feedback, simple side
Just.
As an alternative embodiment, face recognition method further includes, requested based on multiple face alignments and more
The feedback that a face comparison module requests multiple face alignments constructs face alignment log；According to face alignment log
The performance parameter of multiple face alignment modules is obtained, performance parameter characterizes the quality of the performance of face alignment module.Wherein, face
The performance parameter of comparison module includes frequency of use, average response time, success response rate etc..Specifically, recording each face
Comparison module obtains human face recognition request every time, and for the comparison result of human face recognition request feedback, and most pusher
The similarity for the result recommended, the success response rate of the face comparison module is portrayed with the similarity, and record face comparison module is rung
The time of corresponding human face recognition request is answered, and records the number for calling the plugin of the face comparison module, is obtained according to the number
Obtain the frequency of use of the face comparison module.
By using above scheme, the performance of multiple face alignment modules can be monitored, for comparing personnel when needed
The good face alignment module of selection performance carries out face and distinguishes.
In conclusion the embodiment of the invention provides a kind of face recognition methods, which comprises obtain human face recognition
Request, human face recognition request include facial image；And human face recognition request is parsed, obtain multiple face alignment requests；Based on more
The request of multiple face alignments is distributed to multiple face alignment modules respectively, and obtained by the mapping parameters of a face comparison module
Multiple comparison results of each face alignment module for each face alignment request feedback, wherein each comparison result includes
One or multiple comparison images；According to the time of multiple multiple comparison results of face alignment module feedback and every comparison chart
Picture and the similarity size between facial image, asynchronous recommendation comparison result.By the way that the request of multiple face alignments is distributed respectively
At most a face comparison module obtains each face alignment module for multiple comparison knots of each face alignment request feedback
Fruit, multiple comparison results can increase the accuracy of human face recognition, according to multiple multiple comparison results of face alignment module feedback
Time and every similarity size compared between image and facial image, asynchronous recommendation comparison result, according to multiple people
Face comparison module feeds back the time of multiple comparison results and compares the similarity size between image and facial image, referrer
Face recognizes the high comparison result of accuracy, improves the reliability of the comparison result of recommendation, shows comparison knot in a recommended manner
Fruit, can be according to the comparison result Fast Identification face of recommendation, using the asynchronous way of recommendation for ordinary person, can be with
Quickly respond human face recognition request.It solves and existing in the prior art human face recognition personnel is wanted
Seek technical problem high, that human face recognition comparison result accuracy is low, reached raising human face recognition comparison result accuracy, for
Ordinary people can Fast Identification face technical effect.
A kind of face recognition method is provided for above-described embodiment, the also corresponding one kind that provides of the embodiment of the present application is for executing
The executing subject of above-mentioned step, the executing subject can be human face recognition device 200 in Fig. 3.Referring to FIG. 3, the device packet
It includes:
Module 210 is obtained, for obtaining human face recognition request, the human face recognition request includes facial image；
Processing module 220 obtains multiple face alignment requests, is based on multiple people for parsing the human face recognition request
The request of the multiple face alignment is distributed to the multiple face alignment module by the mapping parameters of face comparison module respectively, and
Each face alignment module is obtained for multiple comparison results of each face alignment request feedback, wherein each comparison result
Including one or multiple compare image；According to the time of the multiple multiple comparison results of face alignment module feedback and often
The similarity size compared between image and the facial image is opened, it is asynchronous to recommend the comparison result.
As an alternative embodiment, processing module 220 is specifically used for: each face comparison module has a mapping
Parameter；In the mapping parameters based on multiple face alignment modules, the request of the multiple face alignment is distributed to institute respectively
Multiple face alignment modules are stated, and obtain each face alignment module for multiple comparison knots of each face alignment request feedback
Before fruit, the method also includes: establish the corresponding relationship between each face alignment request and each mapping parameters.
As an alternative embodiment, processing module 220 is specifically used for: it obtains within the scope of setting time, it is described
Multiple comparison results of multiple face alignment module feedbacks；If each comparison result includes that multiple compare image, for described more
The wherein comparison image that one of comparing result in a comparison result includes, in multiple of the multiple comparison result
It compares to obtain in image and compares the comparison image that image is consistent with this, the acquisition comparison image is requested similar to face alignment
Degree；Calculate the average value of the similarity；It is asynchronous to recommend the comparison image according to the size of the average value.
As an alternative embodiment, processing module 220 is specifically also used to: obtaining in preset time range, institute
State multiple comparison results of multiple face alignment module feedbacks, each comparison result further include with it is described multiple compare image
Corresponding comparison score value；It is asynchronous to recommend the corresponding comparison image of the comparison score value according to the size for comparing score value.
As an alternative embodiment, processing module 220 is specifically also used to: being requested based on the multiple face alignment
And the feedback that the multiple face alignment module requests the multiple face alignment, construct face alignment log.
As an alternative embodiment, processing module 220 is specifically also used to: being obtained according to face alignment log multiple
The performance parameter of face alignment module.
About the device in above-described embodiment, wherein modules execute the concrete mode of operation in related this method
Embodiment in be described in detail, no detailed explanation will be given here.
The embodiment of the invention also provides a kind of electronic equipment, as shown in figure 4, include memory 504, processor 502 and
It is stored in the computer program that can be run on memory 504 and on processor 502, the processor 502 executes described program
The step of either Shi Shixian face recognition method described previously method.
Wherein, in Fig. 4, bus architecture (is represented) with bus 500, and bus 500 may include any number of interconnection
Bus and bridge, bus 500 will include the one or more processors represented by processor 502 and what memory 504 represented deposits
The various circuits of reservoir link together.Bus 500 can also will peripheral equipment, voltage-stablizer and management circuit etc. it
Various other circuits of class link together, and these are all it is known in the art, therefore, no longer further retouch to it herein
It states.Bus interface 505 provides interface between bus 500 and receiver 501 and transmitter 503.Receiver 501 and transmitter
503 can be the same element, i.e. transceiver, provide the unit for communicating over a transmission medium with various other devices.Place
It manages device 502 and is responsible for management bus 500 and common processing, and memory 504 can be used for storage processor 502 and execute behaviour
Used data when making.
The embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, the journey
The step of either face recognition method described previously method is realized when sequence is executed by processor.
Algorithm and display are not inherently related to any particular computer, virtual system, or other device provided herein.
Various general-purpose systems can also be used together with teachings based herein.As described above, it constructs required by this kind of system
Structure be obvious.In addition, the present invention is also not directed to any particular programming language.It should be understood that can use various
Programming language realizes summary of the invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention: i.e. required to protect
Shield the present invention claims features more more than feature expressly recited in each claim.More precisely, as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim itself
All as a separate embodiment of the present invention.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments in this include institute in other embodiments
Including certain features rather than other feature, but the combination of the feature of different embodiment means in the scope of the present invention
Within and form different embodiments.For example, in the following claims, embodiment claimed it is any it
One can in any combination mode come using.
Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors
Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice
Microprocessor or digital signal processor (DSP) realize some or all portions in device according to an embodiment of the present invention
The some or all functions of part.The present invention is also implemented as a part or complete for executing method as described herein
The device or device program (for example, computer program and computer program product) in portion.It is such to realize program of the invention
It can store on a computer-readable medium, or may be in the form of one or more signals.Such signal can be with
It downloads from internet website, is perhaps provided on the carrier signal or is provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.
Claims (10)
1. a kind of face recognition method characterized by comprising
Human face recognition request is obtained, the human face recognition request includes facial image；
The human face recognition request is parsed, multiple face alignment requests are obtained；
Based on the mapping parameters of multiple face alignment modules, the request of the multiple face alignment is distributed to the multiple people respectively
Face comparison module, and each face alignment module is obtained for multiple comparison results of each face alignment request feedback, wherein
Each comparison result includes one or multiple compare image；
According to the time of the multiple multiple comparison results of face alignment module feedback and every comparison image and the face
Similarity size between image, it is asynchronous to recommend the comparison result.
2. the method according to claim 1, wherein each face comparison module has a mapping parameters；
In the mapping parameters based on multiple face alignment modules, the request of the multiple face alignment is distributed to respectively described
Multiple face alignment modules, and each face alignment module is obtained for multiple comparison results of each face alignment request feedback
Before, the method also includes:
Establish the corresponding relationship between each face alignment request and each mapping parameters.
3. according to the method described in claim 2, it is characterized in that, described multiple according to the multiple face alignment module feedback
The time of comparison result and every similarity size compared between image and the facial image, it is asynchronous to recommend the comparison
As a result, comprising:
Obtain multiple comparison results of the multiple face alignment module feedback within the scope of setting time；
If each comparison result includes that multiple compare image, for one of comparing result packet in the multiple comparison result
The wherein comparison image included obtains in multiple comparison images of the multiple comparison result and compares what image was consistent with this
Image is compared, the similarity for comparing image and face alignment request is obtained；
Calculate the average value of the similarity；
It is asynchronous to recommend the comparison image according to the size of the average value.
4. according to the method described in claim 2, it is characterized in that, described multiple according to the multiple face alignment module feedback
The time of comparison result and every similarity size compared between image and the facial image, it is asynchronous to recommend the comparison
As a result, further includes:
It obtains in preset time range, multiple comparison results of the multiple face alignment module feedback, each comparison
Result further include with it is described multiple compare that image is corresponding to compare score value；
It is asynchronous to recommend the corresponding comparison image of the comparison score value according to the size for comparing score value.
5. the method according to claim 1, wherein the method also includes:
The multiple face alignment is requested based on the request of the multiple face alignment and the multiple face alignment module
Feedback constructs face alignment log.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
The performance parameter of multiple face alignment modules is obtained according to face alignment log.
7. a kind of human face recognition device characterized by comprising
Module is obtained, for obtaining human face recognition request, the human face recognition request includes facial image；
Processing module obtains multiple face alignment requests, is based on multiple face alignment moulds for parsing the human face recognition request
The request of the multiple face alignment is distributed to the multiple face alignment module respectively, and obtained each by the mapping parameters of block
Multiple comparison results of the face alignment module for each face alignment request feedback, wherein each comparison result includes one
Or multiple compare image；According to the time of the multiple multiple comparison results of face alignment module feedback and every comparison chart
Picture and the similarity size between the facial image, it is asynchronous to recommend the comparison result.
8. a kind of face recognition system characterized by comprising
First interface, for obtaining human face recognition request, the human face recognition request includes facial image；
Processor obtains multiple face alignment requests for parsing the human face recognition request；Based on multiple face alignment modules
Mapping parameters, the multiple face alignment is requested by multiple second interfaces to be distributed to the multiple face alignment mould respectively
Block, and each face alignment module is obtained for multiple comparison results of each face alignment request feedback, wherein each comparison
It as a result include one or multiple comparison images；According to the time of multiple multiple comparison results of face alignment module feedback and often
The similarity size compared between image and the facial image is opened, it is asynchronous to recommend the comparison result；
Multiple face alignment modules, for feeding back multiple comparison results for the request of each face alignment.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor
The step of any one of claim 1-6 the method is realized when row.
10. a kind of electronic equipment, which is characterized in that on a memory and can be in processor including memory, processor and storage
The computer program of upper operation, the processor realize the step of any one of claim 1-6 the method when executing described program
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811632600.4A CN109726680A (en) | 2018-12-28 | 2018-12-28 | Face recognition method, device, system and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811632600.4A CN109726680A (en) | 2018-12-28 | 2018-12-28 | Face recognition method, device, system and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109726680A true CN109726680A (en) | 2019-05-07 |
Family
ID=66296061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811632600.4A Pending CN109726680A (en) | 2018-12-28 | 2018-12-28 | Face recognition method, device, system and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109726680A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488478A (en) * | 2015-12-02 | 2016-04-13 | 深圳市商汤科技有限公司 | Face recognition system and method |
| CN205451095U (en) * | 2015-12-02 | 2016-08-10 | 深圳市商汤科技有限公司 | A face -identifying device |
| CN108446692A (en) * | 2018-06-08 | 2018-08-24 | 南京擎华信息科技有限公司 | Face comparison method, device and system |
| CN108781400A (en) * | 2017-01-06 | 2018-11-09 | 联发科技股份有限公司 | Demand system information transmits process |
| CN108920974A (en) * | 2018-05-23 | 2018-11-30 | 华迪计算机集团有限公司 | A kind of application authorization search method and system for supporting multi-layer |
| CN108960145A (en) * | 2018-07-04 | 2018-12-07 | 北京蜂盒科技有限公司 | Facial image detection method, device, storage medium and electronic equipment |
| CN109033299A (en) * | 2018-07-16 | 2018-12-18 | 深圳市谷熊网络科技有限公司 | It is a kind of by terminal applies to the method, device and equipment of user's recommendation information |
-
2018
- 2018-12-28 CN CN201811632600.4A patent/CN109726680A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488478A (en) * | 2015-12-02 | 2016-04-13 | 深圳市商汤科技有限公司 | Face recognition system and method |
| CN205451095U (en) * | 2015-12-02 | 2016-08-10 | 深圳市商汤科技有限公司 | A face -identifying device |
| CN108781400A (en) * | 2017-01-06 | 2018-11-09 | 联发科技股份有限公司 | Demand system information transmits process |
| CN108920974A (en) * | 2018-05-23 | 2018-11-30 | 华迪计算机集团有限公司 | A kind of application authorization search method and system for supporting multi-layer |
| CN108446692A (en) * | 2018-06-08 | 2018-08-24 | 南京擎华信息科技有限公司 | Face comparison method, device and system |
| CN108960145A (en) * | 2018-07-04 | 2018-12-07 | 北京蜂盒科技有限公司 | Facial image detection method, device, storage medium and electronic equipment |
| CN109033299A (en) * | 2018-07-16 | 2018-12-18 | 深圳市谷熊网络科技有限公司 | It is a kind of by terminal applies to the method, device and equipment of user's recommendation information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111680124B (en) | RPA-based large-scale custom customer demand acquisition and query method | |
| CN110782318A (en) | Marketing method and device based on audio interaction and storage medium | |
| CN107741976B (en) | Intelligent response method, device, medium and electronic equipment | |
| US20200097608A1 (en) | Method and system for service agent assistance of article recommendations to a customer in an app session | |
| CN109165974A (en) | A kind of commercial product recommending model training method, device, equipment and storage medium | |
| CN107800572B (en) | Method and device for upgrading equipment based on neural network | |
| CN107910060A (en) | Method and apparatus for generating information | |
| CN109062807B (en) | Method and device for testing application program, storage medium and electronic device | |
| CN109447851A (en) | One kind prepares for a meal, vegetable providing method, device and equipment | |
| CN107977852A (en) | A kind of intelligent sound purchase guiding system and method | |
| CN108550057A (en) | It attends a banquet request processing method, electronic device, the computer readable storage medium of answering questions | |
| CN113191479A (en) | Method, system, node and storage medium for joint learning | |
| CN107977678A (en) | Method and apparatus for output information | |
| CN109032920A (en) | A kind of method and device for being tested protocol interface | |
| CN109819002A (en) | Data push method and device, storage medium and electronic device | |
| CN113450796B (en) | Voice report generation method, device, equipment and storage medium | |
| CN106356056A (en) | Speech recognition method and device | |
| CN111651989B (en) | Named entity recognition method and device, storage medium and electronic device | |
| TW202121206A (en) | Method and system for automatically identifying valid data acquisition module | |
| CN109726680A (en) | Face recognition method, device, system and electronic equipment | |
| CN110532448A (en) | Document Classification Method, device, equipment and storage medium neural network based | |
| CN111259698A (en) | Method and device for acquiring image | |
| KR20200020127A (en) | Method and apparatus of producing augmented reality contents | |
| CN110163043B (en) | Face detection method, device, storage medium and electronic device | |
| CN110729032A (en) | Information processing method and device and computer storage medium |
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: 20190507 |
|
| RJ01 | Rejection of invention patent application after publication |