CN110462630A - For the optical sensor of recognition of face, device, method and electronic equipment - Google Patents
For the optical sensor of recognition of face, device, method and electronic equipment Download PDFInfo
- Publication number
- CN110462630A CN110462630A CN201980000834.3A CN201980000834A CN110462630A CN 110462630 A CN110462630 A CN 110462630A CN 201980000834 A CN201980000834 A CN 201980000834A CN 110462630 A CN110462630 A CN 110462630A
- Authority
- CN
- China
- Prior art keywords
- pixel unit
- pixel
- facial image
- face
- class
- 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
- 230000003287 optical effect Effects 0.000 title claims abstract description 234
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000001815 facial effect Effects 0.000 claims abstract description 228
- 230000004044 response Effects 0.000 claims description 172
- 238000001228 spectrum Methods 0.000 claims description 86
- 230000003595 spectral effect Effects 0.000 claims description 55
- 238000013135 deep learning Methods 0.000 claims description 17
- 238000012545 processing Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 7
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 238000003786 synthesis reaction Methods 0.000 claims description 6
- 230000015654 memory Effects 0.000 description 26
- 238000013527 convolutional neural network Methods 0.000 description 11
- 238000003860 storage Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000001360 synchronised effect Effects 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000009434 installation Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 239000012237 artificial material Substances 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 102000001554 Hemoglobins Human genes 0.000 description 1
- 108010054147 Hemoglobins Proteins 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 229910003460 diamond Inorganic materials 0.000 description 1
- 239000010432 diamond Substances 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 230000035800 maturation Effects 0.000 description 1
- 230000036564 melanin content Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000007540 photo-reduction reaction Methods 0.000 description 1
- 238000007639 printing Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
- 239000012780 transparent material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/162—Detection; Localisation; Normalisation using pixel segmentation or colour matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Abstract
It is a kind of for the optical sensor of recognition of face, device, method and electronic equipment, it is able to ascend the safety of recognition of face, the optical sensor for being used for recognition of face includes: pixel array, the first pixel unit set in the pixel array includes first kind pixel unit group and the second class pixel unit group, wherein: the first kind pixel unit group includes at least one first kind pixel unit, the first optical filter is arranged in the first kind pixel unit, and first optical filter is used for the optical signal by first band range;The second class pixel unit group includes at least one second class pixel unit, the second optical filter is arranged in the second class pixel unit, second optical filter is used for the optical signal by second band range, and the second band range is different from the first band range;Pixel unit in the first kind pixel unit group and the second class pixel unit group is for receiving the reflected light signal that optical signal launched by light source is reflected from face, and local facial image is obtained according to the reflected light signal, the local facial image is for determining the true and false of the face.
Description
Technical field
This application involves technical field of face recognition, and pass more particularly, to a kind of optics for recognition of face
Sensor, device, method and electronic equipment.
Background technique
Safe and convenient and fast user experience is brought to user using the electronic equipment of face recognition technology still to pass through
User picture (for example, printing or electronics), or the human face data of the forgeries such as 3D face mold of manufacture is recognition of face
A security risk in.Therefore, true and false face how is identified, being one to promote the safety of recognition of face needs to solve
Certainly the problem of.
Summary of the invention
The embodiment of the present application provides a kind of for the optical sensor of recognition of face, device, method and electronic equipment, energy
Identify the true and false of face, enough so as to promote the safety of recognition of face.
In a first aspect, providing a kind of optical sensor for recognition of face, comprising: pixel array, the pixel battle array
The first pixel unit set in column includes first kind pixel unit group and the second class pixel unit group, in which:
The first kind pixel unit group includes at least one first kind pixel unit, the first kind pixel unit setting
First optical filter, first optical filter are used for the optical signal by first band range;
The second class pixel unit group includes at least one second class pixel unit, the second class pixel unit setting
Second optical filter, second optical filter is used for the optical signal by second band range, and the second band range is different
In the first band range;
Pixel unit in the first kind pixel unit group and the second class pixel unit group is for receiving by light source
The reflected light signal that the optical signal of transmitting is reflected from face, and local facial image is obtained according to the reflected light signal, it is described
Local facial image is for determining the true and false of the face.
In some possible implementations, the first pixel unit set further includes third class pixel unit group, institute
Stating third class pixel unit group includes at least one third class pixel unit, and the third class pixel unit setting third filters
Piece, the third optical filter are used for the optical signal by third wavelength band, and the third wavelength band is different from described first
Wavelength band and the second band range, the first kind pixel unit group, the second class pixel unit group and described
Pixel unit in three classes pixel unit group is for receiving the reflection that the optical signal emitted by the light source is reflected from the face
Optical signal, and local facial image is obtained according to the reflected light signal, the local facial image is for determining the face
It is true and false.
In some possible implementations, the first band range, the second band range and the third wave
Segment limit is respectively one of following three kinds of wavelength bands:
Wavelength band including 560nm, the wavelength band including 980nm, the wavelength band including 940nm.
In some possible implementations, the wavelength band of the optical signal of the light source transmitting includes the first band
Range, the second band range and the third wavelength band.
In some possible implementations, the quantity of continuous pixel unit is less than in the first pixel unit set
Or it is equal to first threshold.
In some possible implementations, the quantity of the pixel unit in the first pixel unit set and the picture
The ratio of the total quantity of pixel unit in pixel array is less than the first ratio.
In some possible implementations, pixel unit in the first pixel unit set is discrete be distributed in it is described
In pixel array.
In some possible implementations, other in the pixel array in addition to the first pixel unit set
The facial image of pixel unit acquisition is used for recognition of face.
In some possible implementations, other in the pixel array in addition to the first pixel unit set
Pixel unit is not provided with optical filter.
In some possible implementations, other in the pixel array in addition to the first pixel unit set
The optical filter of pixel unit setting specific band range.
In some possible implementations, the optical filter of the specific band range be include 940nm wavelength band
Optical filter.
Second aspect provides a kind of device for recognition of face, comprising:
Such as the optical sensing for recognition of face in first aspect to second aspect and its any possible implementation
Device, wherein first kind pixel unit group and the second class pixel unit in the first pixel unit set of the optical sensor
Pixel unit in group is used to receive the reflected light signal that optical signal launched by light source is reflected from face, and according to the reflection
Optical signal obtains local facial image;
Processor, for determining the true and false of the face according to the local facial image.
In some possible implementations, the first pixel unit set includes first kind pixel unit group, and second
Class pixel unit group and third class pixel unit group, in which:
The first kind pixel unit group includes at least one first kind pixel unit, the first kind pixel unit setting
First optical filter, first optical filter are used for the optical signal by first band range;
The second class pixel unit group includes at least one second class pixel unit, the second class pixel unit setting
Second optical filter, second optical filter is used for the optical signal by second band range, and the second band range is different
In the first band range;
The third class pixel unit group includes at least one third class pixel unit, the third class pixel unit setting
Third optical filter, the third optical filter are used to be different from by the optical signal of third wavelength band, the third wavelength band
The first band range and the second band range;
Wherein, the first kind pixel unit group, the second class pixel unit group and the third class pixel unit group
In pixel unit for receiving the reflected light signal that the optical signal emitted by the light source is reflected from face, and according to it is described instead
It penetrates optical signal and obtains local facial image, the local facial image is for determining the true and false of the face.
In some possible implementations, the processor is also used to:
According to calibration parameter, to the first partial facial image of first kind pixel unit group acquisition, second class
Second local facial image of pixel unit group acquisition and the third local facial image of third class pixel unit group acquisition
It is calibrated;
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image determines the true and false of the face.
In some possible implementations, the processor is also used to:
According to references object to the first band range, the light of the second band range and the third wavelength band
Relationship between the response of spectrum determines the calibration parameter.In some possible implementations, the optical sensor is also used
In:
When the light source emits optical signal to the references object, by the first kind pixel unit group, described the
Two class pixel unit groups and the third class pixel unit group acquire the references object to the first band range, institute respectively
State the response of the spectrum of second band range and the third wavelength band;
The processor is also used to: according to the references object to the first band range, the second band range
With the response of the spectrum of the third wavelength band, determine that the calibration parameter, the calibration parameter are used for so that described first
The references object of class pixel unit group, the second class pixel unit group and third class pixel unit group acquisition is to institute
State first band range, the response of the spectrum of the second band range and the third wavelength band is in same range.
In some possible implementations, the references object is white object, or yellowish pink object.
In some possible implementations, the processor is also used to:
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image is synthesized, and colored local facial image is obtained, wherein each pixel in the colour local facial image includes
The face is to the first band range, three responses of the spectrum of the second band range and the third wavelength band
Pixel value;
The colored local facial image is subjected to feature extraction, obtains the feature letter of the colored local facial image
Breath;
According to the characteristic information of the colored local facial image, the true and false of the face is determined.
In some possible implementations, the processor is also used to:
Pixel value is used according to what first kind pixel unit in the first kind pixel unit group acquired, determines the colour
The corresponding first response pixel value of the first pixel in the image of local facial, wherein described in the first response pixel value expression
Response of the face to the spectrum of the first band range;
According to the pixel value that the second class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding second response pixel value of pixel, wherein the second response pixel value indicates the face to the second band model
The spectral response enclosed;And
According to the pixel value that the third class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding third of pixel responds pixel value, and the third response pixel value indicates the face to the light of the third wavelength band
Spectrum response.
In some possible implementations, the processor is specifically used for:
It is handled by characteristic information of the deep learning network to the colored local facial image, determines the face
It is true and false.
In some possible implementations, the processor is also used to:
From in the facial image of multiple real human faces that the optical sensor acquires and false face, described first is extracted
Multiple local facial images of pixel unit set acquisition;
Calibration and synthesis processing are carried out to the multiple local facial image, obtain multiple colored local facial images;
The multiple colored local facial image is input to deep learning network to be trained, obtains the deep learning
The model and parameter of network.
In some possible implementations, the processor is also used to:
According to the face of other pixel units acquisition in the pixel array in addition to the first pixel unit set
Image carries out recognition of face.
The third aspect provides a kind of method for recognition of face, comprising:
By pixel unit in the first pixel unit set of optical sensor receive optical signal launched by light source from
The reflected light signal of face reflection, and local facial image is obtained according to the reflected light signal;
Wherein, the first pixel unit set includes first kind pixel unit group and the second class pixel unit group, described
First kind pixel unit group includes at least one first kind pixel unit, and the first optical filter is arranged in the first kind pixel unit,
First optical filter is used for the optical signal by first band range;The second class pixel unit group include at least one
The second optical filter is arranged in two class pixel units, the second class pixel unit, and second optical filter is for passing through second band
The optical signal of range, and the second band range is different from the first band range;
According to the local facial image for determining the true and false of the face.
The first pixel unit set described in some possible implementations further includes third class pixel unit group, described
Third class pixel unit group includes at least one third class pixel unit, and third optical filter is arranged in the third class pixel unit,
The third optical filter is used for the optical signal by third wavelength band, and the third wavelength band is different from the first band
Range and the second band range.
In some possible implementations, the first band range, the second band range and the third wave
Segment limit is respectively one of following three kinds of wavelength bands:
Wavelength band including 560nm, the wavelength band including 980nm, the wavelength band including 940nm.
The wavelength band of the optical signal of the transmitting of the light source described in some possible implementations includes the first band
Range, the second band range and the third wavelength band.
According to the local facial image for determining the true and false of the face described in some possible implementations,
Include:
According to calibration parameter, to the first partial facial image of first kind pixel unit group acquisition, second class
Second local facial image of pixel unit group acquisition and the third local facial image of third class pixel unit group acquisition
It is calibrated;
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image determines the true and false of the face.
In some possible implementations, the method also includes:
According to references object to the first band range, the light of the second band range and the third wavelength band
Relationship between the response of spectrum determines the calibration parameter.
In some possible implementations, it is described according to references object to the first band range, second wave
Relationship between segment limit and the response of the spectrum of the third wavelength band, determines the calibration parameter, comprising:
When the light source emits optical signal to the references object, by the first kind pixel unit group, described the
Two class pixel unit groups and the third class pixel unit group acquire the references object to the first band range, institute respectively
State the response of the spectrum of second band range and the third wavelength band;
According to the references object to the first band range, the second band range and the third wavelength band
Spectrum response, determine the calibration parameter, the calibration parameter is used for so that the first kind pixel unit group, described the
Two class pixel unit groups and the references object of third class pixel unit group acquisition are described to the first band range
The response of the spectrum of second band range and the third wavelength band is in same range.
According to the first partial facial image after calibration, the second game described in some possible implementations
Portion's facial image and third local facial image determine the true and false of the face, comprising:
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image is synthesized, and colored local facial image is obtained, wherein each pixel in the colour local facial image includes
The face is to the first band range, three responses of the spectrum of the second band range and the third wavelength band
Pixel value;
The colored local facial image is subjected to feature extraction, obtains the feature letter of the colored local facial image
Breath;
According to the characteristic information of the colored local facial image, the true and false of the face is determined.
According to the first partial facial image after calibration, the second game described in some possible implementations
Portion's facial image and third local facial image are synthesized, and colored local facial image is obtained, comprising:
Pixel value is used according to what first kind pixel unit in the first kind pixel unit group acquired, determines the colour
The corresponding first response pixel value of the first pixel in the image of local facial, wherein described in the first response pixel value expression
Response of the face to the spectrum of the first band range;
According to the pixel value that the second class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding second response pixel value of pixel, wherein the second response pixel value indicates the face to the second band model
The spectral response enclosed;And
According to the pixel value that the third class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding third of pixel responds pixel value, and the third response pixel value indicates the face to the light of the third wavelength band
Spectrum response.
According to the characteristic information of the colored local facial image described in some possible implementations, determine described in
Face it is true and false, comprising:
It is handled by characteristic information of the deep learning network to the colored local facial image, determines the face
It is true and false.
The method described in some possible implementations further include:
From in the facial image of multiple real human faces that the optical sensor acquires and false face, described first is extracted
Multiple local facial images of pixel unit set acquisition;
Calibration and synthesis processing are carried out to the multiple local facial image, obtain multiple colored local facial images;
The multiple colored local facial image is input to deep learning network to be trained, obtains the deep learning
The model and parameter of network.
The method described in some possible implementations further include:
According to the face of other pixel units acquisition in the pixel array in addition to the first pixel unit set
Image carries out recognition of face.
Described in some possible implementations according in the pixel array remove the first pixel unit set with
The facial image of outer other pixel units acquisition carries out recognition of face, comprising:
If the facial image is matched with the facial image registered and the face is real human face, determine recognition of face at
Function.
The quantity of continuous pixel unit is less than in the first pixel unit set described in some possible implementations
First threshold.
The quantity of pixel unit in the first pixel unit set described in some possible implementations and the picture
The ratio of the total quantity of pixel unit in pixel array is less than the first ratio.
Pixel unit in the first pixel unit set described in some possible implementations is discrete be distributed in it is described
In pixel array.
Other in the pixel array described in some possible implementations in addition to the first pixel unit set
The facial image of pixel unit acquisition is used for recognition of face.
Other in the pixel array described in some possible implementations in addition to the first pixel unit set
Pixel unit is not provided with optical filter.
Other in the pixel array described in some possible implementations in addition to the first pixel unit set
The optical filter of pixel unit setting specific band range.
In some possible implementations, the optical filter of the specific band range be include 940nm wavelength band
Optical filter.
Fourth aspect provides a kind of electronic equipment, including in such as second aspect and its any possible implementation
Device for recognition of face.
5th aspect, provides a kind of computer-readable medium, for storing computer program, the computer program packet
It includes for executing the instruction in the above-mentioned third aspect and its any possible implementation.
6th aspect, provides a kind of computer program product including instruction, when computer runs the computer journey
When the finger of sequence product, the computer execute in the above-mentioned third aspect and its any possible implementation for face
Know method for distinguishing.
Specifically, which can run on the electronic equipment of above-mentioned fourth aspect.
Based on the above-mentioned technical proposal, different types of optical filtering is set by at least two class pixel unit groups in pixel array
Piece, so that the pixel unit that optical filter is arranged can acquire the spectral response of the wavelength band of the optical filter, in this way, once exposing
In photoreduction process, based on this, at least two class pixel unit groups can acquire at least two spectral responses, not need to carry out multi collect
Obtain at least two spectral response, be able to ascend acquisition speed, may further based at least two spectral response into
Row vivo identification is conducive to the safety for promoting recognition of face.
Detailed description of the invention
Fig. 1 is the reflection spectrum curve of human skin.
Fig. 2 is the schematic diagram according to the optical sensor for recognition of face of the embodiment of the present application.
Fig. 3 is schematic figure of the filter set in one of pixel array arrangement mode.
Fig. 4 is the schematic figure of the arrangement mode of the optical filter in filter set.
Fig. 5 is the schematic diagram according to the device for recognition of face of the embodiment of the present application.
Fig. 6 is the schematic flow chart according to the method for recognition of face of the embodiment of the present application.
Fig. 7 is the overall flow figure according to the method for recognition of face of the embodiment of the present application.
Fig. 8 is the schematic diagram according to the electronic equipment of the embodiment of the present application.
Specific embodiment
Below in conjunction with attached drawing, technical solutions in the embodiments of the present application is described.
It should be understood that the embodiment of the present application can be applied to various face identification systems, the application scenarios common as one kind,
Face identification system provided by the embodiments of the present application can be applied in the mobile terminals such as smart phone, tablet computer and door lock, door
Access control system or other electronic equipments.
In traditional face identification system, active false proof is by the way of interaction, for example, being become using blink or expression
Modes, this modes such as change usually require a few frame images of continuous acquisition, reduce recognition speed.Usually, by human skin
The influence of the factors such as skin thickness, hemoglobin concentration, the melanin content of tissue, human skin tissue is to specific band range
Light reflecting properties there is certain particularity, as shown in Figure 1, wavelength band of the skin of human body in 560nm or so,
The wavelength band of 980nm has special spectral response, light of this special spectral response in artificial materials such as paper, molds
It is not present on spectrum response curve.
Accordingly, this application provides a kind of methods of face active false proof, can be obtained by acquiring a frame image wait know
Other target is based further on the spectral response and carries out active false proof to the spectral response of specific band range, is conducive to be promoted and know
Other speed, while the safety of recognition of face can also be promoted.
It it should be understood that the target to be identified in the embodiment of the present application can be face, or may be other portions of human body
Position, such as finger, palm, the embodiment of the present application are not construed as limiting this.
Hereinafter, the Installation practice of the application is discussed in detail in conjunction with Fig. 2 to Fig. 5.
It should be understood that pixel unit group, the quantity of filter plate and arrangement side in the embodiment of the present application as shown below
Formula etc. is merely illustrative, and constitutes any restriction without coping with the application.
Fig. 2 is a kind of schematic diagram of optical sensor 20 for recognition of face provided by the embodiments of the present application,
The optical sensor 20 includes:
Pixel array 200, the first pixel unit set 21 in the pixel array 200 include first kind pixel unit group
With the second class pixel unit group, in which:
The first kind pixel unit group includes at least one first kind pixel unit 211, the first kind pixel unit
211 the first optical filters 221 of setting, first optical filter 221 are used for the optical signal by first band range;
The second class pixel unit group includes at least one second class pixel unit 212, the second class pixel unit
212 the second optical filters 222 of setting, second optical filter 222 are used for the optical signal by second band range, and described second
Wavelength band is different from the first band range;
Pixel unit in the first kind pixel unit group and the second class pixel unit group is for receiving by light source
The reflected light signal that the optical signal of transmitting is reflected from face, and local facial image is obtained according to the reflected light signal, it is described
Local facial image is for determining the true and false of the face.
In the embodiment of the present application, the pixel unit in the first pixel unit set of the optical sensor can be set to
Few two different optical filters, will be arranged the pixel unit of same optical filter as one kind pixel unit group, then first picture
Plain unit set can be divided at least two class pixel unit groups, optionally, pixel unit in a kind of pixel unit group and such
Pixel unit group, which corresponds to the corresponding relationship of the optical filter of type quantitatively, can be one-to-one or many-one, i.e. a picture
The corresponding optical filter of plain unit, or be also possible to multiple pixel units and share an optical filter.
For example, a first kind pixel unit 211 can correspond to first optical filter 221, alternatively, being also possible to multiple
First kind pixel unit 211 corresponds to first optical filter 221;
Similarly, a second class pixel unit 212 can correspond to second optical filter 222, alternatively, being also possible to more
Corresponding second optical filter 222 of a second class pixel unit 212.
Optionally, in some embodiments, optical filter is arranged in the front end optical path of the pixel unit, for example, can be with
Optical filter is arranged in the top for needing to be arranged the pixel unit of optical filter, for example, optical filter is pasted onto the pixel unit
Upper surface, as long as alternatively, filter can also be directly overlayed on the pixel unit can play filter action i.e.
Can, the embodiment of the present application is not construed as limiting this.
Optical filter in the embodiment of the present application only allows the optical signal within the scope of specific band to pass through, in other words, optical filter
It is higher to the transmitance of the optical signal within the scope of specific band, it is greater than 80% or 90%, the light of other wavelength bands is believed
Number transmitance it is lower, for example, less than 10% or 20%.
In the embodiment of the present application, the wavelength band for the optical signal that optical filter passes through can be it is specially designed, as one
A optional implementation can be designed according to the reflection spectrum curve of human skin through the wave for having special spectral response
The optical signal of segment limit, for example, the wavelength band of 560nm or so or other visible-ranges, or, the wave band of 980nm or so
Range, or the wavelength band of recognition of face performance preferably infrared band range, such as 940nm or so can also be penetrated, or
Person can also select suitable wavelength band according to the other biological feature of living body, as long as can have apparent area with prosthese
Indexing.
Therefore, in the embodiment of the present application, different-waveband is set by at least two class pixel unit groups in pixel array
The optical filter of range, so that the pixel unit that optical filter is arranged can acquire the response of the spectrum of the wavelength band of the optical filter,
In this way, based on this, at least two class pixel unit groups can acquire the response of at least two spectrum during single exposure, it is not required to
Multi collect is carried out to obtain the response of at least two spectrum, is able to ascend acquisition speed, it may further be based on this extremely
The response of few two kinds of spectrum carries out vivo identification, is conducive to the safety for promoting recognition of face.
Optionally, in some embodiments, as shown in Fig. 2, the first pixel unit set 21 can be with third class picture
Plain unit group, the third class pixel unit group include at least one third class pixel unit 213, the third class pixel unit
213 setting third optical filters 223, the third optical filter 223 are used for the optical signal by third wavelength band.
Optionally, in the application one embodiment, the first band range, the second band range and described
Three wavelength bands are respectively one of following three kinds of wavelength bands:
Wavelength band including 560nm, the wavelength band including 980nm, the wavelength band including 940nm.
Optionally, in other embodiments, above three wavelength band can be determined in visible light wave range, for example, can
The first band range is arranged, the second band range and the third wavelength band are respectively red spectral band, blue light
One kind of wave band and green light band.For example, it is 440nm~475nm, upper cut-off wave that the wavelength band of blue light, which can be center wave band,
Duan Yuewei 550nm;It is 520nm~550nm that the wavelength band of green light, which can be center wave band, and upper cut-off wave band is about 620nm, under
End wave band 460nm;It is about 550nm that the wavelength band of feux rouges, which can be lower cut-off wave band,.
It should be understood that in the embodiment of the present application, the wavelength band including 560nm can be the wave band of 560nm or so
Range, for example, the wavelength band of 560nm ± 20nm or the wavelength band of 560nm ± 40nm etc., specific wavelength band can
With by the control process of optical filter, the embodiment of the present application is not construed as limiting this, other wavelength bands are similar, here no longer
It repeats.
In the embodiment of the present application, the wavelength band of the optical signal of light source transmitting includes the first band range, described
Second band range and the third wavelength band.In this way, face is irradiated by full wave optical signal, further by a variety of
Optical filter extracts the response of multiple spectrum, and without the light source using multiple and different wave bands, save the cost reduces mould group
Complexity.
Optionally, the people of other pixel units acquisition in the pixel array in addition to the first pixel unit set
Face image is used for recognition of face, such as can be by the facial image mould of the facial image of other described pixel units acquisition and registration
Plate is matched, it is determined whether successful match.
Optionally, in the embodiment of the present application, its in the pixel array in addition to the first pixel unit set
He can be not provided with optical filter by pixel unit, for example, doing transparent processing, transparent material is perhaps arranged or spy also can be set
The optical filter of standing wave segment limit, such as the optical filter of 940nm wavelength band.
Hereinafter, combination Fig. 3 and Fig. 4, illustrate the first pixel unit set, arrangement side of the optical filter in pixel array
Formula.
Optionally, in the embodiment of the present application, continuous pixel unit in the first pixel unit set can be set
Quantity be less than or equal to specific threshold, it is continuous by optical filter in the first pixel unit set by being arranged for example, 6
The number of the pixel unit of covering is less than certain threshold value, and can be avoided influences recognition of face performance.
Optionally, in the embodiment of the present application, the quantity of the pixel unit in the first pixel unit set with it is described
The ratio of the total quantity of pixel unit in pixel array is less than the first ratio, such as 5%, to avoid recognition of face is influenced
Energy.
Optionally, pixel unit in the first pixel unit set is discrete to be distributed in the pixel array, corresponding
Ground, first optical filter, the second optical filter and third optical filter is discrete is distributed in the pixel array.
Optionally, in the application one embodiment, first optical filter, second optical filter, the third filter
Mating plate may be constructed a filter set 220, and the filter set 200 is discrete to be distributed in the pixel array 200.For example,
As shown in figure 3, the filter set 220 can be square, diamond shape, round or other rules or irregular pattern are arranged in
In the pixel array of the optical sensor, as long as not influencing recognition of face performance, the embodiment of the present application is not construed as limiting this.
Optionally, in the embodiment of the present application, the optical filter in a filter set 220 can be discrete, that is, filter
It is separated between piece by the pixel unit for being not provided with optical filter, for example, between the design method h~i or optical filter in Fig. 4
It can be continuously, such as design method a~e in Fig. 4, the embodiment of the present application, which is not particularly limited in a filter set, includes
The first optical filter, the second optical filter, the quantity and arrangement mode of third optical filter.
Fig. 5 is according to the schematic diagram of the device for recognition of face of the embodiment of the present application, as shown in figure 5, should
Device 50 for recognition of face may include:
Optical sensor 51, pixel unit in the first pixel unit set of the optical sensor 51 for receive by
The reflected light signal that the optical signal of light source transmitting is reflected from face, and local facial image is obtained according to the reflected light signal;
Processor 52, for determining the true and false of the face according to the local facial image.
The sensor 51 can be the optical sensor 20 in embodiment illustrated in fig. 2, and illustrating can be with reference to shown in Fig. 2
The related description of embodiment, which is not described herein again.
In the embodiment of the present application, the light source can be the built-in light source of described device 50, or be also possible to described
The external light source of device 50, or the light source in the electronic equipment that described device 50 is installed can also be multiplexed to emit for living
The optical signal of body identification, in this case, described device 50 can not include the light source, the embodiment of the present application does not limit this
It is fixed.
The wavelength band of the light source may include multiple wavelength bands, in some embodiments, the light letter of light source transmitting
It number can be used for recognition of face, i.e. vivo identification and recognition of face uses same light source;It is described in other alternate embodiments
Device 50 for recognition of face can also include another light source for recognition of face, for example, 940nm or so wavelength band
Light source, in this case, can pass through the pixel unit in the first pixel unit set acquire topography carry out living body knowledge
Not, or vivo identification can also be carried out by the facial image that entire pixel array acquires.
Optionally, as one embodiment, the first pixel unit set includes first kind pixel unit group, the second class
Pixel unit group and third class pixel unit group, in which:
The first kind pixel unit group includes at least one first kind pixel unit, the first kind pixel unit setting
First optical filter, first optical filter are used for the optical signal by first band range;
The second class pixel unit group includes at least one second class pixel unit, the second class pixel unit setting
Second optical filter, second optical filter is used for the optical signal by second band range, and the second band range is different
In the first band range;
The third class pixel unit group includes at least one third class pixel unit, the third class pixel unit setting
Third optical filter, the third optical filter are used to be different from by the optical signal of third wavelength band, the third wavelength band
The first band range and the second band range.
Based on above-mentioned set-up mode, first kind pixel unit group can obtain target to be identified to the by the first optical filter
The response (or response of spectrum) of the optical signal of one wavelength band, the second class pixel unit group can be obtained by the second optical filter
Response of the target to be identified to the spectrum of second band range is taken, the third class pixel unit group can pass through third optical filter
Obtain response of the target to be identified to the spectrum of third wavelength band, that is to say, that these three types of pixel unit groups can obtain respectively
Obtain the response of three kinds of different spectrum.
As an optional implementation, the true and false of face, example can be determined according to the response of above-mentioned different spectrum
Such as, the local facial image of the response for the above-mentioned three kinds of spectrum of reflection that every class pixel unit group acquires can be input to and is trained
Convolutional neural networks classify, it is determined whether come from real human face.Alternatively, above-mentioned three classes pixel unit group can also be adopted
The face of collection is compared the response of three kinds of spectrum with the pixel value that the pixel unit of neighbouring not set optical filter acquires, into
Row vivo identification, for example, it may be determined that pixel value that the pixel unit in the first pixel unit set acquires and neighbouring not setting
The ratio between the pixel value of the pixel unit acquisition of optical filter is set, vivo identification is carried out according to the ratio, for example, in the ratio
When within the scope of certain ratio, it is determined as real human face, is otherwise determined as false face, which can be basis
A large amount of real human face and false face data statistics obtains or machine learning obtains.
But artificial material (such as paper) may be with living person to the wave band model to the response of the spectrum of some wavelength band
The response of the spectrum enclosed, which exists, to partly overlap, if the response of the spectrum based on the wavelength band carries out vivo identification, may lead
Cause misrecognition.In the application one embodiment, every class pixel unit group can be determined for multiple wave bands of target to be identified
The response of the spectrum of range, the response of the spectrum of multiple wavelength band may include the target to be identified to first band range
Spectrum response, the target to be identified is to the response of the spectrum of second band range and the target to be identified to third wave
The response of the spectrum of segment limit, that is to say, that the corresponding spectral response of every class pixel unit group can include face to three kinds
The response of the spectrum of different-waveband range, in this way, even if the spectral response of artificial material and some wavelength band of living person exists
Overlapping, can also be distinguished by the response of the spectrum of other wavelength bands, so as to promote the accuracy of vivo identification.
In the embodiment of the present application, for the test object of a pure color, for example, white paper, for different-waveband model
The spectrum enclosed, theoretically, the spectral response of above-mentioned three classes pixel unit group acquisition should be same or similar, still, practical application
In, the size of the pixel unit spectral response collected in the three classes pixel unit group may have certain difference, this
Sample, when synthesizing three kinds of spectral responses, if some spectral response is excessive, may cause other spectral responses cannot be by effective district
Point, in order to promote the discrimination of living body and prosthese, in the embodiment of the present application, the office that every class pixel unit group can also be acquired
Portion's facial image is calibrated.
Optionally, in the embodiment of the present application, the processor 52 is also used to:
According to calibration parameter, to the first partial facial image of first kind pixel unit group acquisition, second class
Second local facial image of pixel unit group acquisition and the third local facial image of third class pixel unit group acquisition
It is calibrated;
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image determines the true and false of the face.
Optionally, in some embodiments, the processor 52 is also used to:
According to the references object to the first band range, the second band range and the third wavelength band
Spectrum response between relationship, determine the calibration parameter.
Using the references object as test object, the calibration parameter is determined, which can be pure color object, example
Such as white paper, yellowish pink object etc., it is expected that the first kind pixel unit group, the second class pixel unit group and the third
The references object of class pixel unit acquisition is to the first band range, the second band range and the third wave band
The response of the spectrum of range is based on this purpose, to the first kind pixel unit group, the second class pixel list in same level
The references object of tuple and the third class pixel unit actual acquisition is to the first band range, the second band
The response of the spectrum of range and the third wavelength band is calibrated, and determines the calibration parameter.
Specifically, light source emits optical signal to the references object, wherein the wavelength band of the optical signal includes described
First band range, the second band range and the third wavelength band pass through described in the optical sensor
A kind of pixel unit group, the second class pixel unit group and the third class pixel unit acquire the references object to described
The response of the spectrum of first band range, the second band range and the third wavelength band, then according to multiple light
Relationship between the response of spectrum determines the calibration parameter.
It is assumed that the pixel value of first kind pixel unit P1 acquisition is 200, neighbouring with the first kind pixel unit P1 the
The pixel value of two class pixel unit P2 acquisition is 100, the third class pixel unit P3 neighbouring with the first kind pixel unit P1
The pixel value of acquisition is 50, and above three pixel value respectively indicates the response of three kinds of spectrum, proportionate relationship 4:2:1, in order to make
These three pixel units acquisition the references object to the response of three kinds of spectrum in same level, school can be carried out to it
Standard, such as can be by the pixel value of the second class pixel unit P2 acquisition multiplied by 2, by the pixel value of third class pixel unit P3 acquisition
Multiplied by 4, in this way, local in the colour that the subsequent pixel value synthesis based on three pixel unit acquisitions includes three kinds of spectral responses
When facial image, every kind of spectral response can be preferably distinguished.
Optionally, in other embodiments, the calibration parameter is preset value, for example, can be according to three kinds of spectrum
The empirical value of response determines, or can also be determined according to above-mentioned calibration steps, for example, determining calibration ginseng through the above steps
After number, the calibration parameter can be prestored, the calibration of the facial image for subsequent acquisition, i.e., the calibration of subsequent facial image
The calibration parameter is all used, the embodiment of the present application is not construed as limiting this.
It should be understood that in the embodiment of the present application, the response of three kinds of spectrum can refer to three kinds of spectral responses in same level
Difference be less than specific threshold or three kinds of spectral responses pixel value it is suitable, in other words, be in same range.
Optionally, in the embodiment of the present application, the school of each pixel unit in the first pixel unit set can be determined
Then quasi- parameter is calibrated according to pixel value of the calibration parameter of each pixel unit to subsequent acquisition, alternatively, can also incite somebody to action
The calibration parameter of each pixel unit is averaged, and unified calibration parameter is obtained, according to the unified calibration parameter to all
The pixel value of pixel unit acquisition is calibrated, and the embodiment of the present application is not construed as limiting specific calibrating mode.
Optionally, in the embodiment of the present application, the processor 52 is also used to:
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image is synthesized, and colored local facial image is obtained, wherein each pixel in the colour local facial image includes
The face is to the first band range, three responses of the spectrum of the second band range and the third wavelength band
Pixel value;
The colored local facial image is subjected to feature extraction, obtains the feature letter of the colored local facial image
Breath;
According to the characteristic information of the colored local facial image, the true and false of the face is determined.
Specifically, colored local facial image may include three Color Channels, such as RGB, a kind of spectral response pair
Answer a Color Channel, that is to say, that each pixel in colored local facial image includes that three pixel values are (three i.e. described
Respond pixel value), respectively correspond three kinds of spectral responses.And every class pixel in the aforementioned three classes pixel unit group of optical filter is set
Unit group can only obtain a kind of spectral response, therefore, obtain the colored local facial image it needs to be determined that other two kinds of light
Spectrum response, in some embodiments, the pixel value that can be acquired according to other neighbouring class pixel unit groups obtains other two kinds
Spectral response.
For example, the pixel value that can be acquired according to first kind pixel unit in the first kind pixel unit group, determines institute
State the corresponding first response pixel value of the first pixel in colored local facial image, wherein the first response pixel value table
Show response of the face to the spectrum of the first band range;
According to the pixel value that the second class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding second response pixel value of pixel, wherein the second response pixel value indicates the face to the second band model
The spectral response enclosed;And
According to the pixel value that the third class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding third of pixel responds pixel value, and the third response pixel value indicates the face to the light of the third wavelength band
Spectrum response.
Assuming that first kind pixel unit group includes 100 pixel units, the second class pixel unit group includes 100 pictures
Plain unit, the third class pixel unit group include 100 pixel units, and the colour local facial image includes 100 pictures
Element, each pixel include three pixel values, and the response of corresponding three kinds of spectrum then can be according to the one of first kind pixel unit group
The pixel value of a first kind pixel unit P1 acquisition determines that a kind of spectrum of the corresponding pixel of P1 in the image of colour local facial is rung
Answer, according to the second class pixel unit P2 neighbouring with P1 and third class pixel unit P3, determine the corresponding pixel of P1 other two
Kind spectral response, may further be using these three spectral responses as the pixel value of three Color Channels of the pixel.According to class
As method can determine each pixel in the colour local facial image three kinds of spectrum response, thus obtain this coloured silk
Color local facial image.
In other embodiments, it can determine that the first kind pixel unit group, the second class pixel unit group are described
The corresponding full spectra image of every class pixel unit group in third class pixel unit group, further according to every class pixel unit group
Corresponding full spectra image determines the colored local facial image, wherein the full spectra image be include three kinds of light
Compose the image of response.
For example, can be according to second class neighbouring with each first kind pixel unit in the first kind pixel unit group
The pixel value of pixel unit acquisition determines that each first kind pixel unit rings the spectrum of the second band range
It answers, obtains the first response image, be equivalent to the spectral response of the corresponding second band range of first kind pixel unit group;It can root
According to the pixel of the third class pixel unit acquisition neighbouring with each first kind pixel unit in the first kind pixel unit group
Value, determines that each first kind pixel unit for the spectral response of the third wavelength band, obtains the second response image,
It is equivalent to the spectral response of the corresponding third wavelength band of first kind pixel unit group.
In another example can be according to neighbouring with the second class pixel unit of each of the second class pixel unit group first
The pixel value of class pixel unit acquisition determines that each second class pixel unit rings the spectrum of the first band range
It answers, obtains third response image, be equivalent to the spectral response of the corresponding first band range of the second class pixel unit group;It can root
According to the pixel acquired with each of the second class pixel unit group neighbouring third class pixel unit of the second class pixel unit
Value, determines that each second class pixel unit for the spectral response of the third wavelength band, obtains the 4th response image,
It is equivalent to the spectral response of the corresponding third wavelength band of the second class pixel unit group.
For another example can be according to neighbouring with each third class pixel unit in the third class pixel unit group first
The pixel value of class pixel unit acquisition determines that each third class pixel unit rings the spectrum of the first band range
It answers, obtains the 5th response image, be equivalent to the spectral response of the corresponding first band range of third class pixel unit group;It can root
According to the pixel of the second class pixel unit acquisition neighbouring with each third class pixel unit in the third class pixel unit group
Value, determines that each third class pixel unit for the spectral response of the second band range, obtains the 6th response image,
It is equivalent to the spectral response of the corresponding second band range of third class pixel unit group.
Further, the processor 52 is also used to:
By the first partial facial image, synthesized with first response image and second response image,
Obtain the corresponding full spectral response image of the first kind pixel unit group;
By the described second local facial image, synthesized with the third response image and the 4th response image,
Obtain the corresponding full spectral response image of the second class pixel unit group;
By third local facial image, synthesized with the 5th response image and the 6th response image,
Obtain the corresponding full spectral response image of the third class pixel unit group.
It should be understood that first response image to the 6th response image is according to the first partial face figure after calibration
What picture, the described second local facial image and third local facial image obtained.
So far, the corresponding full spectral response image of every class pixel unit group is obtained, it is every in the full spectral response image
Three kinds of spectral responses of a pixel correspondence, i.e., each corresponding three pixel values of pixel, it is believed that be RBG value, may further incite somebody to action
The corresponding full spectral response image of the first kind pixel unit group, the corresponding full spectral response of the second class pixel unit group
Image, the corresponding full spectral response image of the third class pixel unit group are recombinated and (in other words, are spliced), and the coloured silk is obtained
Color local facial image (or RGB figure).
Further, which can carry out vivo identification according to the colour local facial, to identify the true of face
Vacation, for example, the processor 52 can extract the characteristic information of the colour local facial, for example, color character information, specifically may be used
Coloration, saturation degree and purity (Hue, Saturation, Value, HSV) information are thought, then by the spy of the colour local facial
Reference breath is input to deep learning network and classifies, and determines the true and false of face.
Optionally, in the embodiment of the present application, which can be convolutional neural networks or other depth
Learning network.By taking convolutional neural networks as an example, illustrate specific training process.
Firstly, building convolutional neural networks structure, such as two layers of convolutional neural networks can be used, or can also use
Three Tiered Network Architecture or more structure etc..
Secondly, the initial training parameter and the condition of convergence of the convolutional neural networks is arranged.
What the initial training parameter can be randomly generated, or obtain based on experience value, or be also possible to according to big
The parameter of the good convolutional neural networks model of the true and false human face data pre-training of amount.
Non-limiting as example, which may include at least one of the following:
It 1, is that the probability from real human face is greater than the first probability, example by the colored local facial spectral discrimination of real human face
Such as, 98%;
2, the colored local facial image of dummy's face is judged as that the probability from dummy's face is greater than the second probability, such as
95%;
It 3, is that the probability from dummy's face is less than third probability, example by the colored local facial spectral discrimination of real human face
Such as, 2%;
4, by the colored local facial image of dummy's face be judged as the probability from real human face less than the 4th probability, such as
3%.
Then, the colored local facial image of a large amount of real human face and dummy's face is inputted to the convolutional neural networks, it should
Convolutional neural networks can be handled above-mentioned colored local facial image based on initial training parameter, be determined to each colour
The judgement of local facial image is as a result, further, according to the judgement as a result, the structure of adjustment convolutional neural networks and/or each
The training parameter of layer, until determining that result meets the condition of convergence, so far, training is completed.Later, subsequent needs can be identified
The colored local facial image of face is input to the convolutional neural networks, which can be used trained parameter
To the colour local facial, image is handled, and determines whether the colour local facial image comes from real human face.
Optionally, in some embodiments, the processor 52 is also used to:
According to the people of other pixel units acquisition in addition to the first pixel unit set in the pixel array
Face image carries out recognition of face.For example, the processor 52 can be in the facial image and registration that other described pixel units acquire
The target to be identified the matched situation of face template under, vivo identification further is carried out to the target to be identified, is waited at this
Identify that target is determining recognition of face success in the case where real human face, thereby executing the operation for triggering the recognition of face, for example,
Carry out the operation such as terminal unlocking or payment.
Optionally, in other embodiments, which can also be the case where the target to be identified be real human face
Under, further judge the face figure of other pixel units acquisition in addition to the first pixel unit set in pixel array
Seem that the no face template with the target to be identified of registration matches, recognition of face success is determined in the event of a match, into one
Step executes the operation for triggering the recognition of face, for example, carrying out the operation such as terminal unlocking or payment.
Know it should be understood that being readily applicable to other biological feature according to the device for recognition of face of the embodiment of the present application
Other scene, such as fingerprint recognition scene, for example, the finger based on the acquisition of partial pixel unit is extremely when acquiring fingerprint image
Few two kinds of spectral responses, are based further at least two spectral response, determine the true and false of finger.
In the embodiment of the present application, the device 50 for recognition of face may include the processor 52, such as at this
Managing unit can be the micro-control unit (Micro ControlUnit, MCU) in the device of the people's face identification, alternatively, at other
In embodiment, which can not include the processor 52, in this case, performed by the processor 52
Function can be by the processor in electronic equipment that the device 50 for recognition of face is installed, such as master control (Host)
Module executes, and the embodiment of the present application is not construed as limiting this.
Above in association with Fig. 2 to Fig. 5, the Installation practice of the application is described in detail, below in conjunction with Fig. 6 to Fig. 7, retouches in detail
State the present processes embodiment, it should be appreciated that embodiment of the method is corresponded to each other with Installation practice, and similar description is referred to
Installation practice.
Fig. 6 is the schematic flow chart of the method for recognition of face of the embodiment of the present application, as shown in fig. 6, this method
60 include:
S61 receives light launched by light source by the pixel unit in the first pixel unit set of optical sensor and believes
Number from face reflect reflected light signal, and according to the reflected light signal obtain local facial image;Wherein, first picture
Plain unit set includes first kind pixel unit group and the second class pixel unit group, and the first kind pixel unit group includes at least
One first kind pixel unit, the first kind pixel unit are arranged the first optical filter, and first optical filter is used for by the
The optical signal of one wavelength band;The second class pixel unit group includes at least one second class pixel unit, second class
The second optical filter is arranged in pixel unit, and second optical filter is used for the optical signal by second band range, and described second
Wavelength band is different from the first band range;
S62, according to the local facial image for determining the true and false of the face.
It should be understood that this method 60 can be executed by the device for recognition of face, such as the device 50 in previous embodiment,
Specifically, S61 can be executed by the optical sensor 51 in the device 50, and S62 can be by the processor 52 in the device 50, example
As MCU is executed;Alternatively, the electronic equipment that this method 60 can also be installed by the device for recognition of face executes, for example,
S62 can be executed by the processor in electronic equipment, such as Host module, and the embodiment of the present application is not construed as limiting this.
Optionally, in some embodiments of the application, the first pixel unit set further includes third class pixel unit
Group, the third class pixel unit group include at least one third class pixel unit, and third is arranged in the third class pixel unit
Optical filter, the third optical filter are used for the optical signal by third wavelength band, and the third wavelength band is different from described
First band range and the second band range.
Optionally, in some embodiments of the application, the first band range, the second band range and described
Three wavelength bands are respectively one of following three kinds of wavelength bands:
Wavelength band including 560nm, the wavelength band including 980nm, the wavelength band including 940nm.
Optionally, in some embodiments of the application, the wavelength band of the optical signal of light source transmitting includes described the
One wavelength band, the second band range and the third wavelength band.
Optionally, described to be used to determine the face according to the local facial image in some embodiments of the application
It is true and false, comprising:
According to calibration parameter, to the first partial facial image of first kind pixel unit group acquisition, second class
Second local facial image of pixel unit group acquisition and the third local facial image of third class pixel unit group acquisition
It is calibrated;
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image determines the true and false of the face.
Optionally, in some embodiments of the application, the method 60 further include:
According to references object to the first band range, the light of the second band range and the third wavelength band
Relationship between the response of spectrum determines the calibration parameter.
Optionally, in some embodiments of the application, it is described according to references object to the first band range, described
Relationship between the response of the spectrum of two wavelength bands and the third wavelength band, determines the calibration parameter, comprising:
When the light source emits optical signal to the references object, by the first kind pixel unit group, described the
Two class pixel unit groups and the third class pixel unit group acquire the references object to the first band range, institute respectively
State the response of the spectrum of second band range and the third wavelength band;
According to the references object to the first band range, the second band range and the third wavelength band
Spectrum response, determine the calibration parameter, the calibration parameter is used for so that the first kind pixel unit group, described the
Two class pixel unit groups and the references object of third class pixel unit group acquisition are described to the first band range
The response of the spectrum of second band range and the third wavelength band is in same range.
Optionally, in some embodiments of the application, the first partial facial image according to after calibration is described
Second local facial image and third local facial image determine the true and false of the face, comprising:
According to the first partial facial image after calibration, the described second local facial image and the third people from part
Face image is synthesized, and colored local facial image is obtained, wherein each pixel in the colour local facial image includes
The face is to the first band range, three responses of the spectrum of the second band range and the third wavelength band
Pixel value;
The colored local facial image is subjected to feature extraction, obtains the feature letter of the colored local facial image
Breath;
According to the characteristic information of the colored local facial image, the true and false of the face is determined.
Optionally, in some embodiments of the application, the first partial facial image according to after calibration is described
Second local facial image and third local facial image are synthesized, and colored local facial image is obtained, comprising:
Pixel value is used according to what first kind pixel unit in the first kind pixel unit group acquired, determines the colour
The corresponding first response pixel value of the first pixel in the image of local facial, wherein described in the first response pixel value expression
Response of the face to the spectrum of the first band range;
According to the pixel value that the second class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding second response pixel value of pixel, wherein the second response pixel value indicates the face to the second band model
The spectral response enclosed;And
According to the pixel value that the third class pixel unit neighbouring with the first kind pixel unit acquires, described first is determined
The corresponding third of pixel responds pixel value, and the third response pixel value indicates the face to the light of the third wavelength band
Spectrum response.
Optionally, in some embodiments of the application, the characteristic information according to the colored local facial image, really
Determine the true and false of the face, comprising:
It is handled by characteristic information of the deep learning network to the colored local facial image, determines the face
It is true and false.
Optionally, in some embodiments of the application, the method 60 further include:
From in the facial image of multiple real human faces that the optical sensor acquires and false face, described first is extracted
Multiple local facial images of pixel unit set acquisition;
Calibration and synthesis processing are carried out to the multiple local facial image, obtain multiple colored local facial images;
The multiple colored local facial image is input to deep learning network to be trained, obtains the deep learning
The model and parameter of network.
Optionally, in some embodiments of the application, the method 60 further include:
According to the face of other pixel units acquisition in the pixel array in addition to the first pixel unit set
Image carries out recognition of face.
Optionally, in some embodiments of the application, it is described according in the pixel array remove first pixel unit
The facial image of other pixel units acquisition other than set carries out recognition of face, comprising:
If the facial image and the facial image template matching registered and the face is real human face, determine that face is known
Cheng Gong not.
Optionally, in some embodiments of the application, the number of continuous pixel unit in the first pixel unit set
Amount is less than first threshold.
Optionally, in some embodiments of the application, the quantity of the pixel unit in the first pixel unit set with
The ratio of the total quantity of pixel unit in the pixel array is less than the first ratio.
Optionally, the discrete distribution of pixel unit in some embodiments of the application, in the first pixel unit set
In the pixel array.
Optionally, in some embodiments of the application, in the pixel array in addition to the first pixel unit set
Other pixel units acquisition facial image be used for recognition of face.
Optionally, in some embodiments of the application, in the pixel array in addition to the first pixel unit set
Other pixel units be not provided with optical filter.
Optionally, in some embodiments of the application, in the pixel array in addition to the first pixel unit set
Other pixel units setting specific band range optical filter.
Optionally, in some embodiments of the application, the optical filter of the specific band range be include 940nm wave band model
The optical filter enclosed.
Hereinafter, illustrating the overall flow of the method for recognition of face according to the embodiment of the present application, such as Fig. 7 in conjunction with Fig. 7
Shown, this method may include following content:
S71 acquires facial image by optical sensor;
Wherein, the facial image include the pixel unit local facial image collected in the first pixel unit set with
And other pixel units facial image collected.
Further, the pixel unit local facial image collected in S72, in the first pixel unit set;
Then in S73, the local facial image is calibrated.
Specific implementation refers to the related description of previous embodiment, and which is not described herein again.
In s 74, it is synthesized according to the local facial image after calibration, obtains colored local facial image;
In S75, the color character information of the colored local facial image, such as HSV information are extracted;
In s 76, classified according to the color character information of the colored local facial image, determine the true of face
It is false.Specifically, which can be input to deep learning network, to determine the true and false of face.
As shown in figure 8, the electronic equipment 80 may include using the embodiment of the present application also provides a kind of electronic equipment 80
In the device 81 of recognition of face, which can be in aforementioned device embodiment for recognition of face
Device 50, it can be used to execute the content in embodiment of the method described in Fig. 6 to Fig. 7, for sake of simplicity, no longer superfluous here
It states.
Optionally, in some embodiments, the electronic equipment 80 can be smart phone, tablet computer, door lock etc. pair
The higher electronic equipment of security requirement.
It should be understood that the processor or processing unit of the embodiment of the present application can be a kind of IC chip, there is signal
Processing capacity.During realization, each step of above method embodiment can be patrolled by the integrated of the hardware in processor
The instruction for collecting circuit or software form is completed.Above-mentioned processor can be general processor, digital signal processor
(Digital Signal Processor, DSP), specific integrated circuit (Application Specific Integrated
Circuit, ASIC), ready-made programmable gate array (Field Programmable Gate Array, FPGA) or other can
Programmed logic device, discrete gate or transistor logic, discrete hardware components.It may be implemented or execute the application implementation
Disclosed each method, step and logic diagram in example.General processor can be microprocessor or the processor can also be with
It is any conventional processor etc..The step of method in conjunction with disclosed in the embodiment of the present application, can be embodied directly in hardware decoding
Processor executes completion, or in decoding processor hardware and software module combination execute completion.Software module can position
In random access memory, flash memory, read-only memory, programmable read only memory or electrically erasable programmable memory, register
In the storage medium of equal this fields maturation.The storage medium is located at memory, and processor reads the information in memory, in conjunction with it
Hardware completes the step of above method.
It is appreciated that the recognition of face of the embodiment of the present application can also include memory, memory can be volatibility and deposit
Reservoir or nonvolatile memory, or may include both volatile and non-volatile memories.Wherein, nonvolatile memory can
Be read-only memory (Read-Only Memory, ROM), programmable read only memory (Programmable ROM, PROM),
Erasable Programmable Read Only Memory EPROM (Erasable PROM, EPROM), electrically erasable programmable read-only memory
(Electrically EPROM, EEPROM) or flash memory.Volatile memory can be random access memory (Random
Access Memory, RAM), it is used as External Cache.By exemplary but be not restricted explanation, many forms
RAM is available, such as static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic
RAM, DRAM), Synchronous Dynamic Random Access Memory (Synchronous DRAM, SDRAM), Double Data Rate synchronous dynamic
Random access memory (Double Data Rate SDRAM, DDR SDRAM), enhanced Synchronous Dynamic Random Access Memory
(Enhanced SDRAM, ESDRAM), synchronized links dynamic random access memory (Synchlink DRAM, SLDRAM) and straight
Meet rambus random access memory (Direct Rambus RAM, DR RAM).It should be noted that system described herein and side
The memory of method is intended to include but is not limited to the memory of these and any other suitable type.
The embodiment of the present application also proposed a kind of computer readable storage medium, the computer-readable recording medium storage one
A or multiple programs, the one or more program include instruction, and the instruction is when by the portable electronic including multiple application programs
When equipment executes, the portable electronic device can be made to execute the content of embodiment of the method.
The embodiment of the present application also proposed a kind of computer program, which includes instruction, when the computer journey
When sequence is computer-executed, computer is allowed to execute the content of embodiment of the method.
The embodiment of the present application also provides a kind of chip, which includes input/output interface, at least one processor, extremely
A few memory and bus, for storing instruction, at least one processor is for calling this extremely for at least one processor
Instruction in a few memory, to execute the content of embodiment of the method.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
Scope of the present application.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that arriving, disclosed systems, devices and methods can
To realize by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
Division, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or group
Part can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown
Or the mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, device or unit it is indirect
Coupling or communication connection can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, institute
Computer software product is stated to be stored in a storage medium, including some instructions are used so that computer equipment (can be with
It is personal computer, server or the network equipment etc.) execute all or part of step of each embodiment the method for the application
Suddenly.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), deposits at random
The various media that can store program code such as access to memory (Random Access Memory, RAM), magnetic or disk.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
Claims (42)
1. a kind of optical sensor for recognition of face characterized by comprising
Pixel array, the first pixel unit set in the pixel array includes first kind pixel unit group and the second class pixel
Unit group, in which:
The first kind pixel unit group includes at least one first kind pixel unit, the first kind pixel unit setting first
Optical filter, first optical filter are used for the optical signal by first band range;
The second class pixel unit group includes at least one second class pixel unit, the second class pixel unit setting second
Optical filter, second optical filter is used for the optical signal by second band range, and the second band range is different from institute
State first band range;
The first kind pixel unit group and the second class pixel unit group are for receiving optical signal launched by light source from people
The reflected light signal of face reflection, and local facial image is obtained according to the reflected light signal, the local facial image is used for
Determine the true and false of the face.
2. optical sensor according to claim 1, which is characterized in that the first pixel unit set further includes third
Class pixel unit group, the third class pixel unit group include at least one third class pixel unit, the third class pixel list
Member setting third optical filter, the third optical filter are used for the optical signal by third wavelength band, the third wavelength band
Different from the first band range and the second band range, the first kind pixel unit group, the second class pixel
Pixel unit in unit group and the third class pixel unit group is for receiving the optical signal emitted by the light source from face
The reflected light signal of reflection, and local facial image is obtained according to the reflected light signal, the local facial image is for true
Determine the true and false of the face.
3. optical sensor according to claim 2, which is characterized in that the first band range, the second band
Range and the third wavelength band are respectively one of following three kinds of wavelength bands:
Wavelength band including 560nm, the wavelength band including 980nm, the wavelength band including 940nm.
4. optical sensor according to claim 2 or 3, which is characterized in that the wave band of the optical signal of the light source transmitting
Range includes the first band range, the second band range and the third wavelength band.
5. optical sensor according to any one of claim 1 to 4, which is characterized in that the first pixel unit collection
The quantity of continuous pixel unit is less than first threshold in conjunction.
6. optical sensor according to any one of claim 1 to 5, which is characterized in that the first pixel unit collection
The ratio of the quantity of pixel unit in conjunction and the total quantity of the pixel unit in the pixel array is less than the first ratio.
7. optical sensor according to any one of claim 1 to 6, which is characterized in that the first pixel unit collection
Pixel unit in conjunction is discrete to be distributed in the pixel array.
8. optical sensor according to any one of claim 1 to 7, which is characterized in that remove institute in the pixel array
The facial image of other pixel units acquisition except the first pixel unit set is stated for recognition of face.
9. optical sensor according to claim 8, which is characterized in that remove the first pixel list in the pixel array
Other pixel units except member set are not provided with optical filter.
10. optical sensor according to claim 8, which is characterized in that remove first pixel in the pixel array
The optical filter of other pixel units setting specific band range except unit set.
11. optical sensor according to claim 10, which is characterized in that the optical filter of the specific band range is packet
Include the optical filter of 940nm wavelength band.
12. a kind of device for recognition of face characterized by comprising
Optical sensor as described in any one of claims 1 to 11;
Wherein, the first kind pixel unit group and the second class pixel unit in the first pixel unit set of the optical sensor
Pixel unit in group is used to receive the reflected light signal that optical signal launched by light source is reflected from face, and according to the reflection
Optical signal obtains local facial image;
Processor, for determining the true and false of the face according to the local facial image.
13. device according to claim 12, which is characterized in that the first pixel unit set includes first kind pixel
Unit group, the second class pixel unit group and third class pixel unit group, in which:
The first kind pixel unit group includes at least one first kind pixel unit, the first kind pixel unit setting first
Optical filter, first optical filter are used for the optical signal by first band range;
The second class pixel unit group includes at least one second class pixel unit, the second class pixel unit setting second
Optical filter, second optical filter is used for the optical signal by second band range, and the second band range is different from institute
State first band range;
The third class pixel unit group includes at least one third class pixel unit, and third is arranged in the third class pixel unit
Optical filter, the third optical filter are used for the optical signal by third wavelength band, and the third wavelength band is different from described
First band range and the second band range;
Wherein, the first kind pixel unit group, in the second class pixel unit group and the third class pixel unit group
Pixel unit is used to receive the reflected light signal that the optical signal emitted by the light source is reflected from the face, and according to described anti-
It penetrates optical signal and obtains local facial image, the local facial image is for determining the true and false of the face.
14. device according to claim 13, which is characterized in that the processor is also used to:
According to calibration parameter, to the first partial facial image of first kind pixel unit group acquisition, the second class pixel
Second local facial image of unit group acquisition and the third local facial image of third class pixel unit group acquisition carry out
Calibration;
According to the first partial facial image after calibration, the described second local facial image and third local facial figure
As determining the true and false of the face.
15. device according to claim 14, which is characterized in that the processor is also used to:
According to references object to the first band range, the spectrum of the second band range and the third wavelength band
Relationship between response determines the calibration parameter.
16. device according to claim 15, which is characterized in that the optical sensor is also used to:
When the light source emits optical signal to the references object, by the first kind pixel unit group, second class
Pixel unit group and the third class pixel unit group acquire the references object to the first band range respectively, and described
The response of the spectrum of two wavelength bands and the third wavelength band;
The processor is specifically used for: according to the references object to the first band range, the second band range and
The response of the spectrum of the third wavelength band determines that the calibration parameter, the calibration parameter are used for so that the first kind
The references object of pixel unit group, the second class pixel unit group and third class pixel unit group acquisition is to described
The response of the spectrum of first band range, the second band range and the third wavelength band is in same range.
17. device described in any one of 4 to 16 according to claim 1, which is characterized in that the processor is also used to:
According to the first partial facial image after calibration, the described second local facial image and third local facial figure
As being synthesized, colored local facial image is obtained, wherein each pixel in the image of the colour local facial includes described
Face is to the first band range, three response pixels of the spectrum of the second band range and the third wavelength band
Value;
The colored local facial image is subjected to feature extraction, obtains the characteristic information of the colored local facial image;
According to the characteristic information of the colored local facial image, the true and false of the face is determined.
18. device according to claim 17, which is characterized in that the processor is also used to:
According to the pixel value that first kind pixel unit in the first kind pixel unit group acquires, the colored local facial is determined
The corresponding first response pixel value of the first pixel in image, wherein the first response pixel value indicates the face to institute
State the response of the spectrum of first band range;
According to the pixel value that the second class pixel unit neighbouring with the first kind pixel unit acquires, first pixel is determined
Corresponding second response pixel value, wherein the second response pixel value indicates the face to the second band range
Spectral response;And
According to the pixel value that the third class pixel unit neighbouring with the first kind pixel unit acquires, first pixel is determined
Corresponding third responds pixel value, and the third response pixel value indicates that the face rings the spectrum of the third wavelength band
It answers.
19. device described in 7 or 18 according to claim 1, which is characterized in that the processor is specifically used for:
It is handled by characteristic information of the deep learning network to the colored local facial image, determines the true of the face
It is false.
20. device according to claim 19, which is characterized in that the processor is also used to:
From in the facial image of multiple real human faces that the optical sensor acquires and false face, first pixel is extracted
Multiple local facial images of unit set acquisition;
Calibration and synthesis processing are carried out to the multiple local facial image, obtain multiple colored local facial images;
The multiple colored local facial image is input to deep learning network to be trained, obtains the deep learning network
Model and parameter.
21. device described in any one of 2 to 20 according to claim 1, which is characterized in that the processor is also used to:
According to the facial image of other pixel units acquisition in the pixel array in addition to the first pixel unit set
Carry out recognition of face.
22. a kind of method for recognition of face characterized by comprising
Optical signal launched by light source is received from face by the pixel unit in the first pixel unit set of optical sensor
The reflected light signal of reflection, and local facial image is obtained according to the reflected light signal, wherein the first pixel unit collection
Close include first kind pixel unit group and the second class pixel unit group, the first kind pixel unit group include at least one first
The first optical filter is arranged in class pixel unit, the first kind pixel unit, and first optical filter is used to pass through first band model
The optical signal enclosed;The second class pixel unit group includes at least one second class pixel unit, the second class pixel unit
Second optical filter is set, and second optical filter is used for the optical signal by second band range, and the second band range
Different from the first band range;
According to the local facial image for determining the true and false of the face.
23. according to the method for claim 22, which is characterized in that the first pixel unit set further includes third class picture
Plain unit group, the third class pixel unit group include at least one third class pixel unit, and the third class pixel unit is set
Third optical filter is set, the third optical filter is used for the optical signal by third wavelength band, and the third wavelength band is different
In the first band range and the second band range.
24. according to the method for claim 23, which is characterized in that the first band range, the second band range
It is respectively one of following three kinds of wavelength bands with the third wavelength band:
Wavelength band including 560nm, the wavelength band including 980nm, the wavelength band including 940nm.
25. the method according to claim 23 or 24, which is characterized in that the wavelength band of the optical signal of the light source transmitting
Including the first band range, the second band range and the third wavelength band.
26. the method according to any one of claim 23 to 25, which is characterized in that described according to the local facial figure
As for determining the true and false of the face, comprising:
According to calibration parameter, to the first partial facial image of first kind pixel unit group acquisition, the second class pixel
Second local facial image of unit group acquisition and the third local facial image of third class pixel unit group acquisition carry out
Calibration;
According to the first partial facial image after calibration, the described second local facial image and third local facial figure
As determining the true and false of the face.
27. according to the method for claim 26, which is characterized in that the method also includes:
According to references object to the first band range, the spectrum of the second band range and the third wavelength band
Relationship between response determines the calibration parameter.
28. according to the method for claim 27, which is characterized in that it is described according to references object to the first band model
It encloses, the relationship between the response of the spectrum of the second band range and the third wavelength band determines the calibration parameter,
Include:
When the light source emits optical signal to the references object, by the first kind pixel unit group, second class
Pixel unit group and the third class pixel unit group acquire the references object to the first band range respectively, and described
The response of the spectrum of two wavelength bands and the third wavelength band;
According to the references object to the first band range, the light of the second band range and the third wavelength band
The response of spectrum determines that the calibration parameter, the calibration parameter are used for so that the first kind pixel unit group, second class
Pixel unit group and the references object of third class pixel unit group acquisition are to the first band range, and described second
The response of the spectrum of wavelength band and the third wavelength band is in same range.
29. the method according to any one of claim 26 to 28, which is characterized in that described according to after calibration
One local facial image, the described second local facial image and third local facial image determine the true and false of the face,
Include:
According to the first partial facial image after calibration, the described second local facial image and third local facial figure
As being synthesized, colored local facial image is obtained, wherein each pixel in the image of the colour local facial includes described
Face is to the first band range, three response pixels of the spectrum of the second band range and the third wavelength band
Value;
The colored local facial image is subjected to feature extraction, obtains the characteristic information of the colored local facial image;
According to the characteristic information of the colored local facial image, the true and false of the face is determined.
30. according to the method for claim 29, which is characterized in that the first partial face figure according to after calibration
Picture, the described second local facial image and third local facial image are synthesized, obtain colored local facial image, wrap
It includes:
Pixel value is used according to what first kind pixel unit in the first kind pixel unit group acquired, determines the colored part
The corresponding first response pixel value of the first pixel in facial image, wherein the first response pixel value indicates the face
Response to the spectrum of the first band range;
According to the pixel value that the second class pixel unit neighbouring with the first kind pixel unit acquires, first pixel is determined
Corresponding second response pixel value, wherein the second response pixel value indicates the face to the second band range
Spectral response;And
According to the pixel value that the third class pixel unit neighbouring with the first kind pixel unit acquires, first pixel is determined
Corresponding third responds pixel value, and the third response pixel value indicates that the face rings the spectrum of the third wavelength band
It answers.
31. the method according to claim 29 or 30, which is characterized in that described according to the colored local facial image
Characteristic information determines the true and false of the face, comprising:
It is handled by characteristic information of the deep learning network to the colored local facial image, determines the true of the face
It is false.
32. according to the method for claim 31, which is characterized in that the method also includes:
From in the facial image of multiple real human faces that the optical sensor acquires and false face, first pixel is extracted
Multiple local facial images of unit set acquisition;
Calibration and synthesis processing are carried out to the multiple local facial image, obtain multiple colored local facial images;
The multiple colored local facial image is input to deep learning network to be trained, obtains the deep learning network
Model and parameter.
33. the method according to any one of claim 22 to 32, which is characterized in that the method also includes:
According to the facial image of other pixel units acquisition in the pixel array in addition to the first pixel unit set
Carry out recognition of face.
34. the method according to any one of claim 22 to 33, which is characterized in that described according in the pixel array
The facial image of other pixel units acquisition in addition to the first pixel unit set carries out recognition of face, comprising:
If the facial image and the facial image template matching registered and the face is real human face, determine recognition of face at
Function.
35. the method according to any one of claim 22 to 34, which is characterized in that in the first pixel unit set
The quantity of continuous pixel unit is less than first threshold.
36. the method according to any one of claim 22 to 35, which is characterized in that in the first pixel unit set
Pixel unit quantity and the pixel unit in the pixel array total quantity ratio less than the first ratio.
37. the method according to any one of claim 22 to 36, which is characterized in that in the first pixel unit set
Pixel unit discrete be distributed in the pixel array.
38. the method according to any one of claim 22 to 37, which is characterized in that except described the in the pixel array
The facial image of other pixel units acquisition except one pixel unit set is used for recognition of face.
39. according to the method for claim 38, which is characterized in that remove the first pixel unit collection in the pixel array
Other pixel units except conjunction are not provided with optical filter.
40. according to the method for claim 38, which is characterized in that remove the first pixel unit collection in the pixel array
The optical filter of other pixel units setting specific band range except conjunction.
41. according to the method for claim 40, which is characterized in that the optical filter of the specific band range is to include
The optical filter of 940nm wavelength band.
42. a kind of electronic equipment characterized by comprising
The device of recognition of face as described in any one of claim 12 to 21.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2019/088653 WO2020237482A1 (en) | 2019-05-27 | 2019-05-27 | Optical sensor, apparatus and method for facial recognition, and electronic device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110462630A true CN110462630A (en) | 2019-11-15 |
Family
ID=68492804
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201980000834.3A Pending CN110462630A (en) | 2019-05-27 | 2019-05-27 | For the optical sensor of recognition of face, device, method and electronic equipment |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN110462630A (en) |
WO (1) | WO2020237482A1 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112825126A (en) * | 2019-11-20 | 2021-05-21 | 上海箩箕技术有限公司 | Fingerprint identification device and detection method thereof |
WO2022141349A1 (en) * | 2020-12-31 | 2022-07-07 | Oppo广东移动通信有限公司 | Image processing pipeline, image processing method, camera assembly, and electronic device |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113418864B (en) * | 2021-06-03 | 2022-09-16 | 奥比中光科技集团股份有限公司 | Multispectral image sensor and manufacturing method thereof |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102110695A (en) * | 2009-11-06 | 2011-06-29 | 索尼公司 | Solid-state Imaging Device,Manufacturing Method and Designing Method Thereof,and Electronic Device |
US20110270092A1 (en) * | 2010-01-29 | 2011-11-03 | Korea Electrotechnology Research Institute | Combined apparatus for detection of multispectral optical image emitted from living body and for light therapy |
WO2013131407A1 (en) * | 2012-03-08 | 2013-09-12 | 无锡中科奥森科技有限公司 | Double verification face anti-counterfeiting method and device |
CN106254785A (en) * | 2015-06-03 | 2016-12-21 | 豪威科技股份有限公司 | Imageing sensor and the method being used for improving non-visible illumination |
CN107205139A (en) * | 2017-06-28 | 2017-09-26 | 重庆中科云丛科技有限公司 | The imaging sensor and acquisition method of multichannel collecting |
CN107609459A (en) * | 2016-12-15 | 2018-01-19 | 平安科技(深圳)有限公司 | A kind of face identification method and device based on deep learning |
US20190111338A1 (en) * | 2017-10-17 | 2019-04-18 | Sony Interactive Entertainment Inc. | Information processing system and information processing method |
CN208819221U (en) * | 2018-09-10 | 2019-05-03 | 杭州海康威视数字技术股份有限公司 | A kind of face living body detection device |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8798374B2 (en) * | 2008-08-26 | 2014-08-05 | The Regents Of The University Of California | Automated facial action coding system |
US20130004028A1 (en) * | 2011-06-28 | 2013-01-03 | Jones Michael J | Method for Filtering Using Block-Gabor Filters for Determining Descriptors for Images |
CN104252622A (en) * | 2014-10-15 | 2014-12-31 | 倪蔚民 | Mobile terminal front-mounting and iris identification integration photoelectric imaging system and method |
CN107330383A (en) * | 2017-06-18 | 2017-11-07 | 天津大学 | A kind of face identification method based on depth convolutional neural networks |
-
2019
- 2019-05-27 CN CN201980000834.3A patent/CN110462630A/en active Pending
- 2019-05-27 WO PCT/CN2019/088653 patent/WO2020237482A1/en active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102110695A (en) * | 2009-11-06 | 2011-06-29 | 索尼公司 | Solid-state Imaging Device,Manufacturing Method and Designing Method Thereof,and Electronic Device |
US20110270092A1 (en) * | 2010-01-29 | 2011-11-03 | Korea Electrotechnology Research Institute | Combined apparatus for detection of multispectral optical image emitted from living body and for light therapy |
WO2013131407A1 (en) * | 2012-03-08 | 2013-09-12 | 无锡中科奥森科技有限公司 | Double verification face anti-counterfeiting method and device |
CN106254785A (en) * | 2015-06-03 | 2016-12-21 | 豪威科技股份有限公司 | Imageing sensor and the method being used for improving non-visible illumination |
CN107609459A (en) * | 2016-12-15 | 2018-01-19 | 平安科技(深圳)有限公司 | A kind of face identification method and device based on deep learning |
CN107205139A (en) * | 2017-06-28 | 2017-09-26 | 重庆中科云丛科技有限公司 | The imaging sensor and acquisition method of multichannel collecting |
US20190111338A1 (en) * | 2017-10-17 | 2019-04-18 | Sony Interactive Entertainment Inc. | Information processing system and information processing method |
CN208819221U (en) * | 2018-09-10 | 2019-05-03 | 杭州海康威视数字技术股份有限公司 | A kind of face living body detection device |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112825126A (en) * | 2019-11-20 | 2021-05-21 | 上海箩箕技术有限公司 | Fingerprint identification device and detection method thereof |
WO2022141349A1 (en) * | 2020-12-31 | 2022-07-07 | Oppo广东移动通信有限公司 | Image processing pipeline, image processing method, camera assembly, and electronic device |
Also Published As
Publication number | Publication date |
---|---|
WO2020237482A1 (en) | 2020-12-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102430083B1 (en) | Fingerprint identification devices and electronic devices | |
CN209168144U (en) | Fingerprint identification device and terminal device | |
CN110337656A (en) | For the optical sensor of recognition of face, device, method and electronic equipment | |
CN109313706A (en) | Fingerprint identification device, method and terminal device | |
CN110462630A (en) | For the optical sensor of recognition of face, device, method and electronic equipment | |
CN102804229B (en) | Determine equipment, finger print input device, determine method and determine program | |
KR101286454B1 (en) | Fake face identification apparatus and method using characteristic of eye image | |
CN111133446B (en) | Fingerprint identification device and electronic equipment | |
CN106236061A (en) | Human body detection device | |
KR20060105783A (en) | Method and device for iris recognition | |
US10504228B2 (en) | Image processing system and method | |
EP3795071B1 (en) | Heart rate detection method and electronic device | |
KR20170078729A (en) | Systems and methods for spoof detection in iris based biometric systems | |
CN107688775A (en) | Binocular camera recognition of face right discriminating system and method based on elevator scene | |
CN207491128U (en) | A kind of RGB+IR image capture devices | |
CN105874461A (en) | Biometric authentication device and biometric authentication method | |
CN211062054U (en) | Biometric imaging arrangement and electronic device | |
CN211628257U (en) | Fingerprint identification device and electronic equipment | |
CN111801687B (en) | Fingerprint identification device and electronic equipment | |
CN114463790A (en) | Optical fingerprint identification and anti-counterfeiting method and system | |
CN114761951A (en) | Image processing and authentication of unclonable features | |
CN112396745B (en) | Gate control method and electronic equipment | |
RU2672279C1 (en) | Person identification method by iris | |
CN107248079A (en) | One kind is traced to the source anti counterfeit appts, system and method | |
CN116912893A (en) | Ear temperature monitoring method, device, platform and storage medium in epidemic situation environment |
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 | ||
AD01 | Patent right deemed abandoned | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20240402 |