CN104077563B - Face identification method and device - Google Patents
Face identification method and device Download PDFInfo
- Publication number
- CN104077563B CN104077563B CN201410240747.4A CN201410240747A CN104077563B CN 104077563 B CN104077563 B CN 104077563B CN 201410240747 A CN201410240747 A CN 201410240747A CN 104077563 B CN104077563 B CN 104077563B
- Authority
- CN
- China
- Prior art keywords
- facial image
- alignment
- specified
- similarity
- area
- 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.)
- Active
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The disclosure is directed to a kind of face identification method and device, belong to technical field of face recognition.Methods described includes：Obtain the first facial image；Determine the similarity of first facial image and specified facial image；Determine interference value of the interference characteristic to the similarity of first facial image and the specified facial image；According to the interference value, the similarity is adjusted.Interference value of the disclosure by the interference characteristic of the first facial image of determination and specified facial image to similarity, and the similarity of the first facial image and specified facial image is adjusted according to the interference value, avoid due to there are deep frame glasses in the first facial image and specified facial image on face, or first face in facial image and specified facial image there is the reasons such as same or analogous hair style, caused by be not that two very high faces are mistaken for that similarity is higher by similarity, improve the accuracy rate of identification.
Description
Technical field
This disclosure relates to technical field of face recognition, more particularly to a kind of face identification method and device.
Background technology
Face is the primary identity that people mutually differentiate, recognized, remembering, recognition of face computer vision, pattern-recognition,
Occupied an important position in multimedia technology research.
In correlation technique, typically two width facial images carry out Face datection, characteristic point to the method for recognition of face successively
Positioning, feature extraction, and similarity measurement is carried out according to the feature of extraction, obtain for weighing two width facial image similarities
Fraction.
When there is face in deep frame glasses, or two width facial images that there is identical or phase in two width facial images on face
As hair style when, original similarity is not two very high faces, it is more likely that can be considered as that similarity is higher, therefore identify
Accuracy rate is relatively low.
The content of the invention
The problem of in order to overcome the accuracy rate identified present in correlation technique relatively low, the disclosure provide a kind of recognition of face
Method and apparatus.The technical scheme is as follows：
According to the first aspect of the embodiment of the present disclosure, there is provided a kind of face identification method, suitable for judging two face figures
The similarity of picture, including：
Obtain the first facial image；
Determine the similarity of first facial image and specified facial image；
Determine interference value of the interference characteristic to the similarity of first facial image and the specified facial image；
According to the interference value, the similarity is adjusted；
The interference characteristic for determining first facial image and the specified facial image is done to the similarity
Value is disturbed, including：
First facial image and the specified facial image are alignd with the average shape model of setting respectively；
Skin analysis is carried out to the specified facial image after first facial image after alignment and alignment respectively,
It is determined that the non-area of skin color of first facial image and the specified facial image after alignment after alignment；
According to the non-colour of skin area of first facial image after the alignment and the specified facial image after alignment
Interference value of the interference characteristic of domain, calculating first facial image and the specified facial image to similarity；
The non-skin of the specified facial image after first facial image according to after the alignment and alignment
Color region, interference value of the interference characteristic to similarity of first facial image and the specified facial image is calculated, including：
Respectively by the non-colour of skin of first facial image after the alignment and the specified facial image after alignment
The characteristic value of the pixel in region is taken as 1, first facial image after the alignment and the specified face figure after alignment
The characteristic value of the area of skin color of picture is taken as 0；
Image friendship is carried out to the specified facial image after first facial image after alignment and alignment；
The spy of pixel corresponding to the specified facial image after first facial image and alignment after statistics alignment
Value indicative is the quantity of 1 pixel；
The quantity that counting statistics obtains with align after first facial image or the specified face figure after aliging
The total ratio of the pixel of picture, the interference characteristic of first facial image and the specified facial image is obtained to similarity
Interference value.
In the first possible implementation of the disclosure, it is described respectively to first facial image after alignment and
The specified facial image after alignment carries out skin analysis, it is determined that the institute after first facial image and alignment after alignment
The non-area of skin color of specified facial image is stated, including：
Choose in first facial image after alignment and the specified facial image after alignment, with the average shape
Region corresponding to area of skin color is set as the first area of skin color in shape faceform；
The features of skin colors of first area of skin color is extracted, and by after first facial image after alignment and alignment
In the specified facial image, the features of skin colors identical region of features of skin colors and first area of skin color is defined as the second skin
Color region；
By first facial image after alignment and alignment after the specified facial image in, except first colour of skin
All areas beyond region and second area of skin color, after first facial image after the alignment and alignment
The specified facial image non-area of skin color.
It is described according to the interference value in second of possible implementation of the disclosure, the similarity is adjusted, is wrapped
Include：
According to predetermined functional relation, according to the interference value, the correction value of the similarity is determined；
The similarity is subtracted into the correction value, the similarity after being adjusted.
According to the second aspect of the embodiment of the present disclosure, there is provided a kind of face identification device, suitable for judging two face figures
The similarity of picture, including：
Acquisition module, for obtaining the first facial image；
Identification module, for determining the similarity of first facial image and specified facial image；
Determining module is disturbed, for determining the interference characteristic of first facial image and the specified facial image to institute
State the interference value of similarity；
Correcting module, for according to the interference value, adjusting the similarity；
The interference determining module includes：
Alignment unit, for by first facial image and the specified facial image average shape with setting respectively
Model aligns；
Area determination unit, for respectively to first facial image after alignment and the specified face after alignment
Image carries out skin analysis, it is determined that the non-skin of the specified facial image after first facial image and alignment after alignment
Color region；
Interference calculation unit, for according to first facial image after the alignment and the nominator after alignment
The non-area of skin color of face image, the interference characteristic of first facial image and the specified facial image is calculated to similarity
Interference value；
The interference calculation unit is used for,
Respectively by the non-colour of skin of first facial image after the alignment and the specified facial image after alignment
The characteristic value of the pixel in region is taken as 1, first facial image after the alignment and the specified face figure after alignment
The characteristic value of the area of skin color of picture is taken as 0；
Image friendship is carried out to the specified facial image after first facial image after alignment and alignment；
The spy of pixel corresponding to the specified facial image after first facial image and alignment after statistics alignment
Value indicative is the quantity of 1 pixel；
The quantity that counting statistics obtains with align after first facial image or the specified face figure after aliging
The total ratio of the pixel of picture, the interference characteristic of first facial image and the specified facial image is obtained to similarity
Interference value.
In the first possible implementation of the disclosure, the area determination unit is used for,
Choose in first facial image after alignment and the specified facial image after alignment, with the average shape
Region corresponding to area of skin color is set as the first area of skin color in shape faceform；
The features of skin colors of first area of skin color is extracted, and by after first facial image after alignment and alignment
In the specified facial image, the features of skin colors identical region of features of skin colors and first area of skin color is defined as the second skin
Color region；
By first facial image after alignment and alignment after the specified facial image in, except first colour of skin
All areas beyond region and second area of skin color, after first facial image after the alignment and alignment
The specified facial image non-area of skin color.
In second of possible implementation of the disclosure, the correcting module includes：
Correction value determining unit, for according to predetermined functional relation, according to the interference value, determining the similarity
Correction value；
Score calculating unit, for the similarity to be subtracted into the correction value, the similarity after being adjusted.
According to the third aspect of the embodiment of the present disclosure, there is provided a kind of face identification device, suitable for judging two face figures
The similarity of picture, including：
Processor；
For storing the memory of processor-executable instruction；
Wherein, the processor is configured as：
Obtain the first facial image；
Determine the similarity of first facial image and specified facial image；
Determine interference value of the interference characteristic to the similarity of first facial image and the specified facial image；
According to the interference value, the similarity is adjusted；
The interference characteristic for determining first facial image and the specified facial image is done to the similarity
Value is disturbed, including：
First facial image and the specified facial image are alignd with the average shape model of setting respectively；
Skin analysis is carried out to the specified facial image after first facial image after alignment and alignment respectively,
It is determined that the non-area of skin color of first facial image and the specified facial image after alignment after alignment；
According to the non-colour of skin area of first facial image after the alignment and the specified facial image after alignment
Interference value of the interference characteristic of domain, calculating first facial image and the specified facial image to similarity；
The non-skin of the specified facial image after first facial image according to after the alignment and alignment
Color region, interference value of the interference characteristic to similarity of first facial image and the specified facial image is calculated, including：
Respectively by the non-colour of skin of first facial image after the alignment and the specified facial image after alignment
The characteristic value of the pixel in region is taken as 1, first facial image after the alignment and the specified face figure after alignment
The characteristic value of the area of skin color of picture is taken as 0；
Image friendship is carried out to the specified facial image after first facial image after alignment and alignment；
The spy of pixel corresponding to the specified facial image after first facial image and alignment after statistics alignment
Value indicative is the quantity of 1 pixel；
The quantity that counting statistics obtains with align after first facial image or the specified face figure after aliging
The total ratio of the pixel of picture, the interference characteristic of first facial image and the specified facial image is obtained to similarity
Interference value.
The technical scheme provided by this disclosed embodiment can include the following benefits：By determining the first facial image
Interference value with the interference characteristic of specified facial image to similarity, and the first facial image is adjusted according to the interference value and specified
The similarity of facial image, avoid due to there are deep frame glasses in the first facial image and specified facial image on face, or
Face has the reasons such as same or analogous hair style in first facial image and specified facial image, caused by by similarity be not
Two very high faces are mistaken for that similarity is higher, improve the accuracy rate of identification.
It should be appreciated that the general description and following detailed description of the above are only exemplary and explanatory, not
The disclosure can be limited.
Brief description of the drawings
Accompanying drawing herein is merged in specification and forms the part of this specification, shows the implementation for meeting the present invention
Example, and for explaining principle of the invention together with specification.
Fig. 1 is a kind of flow chart of face identification method according to an exemplary embodiment；
Fig. 2 is the flow chart of another face identification method according to an exemplary embodiment；
Fig. 3 is a kind of block diagram of face identification device according to an exemplary embodiment；
Fig. 4 is the block diagram of another face identification device according to an exemplary embodiment；
Fig. 5 is a kind of block diagram of face identification device according to an exemplary embodiment.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent and the consistent all embodiments of the present invention.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects being described in detail in claims, of the invention.
Fig. 1 is a kind of flow chart of face identification method according to an exemplary embodiment, as shown in figure 1, face
Recognition methods is used in mobile terminal, the similarity suitable for judging two facial images, comprises the following steps.
In step S101, the first facial image is obtained.
In step s 102, the similarity of the first facial image and specified facial image is determined.
In step s 103, interference of the interference characteristic of the first facial image and specified facial image to similarity is determined
Value.
In the present embodiment, interference characteristic is characteristic value phase in the human face region of the first facial image and specified facial image
The feature of same non-face part.Interference characteristic can include deep frame glasses, hair, beard etc., and the disclosure is not restricted to this.
Interference value is used for the annoyance level for weighing the similarity of interference characteristic pair determination.
In step S104, according to interference value, similarity is adjusted.
The embodiment of the present disclosure is by determining that the interference characteristic of the first facial image and specified facial image is done to similarity
Value is disturbed, and the similarity of the first facial image and specified facial image is adjusted according to the interference value, is avoided due to the first face figure
There is face in deep frame glasses, or the first facial image and specified facial image to have in picture and specified facial image on face
The reasons such as same or analogous hair style, caused by by similarity not to be that two very high faces are mistaken for similarity higher, improve
The accuracy rate of identification.
Fig. 2 is the flow chart of another face identification method according to an exemplary embodiment, as shown in Fig. 2 people
Face recognition method is used in mobile terminal, the similarity suitable for judging two facial images, comprises the following steps.
In step s 201, the first facial image and the second facial image are obtained.
In the present embodiment, the second facial image is to specify facial image, that is, it can be obtained from the external world to specify facial image
Take.In other embodiments, specified facial image can also be prestored in the terminal, and the disclosure is not restricted to this.
In a kind of implementation of the present embodiment, step S201 can include：
Obtain two images；
Respectively two images are carried out with Face datection, the first facial image is determined in piece image in two images,
The second facial image is determined in another piece image in two images.
In actual applications, two images progress Face datection can be used and is based on Adaboost (Adaptive
Boosting, adaptive enhancing) Face datection algorithm.Image is scaled successively according to predetermined ratio first, then every
The subwindow of the 20*20 pixels of individual image differentiates it is face successively, or non-face, finally obtain in image the position of face and
Size.According to the position of face in image and size, intercepted in two images, you can obtain the first facial image and
Two facial images.
In step S202, recognition of face is carried out to the first facial image and the second facial image, obtains the first face figure
The similarity of picture and the second facial image.
In another implementation of the present embodiment, step S202 can include：
The positioning feature point algorithm based on ASM (Active Shape Model, active shape model) is respectively adopted, it is determined that
The shape of the shape of first facial image and the second facial image；
According to the shape of the first facial image, the first facial image is carried out successively Gabor (Jia Bai) wavelet transformation,
PCA (Principal Component Analysis, principal component analysis), LDA (Linear Discriminant
Analysis, linear discriminant analysis), obtain the characteristic information of the first facial image；
According to the shape of the second facial image, Gabor wavelet conversion, PCA, LDA are carried out to the second facial image, obtained
To the characteristic information of the second facial image；
Calculate the COS distance between the characteristic information of the first facial image and the characteristic information of the second facial image, and root
According to COS distance, the similarity of the first facial image and the second facial image is obtained.
In actual applications, when determining the shape of facial image using the positioning feature point algorithm based on ASM, first exist
Initial alignment, then each characteristic point for initial alignment are carried out in image, according to the gray level model of each characteristic point, in image
The accurate location of the middle each characteristic point of search is simultaneously modified.By repeatedly search and amendment, it is determined that shape can be compared with
Reflect face well.
The similarity of first facial image and the second facial image can use fraction representation, such as with 100 points for full marks, 90
It is identical point to represent the first facial image and the second facial image to have 90% region, and similarity is high.
It is to be appreciated that it can be achieved to determine the phase of the first facial image and specified facial image by performing step S202
Like degree.
In step S203, by the first facial image and the second facial image the average shape model pair with setting respectively
Together.
In another implementation of the present embodiment, step S203 can include：
Respectively according to average shape faceform, the shape of shape and the second facial image to the first facial image
Model carries out two dimensional affine conversion.
In actual applications, when carrying out two dimensional affine conversion, shape and the average face shape of face need to be only directed to
Model calculates transfer function.Two are being carried out to the shape of the first facial image and the shape of the second facial image
After tieing up affine transformation, the facial image of size identical first and the second facial image, the i.e. average shape with setting can be obtained
The first facial image and the second facial image of shape model alignment.
In step S204, the colour of skin is carried out to the first facial image after alignment and the second facial image after alignment respectively
Analysis, it is determined that the non-area of skin color of the first facial image and the second facial image after alignment after alignment.
In another implementation of the disclosure, step S204 can include：
Choose respectively in the first facial image after alignment and the second facial image after alignment, with average shape face mould
Region corresponding to area of skin color is set as the first area of skin color in type；
Extract the features of skin colors of the first area of skin color respectively, and by second after the first facial image after alignment and alignment
In facial image, the features of skin colors identical region of features of skin colors and the first area of skin color is defined as the second area of skin color；
Respectively by the first facial image after alignment and the second facial image after alignment, except the first area of skin color and the
All areas beyond two area of skin color, as the first facial image after alignment and the non-skin of the second facial image after alignment
Color region.
It is to be appreciated that the cheek position in face is larger for the probability of area of skin color, therefore can be by the face in face
Cheek position is as setting area of skin color.In actual applications, can be by the characteristic point of cheek part in average shape faceform
It is demarcated as setting area of skin color.
In step S205, according to the non-colour of skin area of the first facial image after alignment and the second facial image after alignment
Interference value of the interference characteristic of domain, the first facial image of calculating and the second facial image to similarity.
In the present embodiment, interference characteristic is characteristic value phase in the human face region of the first facial image and the second facial image
The feature of same non-face part.Interference characteristic can include deep frame glasses, hair, beard etc., and the disclosure is not restricted to this.
Interference value is used for the annoyance level for weighing the similarity of interference characteristic pair determination.
In another implementation of the disclosure, step S205 can include：
Respectively by the pixel of the first facial image after alignment and the non-area of skin color of the second facial image after alignment
Characteristic value is taken as 1, and the characteristic value of the area of skin color of the first facial image after alignment and the second facial image after alignment is taken as
0；
Image friendship is carried out to the first facial image after alignment and the second facial image after alignment；
The characteristic value of pixel is 1 corresponding to the second facial image after the first facial image and alignment after statistics alignment
Pixel quantity；
The quantity that counting statistics obtains with align after the first facial image or the second facial image after aliging pixel
Total ratio, obtain interference value of the interference characteristic to similarity of the first facial image and the second facial image.
In the present embodiment, the area of skin color of the first facial image after alignment and the second facial image after alignment includes
First area of skin color and the second area of skin color.
It is to be appreciated that by performing step S203-S205 successively, you can realize and determine the first facial image and nominator
Interference value of the interference characteristic of face image to similarity.
In step S206, according to interference value, similarity is adjusted.
In another implementation of the disclosure, step S206 can include：
According to predetermined functional relation, according to interference value, the correction value of similarity is determined；
Similarity is subtracted into correction value, the similarity after being adjusted.
For example, the total ratio for the quantity and the pixel of the first facial image or the second facial image that statistics obtains is
80%, similarity is 90 points, then correction value is set into 50 points, similar after the adjustment of the first facial image and the second facial image
Spend and subtract 50 points, i.e., 40 points for 90 points.And for example, obtained quantity and the pixel of the first facial image or the second facial image is counted
Total ratio be 20%, similarity is 80 points, then correction value is set into 10 points, the first facial image and the second facial image
Adjustment after similarity subtract 10 points, i.e., 70 points for 80 points.
The embodiment of the present disclosure is by determining that the interference characteristic of the first facial image and specified facial image is done to similarity
Value is disturbed, and the similarity of the first facial image and specified facial image is adjusted according to the interference value, is avoided due to the first face figure
There is face in deep frame glasses, or the first facial image and specified facial image to have in picture and specified facial image on face
The reasons such as same or analogous hair style, caused by by similarity not to be that two very high faces are mistaken for similarity higher, improve
The accuracy rate of identification.
Fig. 3 is a kind of block diagram of face identification device according to an exemplary embodiment, suitable for judging two people
The similarity of face image, as shown in figure 3, the device includes acquisition module 301, identification module 302, interference determining module 303 and
Correcting module 304.
The acquisition module 301 is configured as obtaining the first facial image.
The identification module 302 is configured to determine that the similarity of the first facial image and specified facial image.
The interference determining module 303 is configured to determine that the interference characteristic of the first facial image and specified facial image to phase
Like the interference value of degree.
The correcting module 304 is configured as, according to interference value, adjusting similarity.
The embodiment of the present disclosure is by determining that the interference characteristic of the first facial image and specified facial image is done to similarity
Value is disturbed, and the similarity of the first facial image and specified facial image is adjusted according to the interference value, is avoided due to the first face figure
There is face in deep frame glasses, or the first facial image and specified facial image to have in picture and specified facial image on face
The reasons such as same or analogous hair style, caused by by similarity not to be that two very high faces are mistaken for similarity higher, improve
The accuracy rate of identification.
Fig. 4 is the block diagram of another face identification device according to an exemplary embodiment, suitable for judging two
The similarity of facial image, as shown in figure 4, the device includes acquisition module 401, identification module 402, interference determining module 403
With correcting module 404.
The acquisition module 401 is configured as obtaining the first facial image.
The identification module 402 is configured to determine that the similarity of the first facial image and specified facial image.
The interference determining module 403 is configured to determine that the interference characteristic of the first facial image and specified facial image to phase
Like the interference value of degree.
The correcting module 404 is configured as, according to interference value, adjusting similarity.
In a kind of implementation of the present embodiment, the interference determining module 403 can include alignment unit 4031, region
Determining unit 4032 and interference calculation unit 4033.
The alignment unit 4031 is configured as the first facial image and specified the facial image average shape with setting respectively
Shape model aligns.
The area determination unit 4032 is configured to the first facial image after alignment and the nominator after alignment
Face image carries out skin analysis, it is determined that the non-colour of skin area of the specified facial image after the first facial image and alignment after alignment
Domain.
The interference calculation unit 4033 is configured as according to the first facial image after alignment and the specified face after alignment
Interference value of the interference characteristic of the non-area of skin color of image, the first facial image of calculating and specified facial image to similarity.
The area determination unit 4032 can be used for,
Choose in the first facial image after alignment and the specified facial image after alignment, and in average shape faceform
Set region corresponding to area of skin color as the first area of skin color；
Extract the features of skin colors of the first area of skin color, and by the first facial image after alignment and the specified face after alignment
In image, the features of skin colors identical region of features of skin colors and the first area of skin color is defined as the second area of skin color；
By the first facial image after alignment and alignment after specified facial image in, except the first area of skin color and the second skin
All areas beyond color region, the non-colour of skin area as the first facial image after alignment and the specified facial image after alignment
Domain.
The interference calculation unit 4033 can be used for,
Respectively by the pixel of the first facial image after alignment and the non-area of skin color of the specified facial image after alignment
Characteristic value is taken as 1, and the characteristic value of the area of skin color of the first facial image after alignment and the specified facial image after alignment is taken as
0；
Image friendship is carried out to the first facial image after alignment and the specified facial image after alignment；
The characteristic value of pixel is 1 corresponding to the specified facial image after the first facial image and alignment after statistics alignment
Pixel quantity；
The quantity that counting statistics obtains with align after the first facial image or the specified facial image after aliging pixel
Total ratio, obtain interference value of the interference characteristic to similarity of the first facial image and specified facial image.
In another implementation of the present embodiment, the correcting module 404 can include correction value determining unit 4041
With score calculating unit 4042.
The correction value determining unit 4041 is configured as, according to predetermined functional relation, according to interference value, determining similarity
Correction value.
The score calculating unit 4042 is configured as similarity subtracting correction value, the similarity after being adjusted.
The embodiment of the present disclosure is by determining that the interference characteristic of the first facial image and specified facial image is done to similarity
Value is disturbed, and the similarity of the first facial image and specified facial image is adjusted according to the interference value, is avoided due to the first face figure
There is face in deep frame glasses, or the first facial image and specified facial image to have in picture and specified facial image on face
The reasons such as same or analogous hair style, caused by by similarity not to be that two very high faces are mistaken for similarity higher, improve
The accuracy rate of identification.
On the device in above-described embodiment, wherein modules perform the concrete mode of operation in relevant this method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
Fig. 5 is a kind of block diagram of device 800 for face identification method according to an exemplary embodiment.Example
Such as, device 800 can be mobile phone, computer, digital broadcast terminal, messaging devices, game console, and flat board is set
It is standby, Medical Devices, body-building equipment, personal digital assistant etc..
Reference picture 5, device 800 can include following one or more assemblies：Processing component 802, memory 804, power supply
Component 806, multimedia groupware 808, audio-frequency assembly 810, I/O (Input/Output, input/output) interface 812, sensor
Component 814, and communication component 816.
The integrated operation of the usual control device 800 of processing component 802, such as communicated with display, call, data, phase
The operation that machine operates and record operation is associated.Treatment element 802 can refer to including one or more processors 820 to perform
Order, to complete all or part of step of above-mentioned method.In addition, processing component 802 can include one or more modules, just
Interaction between processing component 802 and other assemblies.For example, processing component 802 can include multi-media module, it is more to facilitate
Interaction between media component 808 and processing component 802.
Memory 804 is configured as storing various types of data to support the operation in equipment 800.These data are shown
Example includes the instruction of any application program or method for being operated on device 800, contact data, telephone book data, disappears
Breath, picture, video etc..Memory 804 can be by any kind of volatibility or non-volatile memory device or their group
Close and realize, such as SRAM (Static Random Access Memory, static RAM), EEPROM
(Electrically Erasable Programmable Read-Only Memory, the read-only storage of electrically erasable
Device), EPROM (Erasable Programmable Read Only Memory, Erasable Programmable Read Only Memory EPROM), PROM
(Programmable Read-Only Memory, programmable read only memory), and ROM (Read-Only Memory, it is read-only to deposit
Reservoir), magnetic memory, flash memory, disk or CD.
Electric power assembly 806 provides electric power for the various assemblies of device 800.Electric power assembly 806 can include power management system
System, one or more power supplys, and other components associated with generating, managing and distributing electric power for device 800.
Multimedia groupware 808 is included in the screen of one output interface of offer between the device 800 and user.At some
In embodiment, screen can include LCD (Liquid Crystal Display, liquid crystal display) and TP (Touch Panel,
Touch panel).If screen includes touch panel, screen may be implemented as touch-screen, be believed with receiving the input from user
Number.Touch panel includes one or more touch sensors with the gesture on sensing touch, slip and touch panel.The touch passes
Sensor can the not only border of sensing touch or sliding action, but also when detecting related to the touch or slide lasting
Between and pressure.In certain embodiments, multimedia groupware 808 includes a front camera and/or rear camera.Work as equipment
800 are in operator scheme, and during such as screening-mode or video mode, front camera and/or rear camera can receive outside
Multi-medium data.Each front camera and rear camera can be a fixed optical lens system or have focal length
And optical zoom ability.
Audio-frequency assembly 810 is configured as output and/or input audio signal.For example, audio-frequency assembly 810 includes a MIC
(Microphone, microphone), when device 800 is in operator scheme, such as call model, logging mode and speech recognition mode
When, microphone is configured as receiving external audio signal.The audio signal received can be further stored in memory 804
Or sent via communication component 816.In certain embodiments, audio-frequency assembly 810 also includes a loudspeaker, for exporting audio
Signal.
I/O interfaces 812 provide interface between processing component 802 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include but be not limited to：Home button, volume button, start button and lock
Determine button.
Sensor cluster 814 includes one or more sensors, and the state for providing various aspects for device 800 is commented
Estimate.For example, sensor cluster 814 can detect opening/closed mode of equipment 800, the relative positioning of component, such as the group
Part is the display and keypad of device 800, and sensor cluster 814 can be with 800 1 components of detection means 800 or device
Position changes, the existence or non-existence that user contacts with device 800, the orientation of device 800 or acceleration/deceleration and the temperature of device 800
Degree change.Sensor cluster 814 can include proximity transducer, be configured to detect in no any physical contact attached
The presence of nearly object.Sensor cluster 814 can also include optical sensor, such as CMOS (Complementary Metal Oxide
Semiconductor, complementary metal oxide semiconductor) or CCD (Charge-coupled Device, charge coupled cell)
Imaging sensor, for being used in imaging applications.In certain embodiments, the sensor cluster 814 can also include accelerating
Spend sensor, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 816 is configured to facilitate the communication of wired or wireless way between device 800 and other equipment.Device
800 can access the wireless network based on communication standard, such as WiFi (Wireless Fidelity, adopting wireless fidelity technology), 2G
(Second Generation mobile communication technology, second generation mobile communication technology) or 3G
(3rd Generation mobile communication technology, third generation mobile technology), or they
Combination.In one exemplary embodiment, communication component 816 is received from the wide of external broadcasting management system via broadcast channel
Broadcast signal or broadcast related information.In one exemplary embodiment, the communication component 816 also includes NFC (Near Field
Communication, near-field communication) module, to promote junction service.For example, RFID (Radio can be based in NFC module
Frequency Identification, radio frequency identification) technology, IrDA (Infrared Data Association, infrared number
According to association) technology, UWB (Ultra Wideband, ultra wide band) technology, BT (Blue Tooth, bluetooth) technologies and other technologies
To realize.
In the exemplary embodiment, device 800 can be by one or more ASIC (Application Specific
Integrated Circuit, application specific integrated circuit), DSP (Digital Signal Processing, at data signal
Manage device), DSPD (Digital Signal Processing Device, digital signal processing appts), PLD
(Programmable Logic Device, PLD), FPGA (Field-Programmable Gate
Array, field programmable gate array), controller, microcontroller, microprocessor or other electronic components realize, for execution
State method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided
Such as include the memory 804 of instruction, above-mentioned instruction can be performed to complete the above method by the processor 820 of device 800.For example,
The non-transitorycomputer readable storage medium can be ROM, RAM (Ramdom Access Memory, random access memory
Device), CD-ROM (Compact Disc Read-Only Memory, compact disc read-only memory), tape, floppy disk and light data deposit
Store up equipment etc..
A kind of non-transitorycomputer readable storage medium, when the instruction in the storage medium is by terminal (intelligent television)
During computing device so that terminal can perform a kind of face identification method, and this method includes：
Obtain the first facial image；
Determine the similarity of the first facial image and specified facial image；
Determine interference value of the interference characteristic to similarity of the first facial image and specified facial image；
According to interference value, similarity is adjusted.
In a kind of implementation of the present embodiment, the interference characteristic pair of the first facial image and specified facial image is determined
The interference value of similarity, including：
First facial image and specified facial image are alignd with the average shape model of setting respectively；
Skin analysis is carried out to the first facial image after alignment and the specified facial image after alignment respectively, it is determined that alignment
The non-area of skin color of the first facial image afterwards and the specified facial image after alignment；
According to the first facial image after alignment and the non-area of skin color of the specified facial image after alignment, calculate the first
Interference value of the interference characteristic of face image and specified facial image to similarity.
In another implementation of the present embodiment, respectively to specifying after the first facial image after alignment and alignment
Facial image carries out skin analysis, it is determined that the non-colour of skin area of the specified facial image after the first facial image and alignment after alignment
Domain, including：
Choose in the first facial image after alignment and the specified facial image after alignment, and in average shape faceform
Set region corresponding to area of skin color as the first area of skin color；
Extract the features of skin colors of the first area of skin color, and by the first facial image after alignment and the specified face after alignment
In image, the features of skin colors identical region of features of skin colors and the first area of skin color is defined as the second area of skin color；
By the first facial image after alignment and alignment after specified facial image in, except the first area of skin color and the second skin
All areas beyond color region, the non-colour of skin area as the first facial image after alignment and the specified facial image after alignment
Domain.
In another implementation of the present embodiment, according to the first facial image after alignment and the nominator after alignment
The non-area of skin color of face image, interference value of the interference characteristic to similarity of the first facial image and specified facial image is calculated,
Including：
Respectively by the pixel of the first facial image after alignment and the non-area of skin color of the specified facial image after alignment
Characteristic value is taken as 1, and the characteristic value of the area of skin color of the first facial image after alignment and the specified facial image after alignment is taken as
0；
Image friendship is carried out to the first facial image after alignment and the specified facial image after alignment；
The characteristic value of pixel is 1 corresponding to the specified facial image after the first facial image and alignment after statistics alignment
Pixel quantity；
The quantity that counting statistics obtains with align after the first facial image or the specified facial image after aliging pixel
Total ratio, obtain interference value of the interference characteristic to similarity of the first facial image and specified facial image.
In another implementation of the present embodiment, according to interference value, similarity is adjusted, including：
According to predetermined functional relation, according to interference value, the correction value of similarity is determined；
Similarity is subtracted into correction value, the similarity after being adjusted.
Those skilled in the art will readily occur to the present invention its after considering specification and putting into practice invention disclosed herein
Its embodiment.The application be intended to the present invention any modification, purposes or adaptations, these modifications, purposes or
Person's adaptations follow the general principle of the present invention and including the undocumented common knowledges in the art of the disclosure
Or conventional techniques.Description and embodiments are considered only as exemplary, and true scope and spirit of the invention are by following
Claim is pointed out.
It should be appreciated that the invention is not limited in the precision architecture for being described above and being shown in the drawings, and
And various modifications and changes can be being carried out without departing from the scope.The scope of the present invention is only limited by appended claim.
Claims (7)
1. a kind of face identification method, the similarity suitable for judging two facial images, it is characterised in that including：
Obtain the first facial image；
Determine the similarity of first facial image and specified facial image；
Determine interference value of the interference characteristic to the similarity of first facial image and the specified facial image；
According to the interference value, the similarity is adjusted；
The interference characteristic for determining first facial image and the specified facial image to the interference value of the similarity,
Including：
First facial image and the specified facial image are alignd with the average shape model of setting respectively；
Skin analysis is carried out to the specified facial image after first facial image after alignment and alignment respectively, it is determined that
The non-area of skin color of the specified facial image after first facial image and alignment after alignment；
According to first facial image after the alignment and the non-area of skin color of the specified facial image after alignment, meter
Calculate interference value of the interference characteristic to similarity of first facial image and the specified facial image；
The non-colour of skin area of the specified facial image after first facial image according to after the alignment and alignment
Domain, interference value of the interference characteristic to similarity of first facial image and the specified facial image is calculated, including：
Respectively by first facial image after the alignment and the non-area of skin color of the specified facial image after alignment
The characteristic value of pixel be taken as 1, first facial image after the alignment and the specified facial image after alignment
The characteristic value of area of skin color is taken as 0；
Image friendship is carried out to the specified facial image after first facial image after alignment and alignment；
The characteristic value of pixel corresponding to the specified facial image after first facial image and alignment after statistics alignment
It is the quantity of 1 pixel；
The quantity that counting statistics obtains with align after first facial image or the specified facial image after aliging
The total ratio of pixel, the interference characteristic for obtaining first facial image and the specified facial image are done to similarity
Disturb value.
2. according to the method for claim 1, it is characterised in that it is described respectively to first facial image after alignment and
The specified facial image after alignment carries out skin analysis, it is determined that the institute after first facial image and alignment after alignment
The non-area of skin color of specified facial image is stated, including：
Choose in first facial image after alignment and the specified facial image after alignment, with the average shape people
Region corresponding to area of skin color is set as the first area of skin color in face model；
Extract the features of skin colors of first area of skin color, and by described in after first facial image after alignment and alignment
Specify in facial image, the features of skin colors identical region of features of skin colors and first area of skin color is defined as the second colour of skin area
Domain；
By first facial image after alignment and alignment after the specified facial image in, except first area of skin color
With all areas beyond second area of skin color, as first facial image after the alignment and alignment after institute
State the non-area of skin color of specified facial image.
3. method according to claim 1 or 2, it is characterised in that described described similar according to the interference value, adjustment
Degree, including：
According to predetermined functional relation, according to the interference value, the correction value of the similarity is determined；
The similarity is subtracted into the correction value, the similarity after being adjusted.
4. a kind of face identification device, the similarity suitable for judging two facial images, it is characterised in that including：
Acquisition module, for obtaining the first facial image；
Identification module, for determining the similarity of first facial image and specified facial image；
Determining module is disturbed, for determining the interference characteristic of first facial image and the specified facial image to the phase
Like the interference value of degree；
Correcting module, for according to the interference value, adjusting the similarity；
The interference determining module includes：
Alignment unit, for by first facial image and the specified facial image average shape model with setting respectively
Alignment；
Area determination unit, for respectively to first facial image after alignment and the specified facial image after alignment
Skin analysis is carried out, it is determined that the non-colour of skin area of the specified facial image after first facial image and alignment after alignment
Domain；
Interference calculation unit, for according to first facial image after the alignment and the specified face figure after alignment
The non-area of skin color of picture, calculate interference of the interference characteristic of first facial image and the specified facial image to similarity
Value；
The interference calculation unit is used for,
Respectively by first facial image after the alignment and the non-area of skin color of the specified facial image after alignment
The characteristic value of pixel be taken as 1, first facial image after the alignment and the specified facial image after alignment
The characteristic value of area of skin color is taken as 0；
Image friendship is carried out to the specified facial image after first facial image after alignment and alignment；
The characteristic value of pixel corresponding to the specified facial image after first facial image and alignment after statistics alignment
It is the quantity of 1 pixel；
The quantity that counting statistics obtains with align after first facial image or the specified facial image after aliging
The total ratio of pixel, the interference characteristic for obtaining first facial image and the specified facial image are done to similarity
Disturb value.
5. device according to claim 4, it is characterised in that the area determination unit is used for,
Choose in first facial image after alignment and the specified facial image after alignment, with the average shape people
Region corresponding to area of skin color is set as the first area of skin color in face model；
Extract the features of skin colors of first area of skin color, and by described in after first facial image after alignment and alignment
Specify in facial image, the features of skin colors identical region of features of skin colors and first area of skin color is defined as the second colour of skin area
Domain；
By first facial image after alignment and alignment after the specified facial image in, except first area of skin color
With all areas beyond second area of skin color, as first facial image after the alignment and alignment after institute
State the non-area of skin color of specified facial image.
6. the device according to claim 4 or 5, it is characterised in that the correcting module includes：
Correction value determining unit, for according to predetermined functional relation, according to the interference value, determining the amendment of the similarity
Value；
Score calculating unit, for the similarity to be subtracted into the correction value, the similarity after being adjusted.
7. a kind of face identification device, the similarity suitable for judging two facial images, it is characterised in that including：
Processor；
For storing the memory of processor-executable instruction；
Wherein, the processor is configured as：
Obtain the first facial image；
Determine the similarity of first facial image and specified facial image；
Determine interference value of the interference characteristic to the similarity of first facial image and the specified facial image；
According to the interference value, the similarity is adjusted；
The interference characteristic for determining first facial image and the specified facial image to the interference value of the similarity,
Including：
First facial image and the specified facial image are alignd with the average shape model of setting respectively；
Skin analysis is carried out to the specified facial image after first facial image after alignment and alignment respectively, it is determined that
The non-area of skin color of the specified facial image after first facial image and alignment after alignment；
According to first facial image after the alignment and the non-area of skin color of the specified facial image after alignment, meter
Calculate interference value of the interference characteristic to similarity of first facial image and the specified facial image；
The non-colour of skin area of the specified facial image after first facial image according to after the alignment and alignment
Domain, interference value of the interference characteristic to similarity of first facial image and the specified facial image is calculated, including：
Respectively by first facial image after the alignment and the non-area of skin color of the specified facial image after alignment
The characteristic value of pixel be taken as 1, first facial image after the alignment and the specified facial image after alignment
The characteristic value of area of skin color is taken as 0；
Image friendship is carried out to the specified facial image after first facial image after alignment and alignment；
The characteristic value of pixel corresponding to the specified facial image after first facial image and alignment after statistics alignment
It is the quantity of 1 pixel；
The quantity that counting statistics obtains with align after first facial image or the specified facial image after aliging
The total ratio of pixel, the interference characteristic for obtaining first facial image and the specified facial image are done to similarity
Disturb value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410240747.4A CN104077563B (en) | 2014-05-30 | 2014-05-30 | Face identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410240747.4A CN104077563B (en) | 2014-05-30 | 2014-05-30 | Face identification method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104077563A CN104077563A (en) | 2014-10-01 |
| CN104077563B true CN104077563B (en) | 2018-02-02 |
Family
ID=51598811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410240747.4A Active CN104077563B (en) | 2014-05-30 | 2014-05-30 | Face identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104077563B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6572629B2 (en) * | 2015-06-03 | 2019-09-11 | ソニー株式会社 | Information processing apparatus, information processing method, and program |
| CN104899579A (en) * | 2015-06-29 | 2015-09-09 | 小米科技有限责任公司 | Face recognition method and face recognition device |
| CN106919884A (en) * | 2015-12-24 | 2017-07-04 | 北京汉王智远科技有限公司 | Human facial expression recognition method and device |
| 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 |
| CN111339822B (en) * | 2017-07-17 | 2023-06-30 | Oppo广东移动通信有限公司 | Living body detection method and related products |
| CN109272006A (en) * | 2017-07-18 | 2019-01-25 | 北京柯斯元科技有限公司 | Anti-counterfeit sign with random texture pattern decision-making system and determination method |
| CN110462632A (en) * | 2019-06-27 | 2019-11-15 | 深圳市汇顶科技股份有限公司 | The method, apparatus and electronic equipment of recognition of face |
| CN110532746B (en) * | 2019-07-24 | 2021-07-23 | 创新先进技术有限公司 | Face checking method, device, server and readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102034079A (en) * | 2009-09-24 | 2011-04-27 | 汉王科技股份有限公司 | Method and system for identifying faces shaded by eyeglasses |
| CN103020579A (en) * | 2011-09-22 | 2013-04-03 | 上海银晨智能识别科技有限公司 | Face recognition method and system, and removing method and device for glasses frame in face image |
| CN103136533A (en) * | 2011-11-28 | 2013-06-05 | 汉王科技股份有限公司 | Face recognition method and device based on dynamic threshold value |
-
2014
- 2014-05-30 CN CN201410240747.4A patent/CN104077563B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102034079A (en) * | 2009-09-24 | 2011-04-27 | 汉王科技股份有限公司 | Method and system for identifying faces shaded by eyeglasses |
| CN103020579A (en) * | 2011-09-22 | 2013-04-03 | 上海银晨智能识别科技有限公司 | Face recognition method and system, and removing method and device for glasses frame in face image |
| CN103136533A (en) * | 2011-11-28 | 2013-06-05 | 汉王科技股份有限公司 | Face recognition method and device based on dynamic threshold value |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104077563A (en) | 2014-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104077563B (en) | Face identification method and device | |
| CN104156947B (en) | Image partition method, device and equipment | |
| CN104408402B (en) | Face identification method and device | |
| CN107454465A (en) | Video playback progress display method and device, electronic equipment | |
| CN104700353B (en) | Image filters generation method and device | |
| CN107123081A (en) | image processing method, device and terminal | |
| CN106503617A (en) | Model training method and device | |
| CN106548145A (en) | Image-recognizing method and device | |
| CN104036240B (en) | The localization method and device of human face characteristic point | |
| CN107122679A (en) | Image processing method and device | |
| CN105760884B (en) | The recognition methods of picture type and device | |
| CN106934320B (en) | Fingerprint identification method and device | |
| CN105512605A (en) | Face image processing method and device | |
| CN105357425B (en) | Image capturing method and device | |
| CN106295515A (en) | Determine the method and device of human face region in image | |
| CN106971164A (en) | Shape of face matching process and device | |
| CN105528078B (en) | The method and device of controlling electronic devices | |
| CN106980840A (en) | Shape of face matching process, device and storage medium | |
| CN107958223B (en) | Face recognition method and device, mobile equipment and computer readable storage medium | |
| CN107480665A (en) | Character detecting method, device and computer-readable recording medium | |
| CN103886284B (en) | Character attribute information identifying method, device and electronic equipment | |
| CN105354560A (en) | Fingerprint identification method and device | |
| CN106295499A (en) | Age estimation method and device | |
| CN107213642A (en) | Virtual portrait outward appearance change method and device | |
| CN104408404A (en) | Face identification method and apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |