CN109145720A - A kind of face identification method and device - Google Patents
A kind of face identification method and device Download PDFInfo
- Publication number
- CN109145720A CN109145720A CN201810735594.9A CN201810735594A CN109145720A CN 109145720 A CN109145720 A CN 109145720A CN 201810735594 A CN201810735594 A CN 201810735594A CN 109145720 A CN109145720 A CN 109145720A
- Authority
- CN
- China
- Prior art keywords
- image
- target
- facial image
- organic
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
This application discloses a kind of face identification method and devices, this method comprises: identify to target facial image, obtain the corresponding organic image of multiple organs in the target facial image；Obtained organic image is compared with the organic image of target user respectively, obtains the similarity assessment index of the multiple organ；According to the similarity assessment index and corresponding weight of the multiple organ, the target facial image and the overall similarity evaluation index of the target user are obtained；The recognition result to the target facial image is determined according to the overall similarity evaluation index of the target user.
Description
Technical field
This application involves technical field of face recognition more particularly to a kind of face identification methods and device.
Background technique
With the development of face recognition technology, the application scenarios using recognition of face as authentication are more and more.Face
Identification is favored with the advantages that highly-safe, easy to use by user, and the fields such as attendance, safety check are widely used in.
Recognition of face is usually that the facial image of acquisition is done similarity with the facial image prestored to compare, to user's
Identity is verified, however this method usually requires to collect complete facial image, if collected facial image is endless
It is whole, identification can not be quickly made, for example, for the user etc. of the user or wear dark glasses that wear masks, because not acquiring whole person
Face image and can not quickly make identification.
Summary of the invention
The embodiment of the present application provides a kind of face identification method and device, for solving because not acquiring complete facial image
And the problem of identification can not be quickly made to user.
The embodiment of the present application adopts the following technical solutions:
In a first aspect, providing a kind of face identification method, comprising: identify, obtain described to target facial image
The corresponding organic image of multiple organs in target facial image；By obtained organic image respectively with the organic image of target user
It compares, obtains the similarity assessment index of the multiple organ；According to the similarity assessment index of the multiple organ and
Corresponding weight obtains the target facial image and the overall similarity evaluation index of the target user；According to the mesh
Mark the determining recognition result to the target facial image of overall similarity evaluation index of user.
Second aspect provides a kind of face identification device, comprising: organic image obtains module, to target facial image
It is identified, obtains the corresponding organic image of multiple organs in the target facial image；Organ similarity obtains module, will
To organic image compared respectively with the organic image of target user, the similarity assessment for obtaining the multiple organ refers to
Mark；Overall similarity obtains module and obtains the mesh according to the similarity assessment index and corresponding weight of the multiple organ
Mark the overall similarity evaluation index of facial image and the target user；Recognition result determining module is used according to the target
The overall similarity evaluation index at family determines the recognition result to the target facial image.
The third aspect provides a kind of electronic equipment, comprising: memory, processor and is stored on the memory simultaneously
The computer program that can be run on the processor realizes following behaviour when the computer program is executed by the processor
Make: target facial image is identified, obtains the corresponding organic image of multiple organs in the target facial image；It will obtain
Organic image compared respectively with the organic image of target user, obtain the similarity assessment index of the multiple organ；
According to the similarity assessment index and corresponding weight of the multiple organ, obtains the target facial image and the target is used
The overall similarity evaluation index at family；It is determined according to the overall similarity evaluation index of the target user to the target face
The recognition result of image.
Fourth aspect provides a kind of computer readable storage medium, is stored on the computer readable storage medium
Computer program realizes following operation: identifying, obtain to target facial image when the computer program is executed by processor
The corresponding organic image of multiple organs into the target facial image；By obtained organic image respectively with the device of target user
Official's image compares, and obtains the similarity assessment index of the multiple organ；According to the similarity assessment of the multiple organ
Index and corresponding weight obtain the target facial image and the overall similarity evaluation index of the target user；According to
The overall similarity evaluation index of the target user determines the recognition result to the target facial image.
The above-mentioned technical proposal that this specification embodiment provides is carried out between organic image by being based in recognition of face
Comparison, and then the similarity assessment index based on organic image obtains the overall similarity evaluation index of facial image, relative to
Based on the way of contrast of entire facial image, identification can also be quickly made even if collected target facial image is imperfect,
It solves the problems, such as that quickly identification can not be made to user because not acquiring complete facial image.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen
Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is the flow diagram for the face identification method that one embodiment of this specification provides；
Fig. 2 is the flow diagram for the face identification method that another embodiment of this specification provides；
Fig. 3 is the flow diagram for the face identification method that the further embodiment of this specification provides；
Fig. 4 is the structural schematic diagram for the face identification device that one embodiment of this specification provides；
Fig. 5 is the concrete structure schematic diagram for the electronic equipment that one embodiment of this specification provides.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with the application specific embodiment and
Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one
Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall in the protection scope of this application.
As shown in Figure 1, one embodiment of this specification provides a kind of face identification method, for solving in the prior art
The problem of quickly identification can not be made to user because not acquiring complete facial image, as described in Figure 1, the embodiment include such as
Lower step:
S110: identifying target facial image, obtains the corresponding organ of multiple organs in the target facial image
Image.
Target facial image in the embodiment, it can be understood as be the facial image of collected user to be identified.
The step identifies target facial image, i.e., carries out organ identification to collected target facial image, from
And the corresponding organic image of multiple organs is obtained, for example, obtaining eye image, nose image, mouth image and ear image etc..
It should be noted that the corresponding organic image of above-mentioned multiple organs, it can be understood as be the office of target facial image
Portion, with significant feature so as to subgraph for identification, " five be not limited solely in target facial image
The corresponding image of official ", or even can also be the face mask of the image of the chin area of target facial image, target facial image
Image, the birthmark in target facial image or black mole image etc..
It when executing step S110, can be carried out, i.e., sample facial image be carried out first based on the method for model training
Organic region calibration, obtains organ site identification model, is then based on the step of organ site identification model executes S110.When
So, it should be appreciated that step S110 can also realize in other manners, this specification embodiment to this with no restriction.
S120: obtained organic image is compared with the organic image of target user respectively, obtains the multiple device
The similarity assessment index of official.
The organic image of target user in the embodiment specifically can be the use of pre-stored determining identity
The organic image at family, target user can be a user, under the scene suitable for core body；Target user is also possible to multiple use
Family, suitable under the scene of retrieval.
The organic image of target user is corresponding with the organic image of target facial image, is specifically also possible to target user
Eye image, nose image, mouth image and ear image etc..
When the step specifically carries out organic image comparison, the organic image of homolog can be compared respectively, example
Such as, the eye image of the eye image of target facial image and target user is compared；By the nose of target facial image
The nose image of image and target user compare etc., finally obtain the similarity assessment index of above-mentioned multiple organs, example
Such as, the similarity assessment index of eyes, similarity assessment index of nose etc. are respectively obtained.
Above-mentioned similarity assessment index can be used to characterize the similarity degree between two corresponding organic images, specifically
It can be the Euclidean distance between two corresponding organic images (characteristics of image)；It is also possible to carry out above-mentioned Euclidean distance
Obtained numerical value after normalization, such as normalize to 0-100, wherein 0 indicates that similarity degree is minimum, and 100 be similarity degree
Highest.
S130: according to the similarity assessment index and corresponding weight of the multiple organ, the target face figure is obtained
The overall similarity evaluation index of picture and the target user.
Before the embodiment executes, the weight of each organ can be preset, it, can will be some important when specific setting
, the weight for being easy to distinguish the organ of different user setting it is relatively higher, such as eyes；On the contrary, to some secondary organs
Weight setting it is relatively lower, such as nose.
Optionally, step S130 can be executed using following formula:
Overall similarity evaluation index=∑ (organ weight × organ similarity assessment index)/organ number
Wherein, the quantity of organ is multiple, and each organ is provided with weight.
Certainly, for the calculating of overall similarity evaluation index, other mathematical formulaes can also be used, such as to multiple organs
Similarity assessment index directly sum it up and then average, be the equal of that the weight of each organ in multiple organs is equal.
S140: the identification to the target facial image is determined according to the overall similarity evaluation index of the target user
As a result.
Specifically, if target user is a user, and the overall similarity evaluation index of target user is greater than or waits
In preset threshold, determination is identified as function to the target facial image；Or
If target user is a user, and the overall similarity evaluation index of target user is less than preset threshold, really
Determine to the target facial image recognition failures；Or
If target user be multiple users, and in multiple users at least one user overall similarity evaluation index it is big
It is determining that function is identified as to the target facial image in or equal to preset threshold；Or
If target user is multiple users, and the overall similarity evaluation index of multiple users is respectively less than preset threshold,
It determines to the target facial image recognition failures.
The face identification method provided by this specification embodiment, by being based between organic image in recognition of face
It compares, and then the similarity assessment index based on organic image obtains the overall similarity evaluation index of facial image, phase
For the way of contrast based on entire facial image, knowledge can also be quickly made even if collected target facial image is imperfect
Not, it solves the problems, such as that quickly identification can not be made to user because not acquiring complete facial image.
The organic image of target user is mentioned in the S120 of above-described embodiment, wherein the organic image of target user can
To be to be obtained in advance to the facial image progress organ identification of target user, segmentation, in this way, when above-described embodiment executes
The organic image for directly acquiring above-mentioned target user carries out organ identification and segmentation without the facial image in real time to target user
The processing such as processing, thus recognition efficiency when improving recognition of face.Therefore, as a preferred embodiment, in above-described embodiment
Step S110 before, can also include following operating procedure:
Establish facial image database, wherein include the facial image for the user that a large amount of identity determines in facial image database；
Facial image in the facial image database is identified, organic image library is obtained, that is, is directed to facial image database
In each user, the organic image of multiple organs of the available user.The realization process of the step implementation procedure can
Referring to the step S110 of above-described embodiment, wherein the organic image of previously described target user can be positioned at the device
In official's image library.
By the agency of mistake, the organic image of target user can be the device of a designated user in organic image library above
Official's image is also possible to the organic image of multiple users in organic image library.When the organic image of the target user is institute
When stating the organic image of multiple users in organic image library, gone back after the step S120 of above-described embodiment, before step S130
It may include steps of:
For each organ in the multiple organ, will obtain similarity assessment index according to sequence from high to low into
Row sequence, for example, the similarity assessment index of eyes is ranked up according to sequence from high to low；The similarity of nose is commented
Estimate index to be ranked up according to sequence from high to low etc..
The forward multiple candidate users of the sequencing of similarity of each organ are respectively obtained, for example, obtaining ranking for eyes
Forward preceding 5 candidate users, format can be (eyes: Zhang San；Li Si；…；…；King five)；For nose, the row of being similarly obtained
Forward preceding 5 candidate users of name, format can be (nose: Li Si；Zhang San；…；…；Zhao six).
In this way, the step S130 of above-described embodiment can specifically be executed in accordance with the following steps by above-mentioned processing:
According to the similarity assessment index and corresponding weight of the multiple organ, the target facial image is respectively obtained
With the multiple candidate user overall similarity evaluation index.
It should be noted that multiple candidate users that the sequencing of similarity obtained above for each organ is forward, respectively
The identity or collating sequence of the corresponding multiple candidate users of a organ may be different, such as eyes similarity assessment index is highest
It is Zhang San, and it is Li Si that nose similarity assessment index is highest.Therefore, target facial image and multiple candidate users are being calculated
When overall similarity evaluation index, a candidate user identity in an organ can be determined, first with the user identity
Based on, in the similarity evaluation index value for searching the user identity from other organs, finally obtain target facial image and
The overall similarity evaluation index of the user identity.
In addition, mentioned above obtain organic image library, after obtaining organic image library, above-mentioned several embodiments can be with
Include the following steps, feature extraction is carried out to the organic image in organic image library, obtains organic image feature database；On in this way,
The S130 for stating embodiment can specifically be executed in accordance with the following steps:
By the feature of obtained organic image respectively with the organic image feature of target user in organic image feature database into
Row comparison, obtains the similarity assessment index of the multiple organ.Since the organic image for having extracted target user in advance is special
Sign, in recognition of face, the operation for carrying out image characteristics extraction without being directed to target user in real time equally improves recognition of face
When recognition efficiency.
Above-mentioned several embodiments are mentioned obtaining organic image library, in addition, after obtaining organic image library, above-mentioned several realities
Applying example can also include the following steps: for the organic image in organic image library to be trained, and the organ for obtaining multiple organs is known
Other model；Wherein, the S120 of several embodiments can specifically be executed in accordance with the following steps above: by obtained organic image and mesh
The organic image of mark user is separately input in corresponding organ identification model, and the similarity assessment for obtaining the multiple organ refers to
Mark.For be described in detail, be illustrated below with reference to a specific embodiment, the embodiment can be divided into model training part and
Recognition of face part:
Model training part will be introduced first below, as shown in Fig. 2, including the following steps:
S210: facial image database is established.
The embodiment can establish facial image database based on the data of user's history brush face, wherein the facial image of foundation
The corresponding relationship between facial image and user (natural person) is had determined that in library, natural person can be with identity come table
Show, for example is indicated with identification card number or passport No. etc..
S220: identifying the facial image in the facial image database, obtains organic image library.
When identifying to facial image, multiple devices on every facial image can be determined by organ site identification model
The position of official, exports the characteristic point coordinate of multiple designated organs, these characteristic points can depict organ in the area of facial image
Domain.
Characteristic point coordinate based on above-mentioned output, can be in such a way that picture be cut, by a facial image according to device
Official is divided into different organic images and preservation, obtains organic image library, wherein natural person belonging to each organic image
Affiliated natural person is consistent with its original facial image.
S230: the organic image in organic image library is trained, and obtains the organ identification model of multiple organs.
The step can be based on the organic image in organic image library, using such as deep neural network DNN, convolutional Neural net
The machine learning methods such as network RNN are trained for each organ, obtain the organ identification model of each organ, that is, final
Obtain the organ identification model group for multiple organs, wherein different organ identification models can be using identical machine
Learning method training obtains, and is also possible to obtain using different machine learning method training.
Organ identification model in step S230, the organ site identification model being different from step S220, it is specific and
Speech, the organ site identification model in step S220 concern the region of a designated organ, and the device in step S230
Official's identification model is the descriptive model for being directed to organ itself.
For example, for this organ of eyes, the organ site identification model in step S220 be only concerned eyes size and
Region of the eyes in facial image, and the organ identification model in step S230 then may include the various features of inside of eye
It portrays, such as pupil size, eyelash length etc..
It, below will be to the use portion of specific recognition of face based on the organ identification model group that above-mentioned S210-S230 is obtained
Divide and be introduced, as shown in figure 3, including the following steps:
S310: target facial image is obtained.
Target facial image in the step can be arriving of obtaining by way of Image Acquisition.
For the user identity of target facial image, can choose whether to preset:
Mode one: if presetting the user identity of target facial image, whether as judge the target facial image
For the designated user in facial image database, that is, core body, wherein the User Identity of target facial image can be with face
The identity of the natural person of image library is consistent, such as using identification card number etc.；
Mode two: if not setting the user identity of target facial image, the target is exactly retrieved from facial image database
The user identity of facial image.
S320: identifying target facial image, obtains the corresponding organ of multiple organs in the target facial image
Image.
When the step specifically executes, the organ site identification model that can be used using the S220 of model training part is obtained
The corresponding organic image of multiple organs into target facial image.
It specifically can be based on the output data of organ site identification model, by the administrative division map of the organ in target facial image
It saves as cut into organic image to be compared, these organic images to be compared will be subsequent according to respective affiliated organ
Similarity is carried out with the organic image of target user to compare.
S330: obtained organic image is compared with the organic image of target user respectively, obtains the multiple device
The similarity assessment index of official.
When the step specifically executes, each the corresponding organ that can use the organ identification model group that S230 is obtained is known
Other model carries out the operation of feature extraction to organic image to be compared obtained in S320, obtains organic image to be compared
Characteristics of image, and compared with the organic image feature of target user, the similarity assessment for obtaining the multiple organ refers to
Mark.
In practical application, to improve the recognition efficiency of recognition of face organ can be utilized in above-mentioned model training part
Organic image in organic image library is carried out feature extraction by identification model group, obtains organic image feature database, and organic image is special
The composed structure for levying library is consistent with organic image library.
When the step specifically executes, according to the user identity for whether presetting target facial image in S310, two are had
Kind alignments.
Alignments one: if presetting the user identity of target facial image, directly from organic image library or
The organic image or characteristics of image that target user is extracted in organ characteristic library carry out similarity comparison, which avoids entirely
Office's organic image compares, and speed is fast.
Alignments two: if not setting the user identity of target facial image, i.e., when target user is multiple users, then
Organic image to be compared and organic image library or organ characteristic library can be subjected to overall comparison, speed is slow.
The similarity assessment index of the available multiple organs of comparison based on above-mentioned organic image, what one organ compared
Similarity assessment index can be using the Euclidean distance for calculating feature and by the way of normalizing to 0-100, it is believed that 0 is similarity degree
It is minimum, and 100 be similarity degree highest.
As a result format can be with are as follows:
[{ 1 title of organ }: [{ 1 similarity of user }, { 2 similarity of user } ... { user n similarity }], { 2, organ
Claim: [{ 1 similarity of user }, { 2 similarity of user } ... { user n similarity }]]
For convenient for calculate, it is above-mentioned be alignments two when, for the traversal of each organ, similarity assessment will be obtained
Index is ranked up according to sequence from high to low；Respectively obtain the forward multiple candidate use of the sequencing of similarity of each organ
Family.
Shaped like
[left eye: [{ name: Zhang San, similarity: 100 }, { name: Li Si, similarity: 90 }, { name: king five, similar
Degree: 76 }]；
Mouth: [name: Li Si, similarity: 96 }, name: Zhang San, similarity: 90 }, name: king five, similarity:
82}]]
S340: according to the similarity assessment index and corresponding weight of the multiple organ, obtain target facial image and
The overall similarity evaluation index of target user.
The step can do weighting marking based on the similarity assessment index of each organ, obtain target facial image and institute
The overall similarity evaluation index of target user is stated, can specifically use following formula:
Overall similarity evaluation index=∑ (organ weight × organ similarity assessment index)/organ number
S350: the identification to the target facial image is determined according to the overall similarity evaluation index of the target user
As a result.
The step can be based on preset (similarity) preset threshold, it is above-mentioned for alignments for the moment, if whole
Similarity assessment index is greater than or equal to the preset threshold, that is, thinks that target facial image is target user；If overall similarity
Evaluation index is less than the preset threshold, that is, thinks that target facial image is not target user.
It is above-mentioned be alignments two when, all of preset threshold can be greater than or equal in overall similarity evaluation index
In candidate user, it is believed that the maximum candidate user of overall similarity evaluation index is the user identity of target facial image；If
The overall similarity evaluation index of all candidate users is respectively less than preset threshold, then it is assumed that does not detect the target facial image
User identity.
The face identification method provided by this specification embodiment, by being based between organic image in recognition of face
It compares, and then the similarity assessment index based on organic image obtains the overall similarity evaluation index of facial image, phase
For the way of contrast based on entire facial image, knowledge can also be quickly made even if collected target facial image is imperfect
Not, it solves the problems, such as that quickly identification can not be made to user because not acquiring complete facial image.
This specification embodiment passes through the biology to target facial image progress organ identification and each organ of independent draws
Feature is compared, and then does the identification that weighted comprehensive completes target facial image, also solves such as face, eyeprint, iris
The identification of single creature feature can not acquire the problem of input picture leads to not identification under special circumstances, can be improved face knowledge
It is other entirely through rate, bring more flexible usage scenario for face recognition technology.
In this specification embodiment, the organic image of target user is to carry out organ to the facial image of target user in advance
Identification, segmentation obtain, and the facial image without real-time target user carries out the processing such as identification segmentation, thus when improving recognition of face
Recognition efficiency.
Organic image feature of this specification embodiment due to having extracted target user in advance, in recognition of face, nothing
The operation that need to carry out image characteristics extraction for target user in real time, equally improves recognition efficiency when recognition of face.
Above instructions part describes face identification method embodiment in detail, as shown in figure 4, this specification additionally provides
A kind of face identification device, as shown in figure 4, the device includes:
Organic image obtains module 410, can be used for identifying target facial image, obtains the target face figure
The corresponding organic image of multiple organs as in；
Organ similarity obtain module 420, the organic image that can be used for obtain respectively with the organ figure of target user
As comparing, the similarity assessment index of the multiple organ is obtained；
Overall similarity obtains module 430, can be used for the similarity assessment index and correspondence according to the multiple organ
Weight, obtain the target facial image and the overall similarity evaluation index of the target user；
Recognition result determining module 440 can be used for being determined according to the overall similarity evaluation index of the target user
To the recognition result of the target facial image.
The face identification device provided by this specification embodiment, by being based between organic image in recognition of face
It compares, and then the similarity assessment index based on organic image obtains the overall similarity evaluation index of facial image, phase
For the way of contrast based on entire facial image, knowledge can also be quickly made even if collected target facial image is imperfect
Not, it solves the problems, such as that quickly identification can not be made to user because not acquiring complete facial image.
Optionally, as one embodiment, the organic image of the target user is: the organ figure of a designated user
Picture；Or the organic image of multiple users.
Optionally, as one embodiment, described device further includes organic image library building module, can be used for establishing people
Face image library；Facial image in the facial image database is identified, organic image library is obtained；Wherein, the target is used
The organic image at family is located in the organic image library.
Optionally, as one embodiment, when the organic image of the target user is more in the organic image library
When the organic image of a user, described device further includes that candidate user obtains module, be can be used for in the multiple organ
Each organ, similarity assessment index will be obtained and be ranked up according to sequence from high to low；Respectively obtain each organ
The forward multiple candidate users of sequencing of similarity；Wherein, overall similarity obtains module 430, can be used for according to the multiple
The similarity assessment index and corresponding weight of organ, respectively obtain the target facial image and the multiple candidate user is whole
Body similarity assessment index.
Optionally, as one embodiment, described device further includes that characteristic extracting module can be used for organic image
Organic image in library carries out feature extraction, obtains organic image feature database；Wherein, organ similarity obtains module 420, can be with
For the feature of obtained organic image being carried out with the organic image feature of target user in organic image feature database respectively pair
Than obtaining the similarity assessment index of the multiple organ.
Optionally, as one embodiment, described device further includes that model obtains module, be can be used for organic image
Organic image in library is trained, and obtains the organ identification model of multiple organs；Wherein, organ similarity obtains module 420,
The organic image of the organic image and target user that can be used for obtain is separately input in corresponding organ identification model, is obtained
To the similarity assessment index of the multiple organ.
Optionally, as one embodiment, recognition result determining module 440, if can be used for target user is one
User, and the overall similarity evaluation index of target user is greater than or equal to preset threshold, determines to the target facial image
It identifies successfully；Or if target user is a user, and the overall similarity evaluation index of target user is less than preset threshold,
It determines to the target facial image recognition failures；If target user is multiple users, and at least one in multiple users is used
The overall similarity evaluation index at family is greater than or equal to preset threshold, and determination is identified as function to the target facial image；Or such as
Fruit target user is multiple users, and the overall similarity evaluation index of multiple users is respectively less than preset threshold, is determined to described
Target facial image recognition failures.
Corresponding this specification embodiment above is referred to according to the above-mentioned face identification device of this specification embodiment
The process of face identification method, also, each unit/module and other above-mentioned operations and/or function in the face identification device
The corresponding process in face identification method can be realized respectively, for sake of simplicity, details are not described herein.
Below in conjunction with Fig. 5 detailed description according to the electronic equipment of this specification embodiment.With reference to Fig. 5, in hardware view,
Electronic equipment includes processor, optionally, including internal bus, network interface, memory.Wherein, as shown in figure 5, memory
It may include memory, such as high-speed random access memory (Random-Access Memory, RAM), it is also possible to further include non-
Volatile memory (non-volatile memory), for example, at least 1 magnetic disk storage etc..Certainly, which may be used also
It can include hardware required for realizing other business.
Processor, network interface and memory can be connected with each other by internal bus, which can be industry
Standard architecture (Industry Standard Architecture, ISA) bus, Peripheral Component Interconnect standard
(Peripheral Component Interconnect, PCI) bus or expanding the industrial standard structure (Extended
Industry Standard Architecture, EISA) bus etc..The bus can be divided into address bus, data/address bus,
Control bus etc..Only to be indicated with a four-headed arrow in Fig. 5, it is not intended that an only bus or one kind convenient for indicating
The bus of type.
Memory, for storing program.Specifically, program may include program code, and said program code includes calculating
Machine operational order.Memory may include memory and nonvolatile memory, and provide instruction and data to processor.
Processor is from the then operation into memory of corresponding computer program is read in nonvolatile memory, in logical layer
The device of forwarding chat message is formed on face.Processor executes the program that memory is stored, and is specifically used for executing this explanation
The operation of the previously described embodiment of the method for book.
The method that the method, apparatus that above-mentioned Fig. 1 to Fig. 4 illustrated embodiment discloses executes can be applied in processor, or
Person is realized by processor.Processor may be a kind of IC chip, the processing capacity with signal.During realization,
Each step of the above method can be completed by the integrated logic circuit of the hardware in processor or the instruction of software form.On
The processor stated can be at general processor, including central processing unit (Central Processing Unit, CPU), network
Manage device (Network Processor, NP) etc.；Can also be digital signal processor (Digital Signal Processor,
DSP), specific integrated circuit (Application Specific Integrated Circuit, ASIC), field programmable gate
Array (Field-Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or crystalline substance
Body pipe logical device, discrete hardware components.May be implemented or execute disclosed each method in the embodiment of the present application, step and
Logic diagram.General processor can be microprocessor or the processor is also possible to any conventional processor etc..In conjunction with
The step of method disclosed in the embodiment of the present application, can be embodied directly in hardware decoding processor and execute completion, or with decoding
Hardware and software module combination in processor execute completion.Software module can be located at random access memory, flash memory, read-only storage
In the storage medium of this fields such as device, programmable read only memory or electrically erasable programmable memory, register maturation.It should
The step of storage medium is located at memory, and processor reads the information in memory, completes the above method in conjunction with its hardware.
The method that electronic equipment shown in fig. 5 can also carry out Fig. 1 to Fig. 3, and realize face identification method in Fig. 1 to Fig. 3
The function of illustrated embodiment, details are not described herein for the embodiment of the present application.
Certainly, other than software realization mode, other implementations are not precluded in the electronic equipment of the application, for example patrol
Collect device or the mode of software and hardware combining etc., that is to say, that the executing subject of following process flow is not limited to each patrol
Unit is collected, hardware or logical device are also possible to.
This specification embodiment also provides a kind of computer readable storage medium, is stored on computer readable storage medium
Computer program, the computer program realize each process of above-mentioned each embodiment of the method when being executed by processor, and can reach
To identical technical effect, to avoid repeating, which is not described herein again.Wherein, the computer readable storage medium, it is such as read-only
Memory (Read-Only Memory, abbreviation ROM), random access memory (Random Access Memory, abbreviation
RAM), magnetic or disk etc..
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer readable instructions, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), flash memory or other memory techniques, read-only disc read only memory (CD-ROM) (CD-ROM),
Digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or other magnetic storage devices
Or any other non-transmission medium, can be used for storage can be accessed by a computing device information.As defined in this article, it calculates
Machine readable medium does not include temporary computer readable media (transitory media), such as the data-signal and carrier wave of modulation.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including element
There is also other identical elements in process, method, commodity or equipment.
The above is only embodiments herein, are not intended to limit this application.To those skilled in the art,
Various changes and changes are possible in this application.It is all within the spirit and principles of the present application made by any modification, equivalent replacement,
Improve etc., it should be included within the scope of the claims of this application.
Claims (10)
1. a kind of face identification method, comprising:
Target facial image is identified, the corresponding organic image of multiple organs in the target facial image is obtained；
Obtained organic image is compared with the organic image of target user respectively, obtains the similarity of the multiple organ
Evaluation index；
According to the similarity assessment index and corresponding weight of the multiple organ, the target facial image and the mesh are obtained
Mark the overall similarity evaluation index of user；
The recognition result to the target facial image is determined according to the overall similarity evaluation index of the target user.
2. according to the method described in claim 1, the organic image of the target user is:
The organic image of one designated user；Or
The organic image of multiple users.
3. method according to claim 1 or 2, before being identified to target facial image, the method also includes:
Establish facial image database；
Facial image in the facial image database is identified, organic image library is obtained, wherein the device of the target user
Official's image is located in the organic image library.
4. according to the method described in claim 3,
When the organic image of the target user is the organic image of multiple users in the organic image library, the method
Further include: for each organ in the multiple organ, will obtain similarity assessment index according to sequence from high to low into
Row sequence；Respectively obtain the forward multiple candidate users of the sequencing of similarity of each organ；
Wherein, according to the similarity assessment index of the multiple organ and corresponding weight, obtain the target facial image and
The overall similarity evaluation index of the target user, comprising: according to the similarity assessment index and correspondence of the multiple organ
Weight, respectively obtain the target facial image and the multiple candidate user overall similarity evaluation index.
5. according to the method described in claim 3, the method also includes to organic image library after obtaining organic image library
In organic image carry out feature extraction, obtain organic image feature database；
Wherein, obtained organic image is compared with the organic image of target user respectively, obtains the multiple organ
Similarity assessment index, comprising: by the feature of the obtained organic image device with target user in organic image feature database respectively
Official's characteristics of image compares, and obtains the similarity assessment index of the multiple organ.
6. according to the method described in claim 3, after obtaining organic image library, the method also includes: by organic image library
In organic image be trained, obtain the organ identification model of multiple organs；
Wherein, obtained organic image is compared with the organic image of target user respectively, obtains the multiple organ
Similarity assessment index, comprising: the organic image of obtained organic image and target user is separately input to corresponding organ
In identification model, the similarity assessment index of the multiple organ is obtained.
7. according to the method described in claim 1, being determined according to the overall similarity evaluation index of the target user to described
The recognition result of target facial image, comprising:
If target user is a user, and the overall similarity evaluation index of target user is greater than or equal to preset threshold,
Determination is identified as function to the target facial image；
If target user is a user, and the overall similarity evaluation index of target user is less than preset threshold, determination pair
The target facial image recognition failures；
If target user be multiple users, and in multiple users the overall similarity evaluation index of at least one user be greater than or
Equal to preset threshold, determination is identified as function to the target facial image；Or
If target user is multiple users, and the overall similarity evaluation index of multiple users is respectively less than preset threshold, determines
To the target facial image recognition failures.
8. a kind of face identification device, comprising:
Organic image obtains module, identifies to target facial image, obtains multiple organs pair in the target facial image
The organic image answered；
Organ similarity obtains module, and obtained organic image is compared with the organic image of target user respectively, is obtained
The similarity assessment index of the multiple organ；
Overall similarity obtains module, according to the similarity assessment index and corresponding weight of the multiple organ, obtains described
The overall similarity evaluation index of target facial image and the target user；
Recognition result determining module is determined according to the overall similarity evaluation index of the target user to the target face figure
The recognition result of picture.
9. a kind of electronic equipment, comprising: memory, processor and be stored on the memory and can transport on the processor
Capable computer program realizes following operation when the computer program is executed by the processor:
Target facial image is identified, the corresponding organic image of multiple organs in the target facial image is obtained；
Obtained organic image is compared with the organic image of target user respectively, obtains the similarity of the multiple organ
Evaluation index；
According to the similarity assessment index and corresponding weight of the multiple organ, the target facial image and the mesh are obtained
Mark the overall similarity evaluation index of user；
The recognition result to the target facial image is determined according to the overall similarity evaluation index of the target user.
10. a kind of computer readable storage medium, computer program, the meter are stored on the computer readable storage medium
Following operation is realized when calculation machine program is executed by processor:
Target facial image is identified, the corresponding organic image of multiple organs in the target facial image is obtained；
Obtained organic image is compared with the organic image of target user respectively, obtains the similarity of the multiple organ
Evaluation index；
According to the similarity assessment index and corresponding weight of the multiple organ, the target facial image and the mesh are obtained
Mark the overall similarity evaluation index of user；
The recognition result to the target facial image is determined according to the overall similarity evaluation index of the target user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810735594.9A CN109145720A (en) | 2018-07-06 | 2018-07-06 | A kind of face identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810735594.9A CN109145720A (en) | 2018-07-06 | 2018-07-06 | A kind of face identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109145720A true CN109145720A (en) | 2019-01-04 |
Family
ID=64799802
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810735594.9A Pending CN109145720A (en) | 2018-07-06 | 2018-07-06 | A kind of face identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109145720A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619689A (en) * | 2019-08-30 | 2019-12-27 | 恒大智慧科技有限公司 | Automatic sign-in and card-punching method for smart building, computer equipment and storage medium |
| CN110942014A (en) * | 2019-11-22 | 2020-03-31 | 浙江大华技术股份有限公司 | Face recognition rapid retrieval method and device, server and storage device |
| CN112241689A (en) * | 2020-09-24 | 2021-01-19 | 北京澎思科技有限公司 | Face recognition method and device, electronic equipment and computer readable storage medium |
| CN113269935A (en) * | 2021-07-16 | 2021-08-17 | 融讯伟业（北京）科技有限公司 | Visual weighing method and weighing system based on screen-free weighing device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050207622A1 (en) * | 2004-03-16 | 2005-09-22 | Haupt Gordon T | Interactive system for recognition analysis of multiple streams of video |
| JP2007233247A (en) * | 2006-03-03 | 2007-09-13 | Fujifilm Corp | Focus adjusting amount decision device, method and program and imaging apparatus |
| CN106980819A (en) * | 2017-03-03 | 2017-07-25 | 竹间智能科技（上海）有限公司 | Similarity judgement system based on human face five-sense-organ |
| CN107818308A (en) * | 2017-10-31 | 2018-03-20 | 平安科技（深圳）有限公司 | A kind of recognition of face intelligence comparison method, electronic installation and computer-readable recording medium |
| CN108171463A (en) * | 2017-12-01 | 2018-06-15 | 平安科技（深圳）有限公司 | A kind of long-range face label method, electronic device and computer readable storage medium |
-
2018
- 2018-07-06 CN CN201810735594.9A patent/CN109145720A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050207622A1 (en) * | 2004-03-16 | 2005-09-22 | Haupt Gordon T | Interactive system for recognition analysis of multiple streams of video |
| JP2007233247A (en) * | 2006-03-03 | 2007-09-13 | Fujifilm Corp | Focus adjusting amount decision device, method and program and imaging apparatus |
| CN106980819A (en) * | 2017-03-03 | 2017-07-25 | 竹间智能科技（上海）有限公司 | Similarity judgement system based on human face five-sense-organ |
| CN107818308A (en) * | 2017-10-31 | 2018-03-20 | 平安科技（深圳）有限公司 | A kind of recognition of face intelligence comparison method, electronic installation and computer-readable recording medium |
| CN108171463A (en) * | 2017-12-01 | 2018-06-15 | 平安科技（深圳）有限公司 | A kind of long-range face label method, electronic device and computer readable storage medium |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619689A (en) * | 2019-08-30 | 2019-12-27 | 恒大智慧科技有限公司 | Automatic sign-in and card-punching method for smart building, computer equipment and storage medium |
| CN110942014A (en) * | 2019-11-22 | 2020-03-31 | 浙江大华技术股份有限公司 | Face recognition rapid retrieval method and device, server and storage device |
| CN112241689A (en) * | 2020-09-24 | 2021-01-19 | 北京澎思科技有限公司 | Face recognition method and device, electronic equipment and computer readable storage medium |
| CN113269935A (en) * | 2021-07-16 | 2021-08-17 | 融讯伟业（北京）科技有限公司 | Visual weighing method and weighing system based on screen-free weighing device |
| CN113269935B (en) * | 2021-07-16 | 2021-11-30 | 融讯伟业（北京）科技有限公司 | Visual weighing method and weighing system based on screen-free weighing device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20190377930A1 (en) | Method and System for Face Recognition Via Deep Learning | |
| TWI752418B (en) | Server, client, user authentication method and system | |
| CN109145720A (en) | A kind of face identification method and device | |
| WO2019119505A1 (en) | Face recognition method and device, computer device and storage medium | |
| CN109492627B (en) | Scene text erasing method based on depth model of full convolution network | |
| CN110110715A (en) | Text detection model training method, text filed, content determine method and apparatus | |
| US10169683B2 (en) | Method and device for classifying an object of an image and corresponding computer program product and computer-readable medium | |
| CN110851835A (en) | Image model detection method and device, electronic equipment and storage medium | |
| CN107333071A (en) | Video processing method and device, electronic equipment and storage medium | |
| CN104573652A (en) | Method, device and terminal for determining identity identification of human face in human face image | |
| CN106484837A (en) | The detection method of similar video file and device | |
| CN107871314B (en) | Sensitive image identification method and device | |
| CN109086734A (en) | The method and device that pupil image is positioned in a kind of pair of eye image | |
| CN111951154B (en) | Picture generation method and device containing background and medium | |
| CN115631112B (en) | Building contour correction method and device based on deep learning | |
| CN109299276A (en) | One kind converting the text to word insertion, file classification method and device | |
| CN109064217A (en) | Method, apparatus and electronic equipment are determined based on the core body strategy of user gradation | |
| CN112613471B (en) | Face living body detection method, device and computer readable storage medium | |
| CN110046340A (en) | The training method and device of textual classification model | |
| CN109597987A (en) | A kind of text restoring method, device and electronic equipment | |
| CN110502961A (en) | A kind of facial image detection method and device | |
| CN108734099A (en) | Auth method and device, electronic equipment, computer program and storage medium | |
| CN113743160A (en) | Method, apparatus and storage medium for biopsy | |
| CN109886186A (en) | A kind of face identification method and device | |
| CN113536845B (en) | Face attribute identification method and device, storage medium and intelligent equipment |
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 | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40002370 Country of ref document: HK |
|
| TA01 | Transfer of patent application right |
Effective date of registration: 20200922 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Innovative advanced technology Co.,Ltd. Address before: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant before: Advanced innovation technology Co.,Ltd. Effective date of registration: 20200922 Address after: Cayman Enterprise Centre, 27 Hospital Road, George Town, Grand Cayman, British Islands Applicant after: Advanced innovation technology Co.,Ltd. Address before: A four-storey 847 mailbox in Grand Cayman Capital Building, British Cayman Islands Applicant before: Alibaba Group Holding Ltd. |
|
| TA01 | Transfer of patent application right | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190104 |
|
| RJ01 | Rejection of invention patent application after publication |