WO2020237483A1 - 用于人脸识别的光学传感器、装置、方法和电子设备 - Google Patents

用于人脸识别的光学传感器、装置、方法和电子设备 Download PDF

Info

Publication number
WO2020237483A1
WO2020237483A1 PCT/CN2019/088654 CN2019088654W WO2020237483A1 WO 2020237483 A1 WO2020237483 A1 WO 2020237483A1 CN 2019088654 W CN2019088654 W CN 2019088654W WO 2020237483 A1 WO2020237483 A1 WO 2020237483A1
Authority
WO
WIPO (PCT)
Prior art keywords
face image
face
pixel
range
waveband
Prior art date
Application number
PCT/CN2019/088654
Other languages
English (en)
French (fr)
Inventor
吴勇辉
潘雷雷
Original Assignee
深圳市汇顶科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市汇顶科技股份有限公司 filed Critical 深圳市汇顶科技股份有限公司
Priority to CN201980000833.9A priority Critical patent/CN110337656A/zh
Priority to PCT/CN2019/088654 priority patent/WO2020237483A1/zh
Publication of WO2020237483A1 publication Critical patent/WO2020237483A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/161Detection; Localisation; Normalisation
    • G06V40/166Detection; Localisation; Normalisation using acquisition arrangements
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/168Feature extraction; Face representation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/16Human faces, e.g. facial parts, sketches or expressions
    • G06V40/172Classification, e.g. identification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/40Spoof detection, e.g. liveness detection
    • G06V40/45Detection of the body part being alive

Definitions

  • This application relates to the technical field of face recognition, and more specifically, to an optical sensor, device, method, and electronic device for face recognition.
  • the embodiments of the present application provide an optical sensor, a device, a method, and an electronic device for face recognition, which can recognize the true and false of a face, thereby improving the security of face recognition.
  • an optical sensor for face recognition including:
  • a pixel array where the first set of pixel units in the pixel array is used to receive the reflected light signal from the face of at least two light signals sequentially emitted by the light source, and obtain at least two face images according to the reflected light signal,
  • the at least two optical signals correspond to different waveband ranges
  • the at least two human face images are used to determine the authenticity of the human face.
  • the at least two optical signals include a first optical signal, a second optical signal, and a third optical signal, where the first optical signal corresponds to a first wavelength range, and the second optical signal
  • the signal corresponds to the second waveband range
  • the third optical signal corresponds to the third waveband range, the first waveband range, the second waveband range and the third waveband range are different in pairs
  • the at least two face images Including a first face image, a second face image, and a third face image, corresponding to the first light signal, the second light signal and the third light signal, the first face image, The second face image and the third face image are used to determine the authenticity of the face.
  • the first waveband range, the second waveband range, and the third waveband range are respectively one of the following three waveband ranges: including a 560nm waveband and a 980nm waveband Range, including the 940nm band range.
  • the pixel units in the first pixel unit set are not provided with filters.
  • other pixel units in the pixel array except the first set of pixel units are not provided with filters, or are provided with filters in a specific wavelength range.
  • the filter in the specific wavelength range includes a filter in the 940 nm wavelength range.
  • the first set of pixel units includes some pixel units in the pixel array.
  • the ratio of the number of pixel units in the first pixel unit set to the total number of pixel units in the pixel array is less than a first ratio.
  • the pixel units in the first pixel unit set are discretely distributed in the pixel array.
  • the number of consecutive pixel units in the first pixel unit set is less than or equal to a certain threshold.
  • the first set of pixel units includes all pixel units in the pixel array.
  • the face images collected by other pixel units in the pixel array except the first pixel unit set are used for face recognition.
  • a device for face recognition including:
  • optical sensor for face recognition in the first aspect to the second aspect and any possible implementation manner thereof;
  • a light source for emitting at least two optical signals in sequence, the at least two optical signals corresponding to different waveband ranges
  • the first set of pixel units of the optical sensor is used to receive the reflected light signal from the face of the at least two light signals sequentially emitted by the light source, and obtain at least two face images according to the reflected light signal ,
  • the at least two human face images are used to determine the authenticity of the human face.
  • the light source is specifically used for:
  • the first optical signal, the second optical signal, and the third optical signal are sequentially emitted to the human face, wherein the first optical signal corresponds to a first waveband range, the second optical signal corresponds to a second waveband range, and the The third optical signal corresponds to a third waveband range, and the first waveband range, the second waveband range, and the third waveband range are two different.
  • the light source is a light source
  • the wavelength range of the optical signal that the light source can emit includes the first wavelength range, the second wavelength range, and the third wavelength range.
  • the light source includes a first sub-light source, a second sub-light source, and a third sub-light source, which are respectively used for emitting the first waveband range, the second waveband range, and the third waveband Range of light signal.
  • the device further includes:
  • the light source control module is configured to control the light source to sequentially emit the first light signal, the second light signal and the third light signal to the face.
  • the optical sensor is specifically used for:
  • the first light signal, the second light signal and the third light signal are respectively reflected from the face of the person through the first pixel unit set, and the first person is respectively obtained according to the reflected light signal. Face image, second face image and third face image.
  • the device further includes:
  • the processor is configured to determine the authenticity of the human face according to the first human face image, the second human face image, and the third human face image.
  • the processor is further configured to:
  • the authenticity of the face is determined.
  • the processor is further configured to:
  • the calibration parameter is determined according to the relationship between the spectral response of the reference object to the first waveband range, the second waveband range and the third waveband range.
  • the optical sensor is also used for:
  • the light source sequentially emits light signals in the first waveband range, the second waveband range, and the third waveband range to the reference object, respectively collecting three images of the reference object;
  • the processor is also used for:
  • the calibration parameter is determined according to the relationship between the pixel values of the three images, and the calibration parameter is used to make the pixel values of the three images within the same range.
  • the processor is further configured to:
  • the second face image and the third face image are synthesized to obtain a color face image, wherein each pixel in the color face image includes Three response pixel values of the human face to the spectrum of the first waveband range, the second waveband range, and the third waveband range;
  • the authenticity of the face is determined.
  • the processor is specifically configured to:
  • the first response pixel value corresponding to the first pixel in the color face image is determined, wherein the The first response pixel value represents the response of the face to the spectrum in the first waveband range, and the first pixel corresponds to the first pixel unit;
  • the third response pixel value corresponding to the first pixel is determined according to the sampling pixel value of the first pixel unit in the first pixel unit set in the third face image, where the third response pixel value represents The response of the face to the spectrum in the third waveband range.
  • the processor is specifically configured to:
  • the feature information of the color face image is processed through a deep learning network to determine the authenticity of the face.
  • the processor is further configured to:
  • Face recognition is performed according to face images collected by other pixel units in the pixel array except the first pixel unit set.
  • the processor is specifically configured to:
  • the processor is further configured to:
  • the face image matches the registered face image template and the face is a real face, it is determined that the face recognition is successful.
  • a method for face recognition includes:
  • the first pixel unit set of the optical sensor sequentially receives the reflected light signals of the at least two light signals from the face, and obtains at least two light signals according to the reflected light signals.
  • Personal face image wherein the at least two optical signals correspond to different waveband ranges;
  • the authenticity of the human face is determined according to the at least two human face images.
  • the light source sequentially emits at least two kinds of light signals to the human face, including:
  • the light source sequentially emits a first optical signal, a second optical signal, and a third optical signal to the human face, wherein the first optical signal corresponds to a first waveband range, and the second optical signal corresponds to a second waveband range , The third optical signal corresponds to a third waveband range.
  • the first waveband range, the second waveband range, and the third waveband range are respectively one of the following three waveband ranges:
  • the light source is a light source
  • the wavelength range of the optical signal that the light source can emit includes the first wavelength range, the second wavelength range, and the third wavelength range.
  • the light source includes a first sub-light source, a second sub-light source, and a third sub-light source, which are respectively used for emitting the first waveband range, the second waveband range, and the third waveband Range of light signal.
  • the method further includes:
  • the first set of pixel units of the optical sensor sequentially receives the reflected light signals of the at least two light signals reflected from the human face, and acquires at least two people according to the reflected light signals.
  • Face images including:
  • the pixel units in the first pixel unit set sequentially receive the first light signal, the second light signal and the third light signal reflected light signals from the human face, and respectively, according to the reflected light signals Acquire the first face image, the second face image and the third face image.
  • the method further includes:
  • the authenticity of the face is determined.
  • the determining the authenticity of the human face according to the first human face image, the second human face image, and the third human face image includes:
  • the authenticity of the face is determined.
  • the method further includes:
  • the calibration parameter is determined according to the relationship between the spectral response of the reference object to the first waveband range, the second waveband range and the third waveband range.
  • the determining the calibration parameter according to the relationship between the spectral response of the reference object to the first waveband range, the second waveband range, and the third waveband range includes :
  • the light source sequentially emits light signals in the first waveband range, the second waveband range, and the third waveband range to the reference object, they are respectively collected by the pixel units in the first pixel unit set Three images of the reference object;
  • the calibration parameter is determined according to the relationship between the pixel values of the three images, and the calibration parameter is used to make the pixel values of the three images within the same range.
  • the determining the authenticity of the human face based on the calibrated first human face image, the second human face image, and the third human face image includes:
  • the second face image and the third face image are synthesized to obtain a color face image, wherein each pixel in the color face image includes Three response pixel values of the human face to the spectrum of the first waveband range, the second waveband range, and the third waveband range;
  • the authenticity of the face is determined.
  • the synthesizing according to the calibrated first face image, the second face image and the third face image to obtain a color face image includes:
  • the first response pixel value corresponding to the first pixel in the color face image is determined, wherein
  • the first response pixel value represents the response of the face to the spectrum in the first waveband range, and the first pixel corresponds to the first pixel unit;
  • the third response pixel value corresponding to the first pixel is determined according to the sampling pixel value of the first pixel unit in the first pixel unit set in the third face image, where the third response pixel value represents The response of the face to the spectrum in the third waveband range.
  • the determining the authenticity of the human face according to the feature information of the color human face image includes:
  • the feature information of the color face image is processed through a deep learning network to determine the authenticity of the face.
  • the method further includes:
  • Face recognition is performed according to face images collected by other pixel units in the pixel array except the first pixel unit set.
  • the performing face recognition based on the face images collected by other pixel units in the pixel array except the first pixel unit set includes:
  • the face image collected by other pixel units in the pixel array except the first pixel unit set is matched with a registered face image template to determine whether the face recognition is successful ,include:
  • the face image matches the registered face image template and the face is a real face, it is determined that the face recognition is successful.
  • an electronic device including the apparatus for face recognition as in the second aspect to the second aspect and any possible implementation manner thereof.
  • a computer-readable medium for storing a computer program, and the computer program includes instructions for executing the foregoing third aspect and any possible implementation manner thereof.
  • a computer program product including instructions is provided.
  • the computer runs the instructions of the computer program product, the computer executes the third aspect and any one of its possible implementations for humans. Face recognition method.
  • the computer program product can run on the electronic device of the fourth aspect.
  • the light source emits light signals in at least two wavebands in sequence, and at least two spectral responses of the target to be identified can be acquired through multiple collections by the optical sensor, and the living body identification can be performed based on the at least two spectral responses , which helps to improve the security of face recognition.
  • Figure 1 is the reflectance spectrum curve of human skin.
  • Fig. 2 is a schematic structural diagram of an optical sensor for face recognition according to an embodiment of the present application.
  • Fig. 3 is a schematic structural diagram of an apparatus for face recognition according to an embodiment of the present application.
  • Fig. 4 is a schematic flowchart of a method for face recognition according to an embodiment of the present application.
  • Fig. 5 is an overall flowchart of a method for face recognition according to an embodiment of the present application.
  • Fig. 6 is a schematic block diagram of an electronic device according to an embodiment of the present application.
  • the embodiments of this application can be applied to various face recognition systems.
  • the face recognition system provided in the embodiments of this application can be applied to mobile terminals such as smart phones and tablet computers, door locks, etc. Access control system or other electronic equipment.
  • in-vivo anti-counterfeiting uses interactive methods, such as blinking or changing facial expressions. This method usually requires continuous collection of several frames of images, which reduces the recognition speed.
  • human skin tissue has a certain specificity in the light reflection performance of a specific wavelength range.
  • human skin is at 560nm
  • the 980nm wavelength range has a special spectral response, which does not exist on the spectral response curve of artificial materials such as paper and molds.
  • the present application provides a method for in vivo anti-counterfeiting of a human face, which can collect the response of the target to be identified to multiple spectra of specific band ranges, and further based on the spectral response to perform in vivo anti-counterfeiting, which is beneficial to improve the recognition speed, and at the same time Improve the security of face recognition.
  • the target to be recognized in the embodiments of the present application may be a human face, or may also be other parts of the human body, such as fingers and palms, which are not limited in the embodiments of the present application.
  • FIG. 2 is a schematic structural diagram of an optical sensor 20 for face recognition provided by an embodiment of the present application.
  • the optical sensor 20 includes:
  • the pixel array 200, the first set of pixel units 21 in the pixel array is used to receive the reflected light signal reflected from the human face by at least two light signals sequentially emitted by the light source, and obtain at least two human faces according to the reflected light signal An image, wherein the at least two light signals correspond to different waveband ranges, and the at least two face images are used to determine the authenticity of the face.
  • the response of the target to be identified to the spectrum of at least two wavebands can be obtained by performing multiple collections based on the optical signals of at least two different wavebands.
  • the living body recognition can be performed based on the response of the at least two kinds of spectra, which is beneficial to improve the security of face recognition.
  • the wavelength range of the light source may be specially designed.
  • the wavelength range of the light source may be designed according to the reflection spectrum curve of human skin, for example, the light source may be designed
  • the waveband range of the waveband is around 560nm, or other visible light ranges, or around 980nm, or it can also be an infrared waveband with better face recognition performance, such as around 940nm, or it can be based on the body’s
  • select the appropriate waveband range as long as it can be clearly distinguished from the prosthesis.
  • the at least two face images may be collected by all pixel units in the pixel array of the optical sensor, or may also be collected by some pixel units in the pixel array, that is, The entire pixel array is used for in vivo identification, or only a part of the pixel units can be used for in vivo identification.
  • the latter implementation method is beneficial to reduce the power consumption of the module and increase the collection speed and recognition speed.
  • the face images collected by other pixel units in the pixel array except the first pixel unit set are used for face recognition.
  • all the pixel units can be used for face recognition.
  • the face image collected by the other pixel units is matched with the registered face image template to determine whether the matching is successful, that is, the face image collected by the pixel unit in the first pixel unit set can be used for living body anti-counterfeiting, the pixel Other pixel units in the array except for the first set of pixel units can be used for face recognition to determine whether the collected face image matches the registered face image template.
  • the sensor driving module can turn off other pixel units in the pixel array except the first set of pixel units, thereby reducing the power consumption of the module and improving Acquisition speed and recognition speed.
  • the number of pixel units in the first pixel unit set and the total number of pixel units in the pixel array are The ratio is smaller than the first ratio, such as 5%, to avoid affecting the face recognition performance.
  • the pixel units in the first pixel unit set are discretely distributed in the pixel array. It can ensure that the living body recognition function does not affect the face recognition performance.
  • the number of consecutive pixel units in the first pixel unit set may be set to be less than or equal to a specific threshold, for example, 6, by setting the number of consecutive pixel units in the first pixel unit set.
  • the number of pixel units is less than a certain threshold, which can avoid affecting the face recognition performance.
  • the pixel units in the first pixel unit set may not be provided with filters, for example, transparent processing or transparent materials are provided, which is beneficial to ensure that the first pixel unit set can collect the light source The transmitted optical signal in the band range.
  • the pixel units in the first pixel unit set may also be provided with filters of a specific wavelength band, for example, used to filter stray light that affects face recognition and living body recognition.
  • other pixel units in the pixel array except for the first set of pixel units are not provided with filters, or are provided with filters of a specific wavelength range, for example, infrared wavelength filters, such as 940 nm Filters for the left and right bands.
  • the at least two optical signals include a first optical signal, a second optical signal, and a second optical signal, wherein the first optical signal corresponds to a first wavelength range, and the first optical signal
  • the two optical signals correspond to the second waveband range
  • the third optical signal corresponds to the third waveband range
  • the first waveband range, the second waveband range and the third waveband range are different in pairs
  • the unit set 21 is specifically used for:
  • the light source sequentially emits the first optical signal, the second optical signal and the third optical signal, it sequentially receives the first optical signal, the second optical signal and the third optical signal
  • the reflected light signal is reflected from the human face, and the first face image, the second face image, and the third face image are respectively acquired according to the reflected light signal, wherein the first face image is
  • the second face image and the third face image are used to determine the authenticity of the face, that is, the three face images can be used for living body recognition.
  • the first waveband range, the second waveband range, and the third waveband range are respectively one of the following three waveband ranges:
  • the above-mentioned three waveband ranges may be determined in the visible light waveband.
  • the first waveband range, the second waveband range, and the third waveband range may be red light wavebands, blue light wavebands, and blue light wavebands, respectively.
  • One of wave band and green light wave band For example, the wavelength range of blue light may be 440nm-475nm in the center band, and the upper cut-off band is about 550nm; the wavelength range of green light may be 520nm-550nm in the center band, the upper cut-off band is about 620nm, and the lower cut-off band is about 460nm; The wavelength range of the red light may be about 550nm in the lower cutoff wavelength range.
  • the photoelectric sensor can collect the face image based on the light signals in the three wavebands, so as to obtain the response of the face to the spectra of the three wavebands, and further can be based on the response of the three spectra Carrying out living body recognition helps to improve the security of face recognition.
  • Fig. 3 is a schematic structural diagram of an apparatus for face recognition according to an embodiment of the present application.
  • the apparatus 30 may include: a light source 31 for emitting at least two optical signals in sequence. The two optical signals correspond to different band ranges;
  • the optical sensor 32 the first set of pixel units of the optical sensor 32 is used to receive the at least two types of light signals sequentially emitted by the light source and reflect the reflected light signals from the human face, and obtain at least two light signals according to the reflected light signals.
  • Personal face images, and the at least two face images are used to determine the authenticity of the face.
  • optical sensor 32 may correspond to the optical sensor 20 in the embodiment shown in FIG. 2, and for specific implementation, reference may be made to the relevant description of the foregoing embodiment, which will not be repeated here.
  • the at least two optical signals may include a first optical signal, a second optical signal, and a third optical signal.
  • the first optical signal corresponds to a first wavelength range
  • the second optical signal corresponds to the second waveband range
  • the third optical signal corresponds to the third waveband range.
  • the implementation of the first waveband range, the second waveband range, and the third waveband range may refer to the related description of the foregoing embodiment, and details are not described herein again.
  • the light source 31 is a light source
  • the wavelength range of the optical signal that the light source 31 can emit includes the first wavelength range, the second wavelength range, and the third wavelength range. range. That is, the above three kinds of optical signals are emitted by the same light source.
  • the frequency of the light source can be controlled to emit optical signals in different wavelength ranges.
  • the light source 31 is a light source combination including a first sub-light source, a second sub-light source, and a third sub-light source, wherein the first sub-light source, the second sub-light source and The third sub-light source is respectively used to emit light signals in the first waveband range, the second waveband range and the third waveband range, that is, the above three optical signals are emitted by different light sources, each The light source can be used to emit a range of light signals.
  • the device 30 further includes:
  • the light source control module is configured to control the light source 31 to sequentially emit the first light signal, the second light signal, and the third light signal to the face.
  • the light source control module may control the light source 31 to sequentially emit light signals in different wavelength ranges, or if the light source includes multiple light sources, the light source control module may control the Multiple light sources emit light in sequence to emit light signals in different wavelength ranges.
  • the light source 31 may be a built-in light source of the device 30, or it may be an external light source of the device 30, or it may be reused with a light source in an electronic device installed in the device 30 to emit In this case, the device 30 may not include the light source 31, which is not limited in the embodiment of the present application.
  • the optical sensor 32 is specifically used for:
  • the third light signal is a reflected light signal reflected from the human face, and a first face image, a second face image, and a third face image are respectively obtained according to the reflected light signal.
  • the apparatus 30 may further include:
  • the processor 33 is configured to determine the authenticity of the human face according to the first human face image, the second human face image, and the third human face image.
  • the first pixel unit set includes only part of the pixel units of the pixel array, the first face image, the second face image, and the third face image may be partial face images, Or if the first pixel unit set includes all pixel units of the pixel array, the first face image, the second face image, and the third face image may be complete face images.
  • the first face image is the response of the face to the spectrum in the first waveband
  • the second face image is the response of the face to the spectrum in the second waveband
  • the third face image is The response of the face to the spectrum in the third waveband range. Since the face has special requirements for the three spectra, based on the responses of the three spectra, the true and false faces can be effectively distinguished.
  • the response of artificial materials (such as paper) to the spectrum of a certain wavelength range may partially overlap with the response of a living person to the spectrum of that wavelength range. If a living body is identified based on the response of the spectrum of this wavelength range, it may cause misidentification. .
  • the response of the target to be identified to the spectrum of multiple band ranges may be collected through the first pixel unit set, and the response of the spectrum of the multiple bands may include the spectrum of the target to be identified to the first wavelength range.
  • the spectral response of can include the human face’s response to the spectrum of three different bands. In this way, even if the artificial material overlaps with the spectral response of a certain band of a living person, it can be measured by the response of the spectrum of other bands. Distinguish, so as to improve the accuracy of living body recognition.
  • the spectral response collected by the pixel units in the first pixel unit set has little difference That is to say, the first pixel unit set has the same or similar response to the light signals in the three waveband ranges.
  • the three types of spectral responses collected by the first pixel unit set may have a certain magnitude In this way, when synthesizing the three spectral responses, if a certain spectral response is too large, other spectral responses may not be effectively distinguished.
  • the processor 33 is further configured to:
  • the authenticity of the face is determined.
  • the reference object is used as a test object to determine the calibration parameters.
  • the reference object may be a solid-color object, such as a white paper, a flesh-colored object, etc., and it is expected that the reference object collected by the first pixel unit set is Wave band range, the spectral response of the second wave band range and the third wave band range are at the same level.
  • the reference object actually collected by the first set of pixel units is relative to the first wave band range Calibrating the spectral response of the second waveband range and the third waveband range to determine the calibration parameter.
  • the first pixel unit set collects all the light signals.
  • the three images of the reference object are marked as the first image, the second image and the third image.
  • the processor 33 determines the calibration parameter according to the relationship between the pixel values of the corresponding pixel units in the three images, and the calibration parameter is used to make the pixel values of the three images within the same range .
  • the pixel value of the first pixel unit P1 in the first image is 200
  • the pixel value collected by the first pixel unit P1 in the second image is 100
  • the pixel value collected by the first pixel unit P1 in the third image The value is 50.
  • the above three pixel values respectively represent the response of the three spectra, and the ratio relationship is 4:2:1.
  • the pixel value collected by the first pixel unit P1 in the second image can be multiplied by 2
  • the pixel value collected by the first pixel unit P1 in the third image can be multiplied by 4
  • the calibration parameter is a preset value, for example, it can be determined according to the empirical value of the three kinds of spectral responses, or it can also be determined according to the above-mentioned calibration step, for example, after the above-mentioned step is determined
  • the calibration parameter can be pre-stored and used for the calibration of the subsequently collected face image, that is, the subsequent calibration of the face image all adopt the calibration parameter, which is not limited in the embodiment of the present application.
  • the response of the three spectra within the same range may mean that the difference of the three spectral responses is smaller than a certain threshold, or the pixel values of the three spectral responses are equivalent, or in other words, at the same level.
  • the calibration parameter of each pixel unit in the first pixel unit set can be determined, and then the subsequently collected pixel values can be calibrated according to the calibration parameter of each pixel unit, or, The calibration parameters of each pixel unit are averaged to obtain a unified calibration parameter, and the pixel values collected by all the pixel units in the first pixel unit set are calibrated according to the unified calibration parameter.
  • the embodiment of the present application deals with specific calibration methods Not limited.
  • the processor 33 is further configured to:
  • the second face image and the third face image are synthesized to obtain a color face image, wherein each pixel in the color face image includes Three response pixel values of the spectrum of the face to the first waveband range, the second waveband range, and the third waveband range;
  • the authenticity of the face is determined.
  • the processor 33 may determine the first pixel corresponding to the first pixel in the color face image according to the sampled pixel value of the first pixel unit in the first pixel unit set in the first face image.
  • Response pixel value wherein the first response pixel value represents the response of the face to the spectrum in the first waveband range, and the first pixel corresponds to the first pixel unit;
  • the third response pixel value corresponding to the first pixel is determined according to the sampling pixel value of the first pixel unit in the first pixel unit set in the third face image, where the third response pixel value represents The response of the face to the spectrum in the third waveband range.
  • the color face image may include three color channels, such as RGB, one spectral response corresponds to one color channel, that is, each pixel in the color face image includes three pixel values (that is, The three response pixel values) correspond to three spectral responses respectively.
  • the three spectral responses of the first pixel unit P1 in the first pixel unit set as an example, the first face image after calibration, the second face image and the third person
  • the three pixel values corresponding to the first pixel unit P1 in the face image are used as the pixel values of the three color channels of the corresponding pixels in the color face image.
  • the three-spectrum responses of each pixel unit in the first pixel unit set can be determined, so as to obtain the color face image.
  • the processor 33 may perform live body recognition based on the color face to identify the true or false of the face. For example, the processor 33 may extract feature information of the color face, for example, color feature information. For chroma, saturation, and purity (Hue, Saturation, Value, HSV) information, then input the feature information of the color partial face into the deep learning network for classification to determine the true and false of the face.
  • feature information of the color face for example, color feature information.
  • chroma, saturation, and purity (Hue, Saturation, Value, HSV) information then input the feature information of the color partial face into the deep learning network for classification to determine the true and false of the face.
  • the deep learning network may be a convolutional neural network or other deep learning networks.
  • the convolutional neural network Take the convolutional neural network as an example to illustrate the specific training process.
  • a convolutional neural network structure For example, a two-layer convolutional neural network can be used, or a three-layer network structure or a multi-layer network structure can also be used.
  • the initial training parameters may be randomly generated, or acquired based on empirical values, or may be parameters of a convolutional neural network model pre-trained based on a large amount of true and false face data.
  • the convergence condition may include at least one of the following:
  • the probability of judging a color face image of a real human face as being from a real human face is greater than the first probability, for example, 98%;
  • the probability that the color face image of the fake face is judged to be from the fake face is greater than the second probability, for example, 95%;
  • the probability of judging a color face image of a real face as a fake face is less than the third probability, for example, 2%;
  • the probability that the color face image of the fake face is judged to be from the real face is less than the fourth probability, for example, 3%.
  • the convolutional neural network can process the feature information of the color face images based on the initial training parameters to determine For the judgment result of each color face image, further, according to the judgment result, adjust the structure of the convolutional neural network and/or the training parameters of each layer until the judgment result meets the convergence condition, and the training is completed.
  • the feature information of the color face image of the face to be recognized can be input into the convolutional neural network, and the convolutional neural network can use the trained parameters to process the feature information of the color face image to determine Whether the color face image is from a real face.
  • the processor 33 is further configured to:
  • Face recognition is performed according to face images collected by other pixel units in the pixel array except the first pixel unit set.
  • the processor 33 may further perform live body recognition on the target to be recognized when the face image collected by the other pixel unit matches the registered face image template of the target to be recognized.
  • live body recognition it is determined that the face recognition is successful, and the operation that triggers the face recognition is performed, for example, operations such as terminal unlocking or payment.
  • the processor 33 may also determine whether the face image collected by other pixel units in the pixel array except for the first pixel unit set is the same as the real face.
  • the registered face image template of the target to be recognized is matched, and in the case of matching, it is determined that the face recognition is successful, and further operations that trigger the face recognition are performed, for example, operations such as terminal unlocking or payment.
  • the device 30 for face recognition can also be applied to other biometric recognition scenarios, such as fingerprint recognition scenarios.
  • biometric recognition scenarios For example, when a fingerprint image is collected, at least two of the fingers collected by a partial pixel unit are used. Based on the at least two spectral responses, the authenticity of the finger is determined.
  • the apparatus 30 for face recognition may include the processor 33.
  • the processor 33 may be a micro control unit (Micro Control Unit, in the apparatus 30 for face recognition). MCU), or, in other embodiments, the device 30 for face recognition may not include the processor 33.
  • the function performed by the processor 33 may be determined by the device for face recognition.
  • the processor in the electronic device installed in the apparatus 30 is executed by, for example, a host module, which is not limited in the embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a method 40 for face recognition according to another embodiment of the present application. As shown in FIG. 4, the method 40 includes:
  • the first pixel unit set of the optical sensor sequentially receives the reflected light signals of the at least two light signals from the human face, and obtains them according to the reflected light signals.
  • S42 Determine the authenticity of the human face according to the at least two human face images.
  • the method 40 may be executed by a device for face recognition, for example, the device 30 in the foregoing embodiment.
  • S41 may be executed by an optical sensor in the device 30, and S42 may be processed by the device 30.
  • the method 40 can also be executed by the electronic device installed in the apparatus for face recognition.
  • S42 can be executed by the processor in the electronic device, such as the Host module. The embodiment of the present application There is no restriction on this.
  • the light source sequentially emits at least two optical signals to the human face, including: the light source sequentially emits a first optical signal, a second optical signal, and a third optical signal to the human face , wherein the first optical signal corresponds to a first waveband range, the second optical signal corresponds to a second waveband range, and the third optical signal corresponds to a third waveband range.
  • the first waveband range, the second waveband range, and the third waveband range are respectively one of the following three waveband ranges: a waveband range including 560nm, including 980nm
  • the wave band range includes the 940nm wave band range.
  • the light source is a light source
  • the wavelength range of the optical signal that the light source can emit includes the first wavelength range, the second wavelength range, and the third wavelength range.
  • the light source includes a first sub-light source, a second sub-light source, and a third sub-light source, which are respectively used for emitting the first waveband range, the second waveband range, and the first sub-light source.
  • the method 40 further includes:
  • the first set of pixel units of the optical sensor sequentially receives the at least two types of light signals reflected from the human face, and obtains at least Two face images, including:
  • the pixel units in the first pixel unit set sequentially receive the first light signal, the second light signal and the third light signal reflected light signals from the human face, and respectively, according to the reflected light signals Acquire the first face image, the second face image and the third face image.
  • the method 40 further includes:
  • the authenticity of the face is determined.
  • the determining the authenticity of the face based on the first face image, the second face image, and the third face image includes:
  • the authenticity of the face is determined.
  • the method 40 further includes:
  • the calibration parameter is determined according to the relationship between the spectral response of the reference object to the first waveband range, the second waveband range and the third waveband range.
  • the calibration parameter is determined based on the relationship between the spectral response of the reference object to the first waveband range, the second waveband range, and the third waveband range ,include:
  • the light source sequentially emits light signals in the first waveband range, the second waveband range, and the third waveband range to the reference object
  • the light signals are respectively collected by the pixel units in the first pixel unit set Three images of the reference object;
  • the calibration parameter is determined according to the relationship between the pixel values of the three images, and the calibration parameter is used to make the pixel values of the three images within the same range.
  • the determining the authenticity of the human face based on the calibrated first human face image, the second human face image, and the third human face image includes :
  • the second face image and the third face image are synthesized to obtain a color face image, wherein each pixel in the color face image includes Three response pixel values of the spectrum of the face to the first waveband range, the second waveband range, and the third waveband range;
  • the authenticity of the face is determined.
  • the synthesizing the first face image, the second face image, and the third face image according to the calibrated first face image to obtain a color face image includes:
  • the first response pixel value corresponding to the first pixel in the color face image is determined, wherein
  • the first response pixel value represents the response of the face to the spectrum in the first waveband range, and the first pixel corresponds to the first pixel unit;
  • the third response pixel value corresponding to the first pixel is determined according to the sampling pixel value of the first pixel unit in the first pixel unit set in the third face image, where the third response pixel value represents The response of the face to the spectrum in the third waveband range.
  • the determining the authenticity of the human face according to the characteristic information of the color facial image includes: processing the characteristic information of the color facial image through a deep learning network To determine the authenticity of the face.
  • the method 40 further includes:
  • Face recognition is performed according to face images collected by other pixel units in the pixel array except for the first pixel unit set.
  • the performing face recognition based on the face image collected by other pixel units in the pixel array except the first pixel unit set includes:
  • the face image matches the registered face image and the face is a real face, it is determined that the face recognition is successful.
  • the method may include the following contents:
  • the second face image and the third face image are synthesized to obtain a color face image
  • classification is performed according to the color feature information of the color face image, and the authenticity of the face is determined.
  • the color face image can be input to a deep learning network to determine the authenticity of the face.
  • an embodiment of the present application also provides an electronic device 60.
  • the electronic device 60 may include a device 61 for face recognition, and the device 61 for face recognition may be the aforementioned device embodiment.
  • the apparatus 30 for face recognition in FIG. 4 can be used to execute the content in the method embodiments described in FIG. 4 to FIG. 5. For brevity, details are not repeated here.
  • the processor or processing unit in the embodiment of the present application may be an integrated circuit chip with signal processing capability.
  • the steps of the foregoing method embodiments can be completed by hardware integrated logic circuits in the processor or instructions in the form of software.
  • the aforementioned processor may be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (ASIC), a ready-made programmable gate array (Field Programmable Gate Array, FPGA) or other Programming logic devices, discrete gates or transistor logic devices, discrete hardware components.
  • DSP Digital Signal Processor
  • ASIC application specific integrated circuit
  • FPGA Field Programmable Gate Array
  • the methods, steps, and logical block diagrams disclosed in the embodiments of the present application can be implemented or executed.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor.
  • the software module can be located in a mature storage medium in the field such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
  • the face recognition in the embodiments of the present application may further include a memory
  • the memory may be a volatile memory or a nonvolatile memory, or may include both volatile and nonvolatile memory.
  • the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), and electrically available Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • the volatile memory may be a random access memory (Random Access Memory, RAM), which is used as an external cache.
  • RAM random access memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM DDR SDRAM
  • ESDRAM enhanced synchronous dynamic random access memory
  • Synchlink DRAM SLDRAM
  • DR RAM Direct Rambus RAM
  • the embodiment of the present application also proposes a computer-readable storage medium that stores one or more programs, and the one or more programs include instructions.
  • the instructions When the instructions are included in a portable electronic device that includes multiple application programs When executed, the portable electronic device can be made to execute the method of the embodiment shown in FIG. 4 to FIG. 5.
  • the embodiment of the present application also proposes a computer program, which includes instructions.
  • the computer program When the computer program is executed by a computer, the computer can execute the method of the embodiments shown in FIGS. 4 to 5.
  • An embodiment of the present application also provides a chip that includes an input and output interface, at least one processor, at least one memory, and a bus.
  • the at least one memory is used to store instructions, and the at least one processor is used to call the at least one memory. To execute the method of the embodiment shown in FIG. 4 to FIG. 5.
  • the disclosed system, device, and method may be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or It can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of the present application essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

Landscapes

  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Collating Specific Patterns (AREA)

Abstract

一种用于人脸识别的光学传感器(20)、装置、方法和电子设备,能够提升人脸识别的安全性,光学传感器(20)包括:像素阵列(200),像素阵列(200)中的第一像素单元集合(21)用于接收由光源依次发射的至少两种光信号从人脸反射的反射光信号,并根据反射光信号获取至少两个人脸图像,其中,至少两种光信号对应不同的波段范围,至少两个人脸图像用于确定人脸的真假。

Description

用于人脸识别的光学传感器、装置、方法和电子设备 技术领域
本申请涉及人脸识别技术领域,并且更具体地,涉及一种用于人脸识别的光学传感器、装置、方法和电子设备。
背景技术
采用人脸识别技术的电子设备给用户带来了安全和便捷的用户体验,但是,通过用户照片(例如,打印的或电子的),或者人工制造的3D人脸模具等伪造的人脸数据是人脸识别应用中的一个安全隐患。因此,如何识别真假人脸,以提升人脸识别的安全性是一项亟需解决的问题。
发明内容
本申请实施例提供了一种用于人脸识别的光学传感器、装置、方法和电子设备,能够识别人脸的真假,从而能够提升人脸识别的安全性。
第一方面,提供了一种用于人脸识别的光学传感器,包括:
像素阵列,所述像素阵列中的第一像素单元集合用于接收由光源依次发射的至少两种光信号从人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,其中,所述至少两种光信号对应不同的波段范围,所述至少两个人脸图像用于确定所述人脸的真假。
在一些可能的实现方式中,所述至少两种光信号包括第一光信号,第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围,所述第一波段范围,所述第二波段范围和所述第三波段范围两两不同;所述至少两个人脸图像包括第一人脸图像,第二人脸图像和第三人脸图像,分别对应所述第一光信号,所述第二光信号和所述第三光信号,所述第一人脸图像,所述第二人脸图像和所述第三人脸图像用于确定所述人脸的真假。
在一些可能的实现方式中,所述第一波段范围,所述第二波段范围和所述第三波段范围分别为以下三种波段范围中的一种:包括560nm的波段范围,包括980nm的波段范围,包括940nm的波段范围。
在一些可能的实现方式中,所述第一像素单元集合中的像素单元不设置 滤光片。
在一些可能的实现方式中,所述像素阵列中除所述第一像素单元集合之外的其他像素单元不设置滤光片,或设置特定波段范围的滤光片。
在一些可能的实现方式中,所述特定波段范围的滤光片为包括940nm波段范围的滤光片。
在一些可能的实现方式中,所述第一像素单元集合包括所述像素阵列中的部分像素单元。
在一些可能的实现方式中,所述第一像素单元集合中的像素单元的数量与所述像素阵列中的像素单元的总数量的比例小于第一比值。
在一些可能的实现方式中,所述第一像素单元集合中的像素单元离散分布在所述像素阵列中。
在一些可能的实现方式中,所述第一像素单元集合中连续的像素单元的数量小于或等于特定阈值。
在一些可能的实现方式中,所述第一像素单元集合包括所述像素阵列中的全部像素单元。
在一些可能的实现方式中,所述像素阵列中除所述第一像素单元集合之外的其他像素单元采集的人脸图像用于人脸识别。
第二方面,提供了一种用于人脸识别的装置,包括:
如第一方面至第二方面及其任一可能的实现方式中的人脸识别的光学传感器;
光源,用于依次发射至少两种光信号,所述至少两种光信号对应不同的波段范围;
其中,所述光学传感器的第一像素单元集合用于接收所述光源依次发射的所述至少两种光信号从人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,所述至少两个人脸图像用于确定所述人脸的真假。
在一些可能的实现方式中,所述光源具体用于:
向所述人脸依次发射第一光信号,第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围,所述第一波段范围,所述第二波段范围和所述第三波段范围两两不同。
在一些可能的实现方式中,所述光源为一个光源,所述光源能够发射的 光信号的波段范围包括所述第一波段范围,所述第二波段范围和第三波段范围。
在一些可能的实现方式中,所述光源包括第一子光源,第二子光源和第三子光源,分别用于发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号。
在一些可能的实现方式中,所述装置还包括:
光源控制模块,用于控制所述光源向所述人脸依次发射所述第一光信号,所述第二光信号和所述第三光信号。
在一些可能的实现方式中,所述光学传感器具体用于:
通过所述第一像素单元集合依次接收所述第一光信号,第二光信号和第三光信号分别从所述人脸反射的反射光信号,并根据所述反射光信号分别获取第一人脸图像,第二人脸图像和第三人脸图像。
在一些可能的实现方式中,所述装置还包括:
处理器,用于根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
在一些可能的实现方式中,所述处理器还用于:
根据校准参数,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
在一些可能的实现方式中,所述处理器还用于:
根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数。在一些可能的实现方式中,所述光学传感器还用于:
在所述光源向所述参考对象依次发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号时,分别采集所述参考对象的三幅图像;
所述处理器还用于:
根据所述三幅图像的像素值的关系,确定所述校准参数,所述校准参数用于使得所述三幅图像的像素值在同一范围内。
在一些可能的实现方式中,所述处理器还用于:
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图 像进行合成,得到彩色人脸图像,其中,所述彩色人脸图像中的每个像素包括所述人脸对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的三个响应像素值;
将所述彩色人脸图像进行特征提取,得到所述彩色人脸图像的特征信息;
根据所述彩色人脸图像的特征信息,确定所述人脸的真假。
在一些可能的实现方式中,所述处理器具体用于:
根据第一像素单元集合中的第一像素单元在所述第一人脸图像中的采样像素值,确定所述彩色人脸图像中的第一像素对应的第一响应像素值,其中,所述第一响应像素值表示所述人脸对所述第一波段范围的光谱的响应,所述第一像素对应所述第一像素单元;
根据第一像素单元集合中的第一像素单元在所述第二人脸图像中的采样像素值,确定所述第一像素对应的第二响应像素值,其中,所述第二响应像素值表示所述人脸对所述第二波段范围的光谱的响应;以及
根据第一像素单元集合中的第一像素单元在所述第三人脸图像中的采样像素值,确定所述第一像素对应的第三响应像素值,其中,所述第三响应像素值表示所述人脸对所述第三波段范围的光谱的响应。
在一些可能的实现方式中,所述处理器具体用于:
通过深度学习网络对所述彩色人脸图像的特征信息进行处理,确定所述人脸的真假。
在一些可能的实现方式中,所述处理器还用于:
根据所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像进行人脸识别。
在一些可能的实现方式中,所述处理器具体用于:
将所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定人脸识别是否成功。
在一些可能的实现方式中,所述处理器还用于:
在所述人脸图像与注册的所述人脸图像模板匹配且所述人脸为真实人脸的情况下,确定人脸识别成功。
第三方面,提供了一种用于人脸识别的方法,所述方法包括:
在光源依次发射至少两种光信号时,通过光学传感器的第一像素单元集 合依次接收所述至少两种光信号从所述人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,其中,所述至少两种光信号对应不同的波段范围;
根据所述至少两个人脸图像确定所述人脸的真假。
在一些可能的实现方式中,所述光源依次向人脸发射至少两种光信号,包括:
所述光源依次向所述人脸发射第一光信号,第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围。
在一些可能的实现方式中,所述第一波段范围,所述第二波段范围和所述第三波段范围分别为以下三种波段范围中的一种:
包括560nm的波段范围,包括980nm的波段范围,包括940nm的波段范围。
在一些可能的实现方式中,所述光源为一个光源,所述光源能够发射的光信号的波段范围包括所述第一波段范围,所述第二波段范围和第三波段范围。
在一些可能的实现方式中,所述光源包括第一子光源,第二子光源和第三子光源,分别用于发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号。
在一些可能的实现方式中,所述方法还包括:
控制所述光源向所述人脸依次发射所述第一光信号,所述第二光信号和所述第三光信号。
在一些可能的实现方式中,所述通过光学传感器的第一像素单元集合依次接收所述至少两种光信号从所述人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,包括:
通过所述第一像素单元集合中的像素单元依次接收所述第一光信号,第二光信号和第三光信号分别从所述人脸反射的反射光信号,并根据所述反射光信号分别获取第一人脸图像,第二人脸图像和第三人脸图像。
在一些可能的实现方式中,所述方法还包括:
根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
在一些可能的实现方式中,所述根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假,包括:
根据校准参数,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像,确定所述人脸的真假。
在一些可能的实现方式中,所述方法还包括:
根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数。
在一些可能的实现方式中,所述根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数,包括:
在所述光源向所述参考对象依次发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号时,通过所述第一像素单元集合中的像素单元分别采集所述参考对象的三幅图像;
根据所述三幅图像的像素值的关系,确定所述校准参数,所述校准参数用于使得所述三幅图像的像素值在同一范围内。
在一些可能的实现方式中,所述根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像,确定所述人脸的真假,包括:
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,其中,所述彩色人脸图像中的每个像素包括所述人脸对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的三个响应像素值;
将所述彩色人脸图像进行特征提取,得到所述彩色人脸图像的特征信息;
根据所述彩色人脸图像的特征信息,确定所述人脸的真假。
在一些可能的实现方式中,所述根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,包括:
根据第一像素单元集合中的第一像素单元在所述第一人脸图像中的采样像素值,确定所述彩色人脸图像中的第一像素对应的第一响应像素值,其中,所述第一响应像素值表示所述人脸对所述第一波段范围的光谱的响应, 所述第一像素对应所述第一像素单元;
根据第一像素单元集合中的第一像素单元在所述第二人脸图像中的采样像素值,确定所述第一像素对应的第二响应像素值,其中,所述第二响应像素值表示所述人脸对所述第二波段范围的光谱的响应;以及
根据第一像素单元集合中的第一像素单元在所述第三人脸图像中的采样像素值,确定所述第一像素对应的第三响应像素值,其中,所述第三响应像素值表示所述人脸对所述第三波段范围的光谱的响应。
在一些可能的实现方式中,所述根据所述彩色人脸图像的特征信息,确定所述人脸的真假,包括:
通过深度学习网络对所述彩色人脸图像的特征信息进行处理,确定所述人脸的真假。
在一些可能的实现方式中,所述方法还包括:
根据所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像进行人脸识别。
在一些可能的实现方式中,所述根据所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像进行人脸识别,包括:
将所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定人脸识别是否成功。
在一些可能的实现方式中,所述将所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定人脸识别是否成功,包括:
在所述人脸图像与注册的人脸图像模板匹配且所述人脸为真实人脸的情况下,确定人脸识别成功。
第四方面,提供了一种电子设备,包括如第二方面至第二方面及其任一可能的实现方式中的用于人脸识别的装置。
第五方面,提供了一种计算机可读介质,用于存储计算机程序,所述计算机程序包括用于执行上述第三方面及其任一可能的实现方式中的指令。
第六方面,提供了一种包括指令的计算机程序产品,当计算机运行所述计算机程序产品的所述指令时,所述计算机执行上述第三方面及其任一可能的实现方式中的用于人脸识别的方法。
具体地,该计算机程序产品可以运行于上述第四方面的电子设备上。
基于上述技术方案,通过光源依次发射至少两种波段范围的光信号,通过光学传感器进行多次采集即可获取待识别目标的至少两种光谱响应,进一步可以基于该至少两种光谱响应进行活体识别,有利于提升人脸识别的安全性。
附图说明
图1是人体皮肤的反射光谱曲线。
图2是根据本申请实施例的用于人脸识别的光学传感器的示意性结构图。
图3是根据本申请实施例的用于人脸识别的装置的示意性结构图。
图4是根据本申请实施例的用于人脸识别的方法的示意性流程图。
图5是根据本申请实施例的用于人脸识别的方法的整体流程图。
图6是根据本申请实施例的电子设备的示意性框图。
具体实施方式
下面将结合附图,对本申请实施例中的技术方案进行描述。
应理解,本申请实施例可以应用于各种人脸识别系统,作为一种常见的应用场景,本申请实施例提供的人脸识别系统可以应用在智能手机、平板电脑等移动终端和门锁、门禁系统或者其他电子设备。
在传统的人脸识别系统中,活体防伪采用交互的方式,例如,采用眨眼,或表情变化等方式,这种方式通常需要连续采集几帧图像,降低了识别速度。
通常来说,受人体皮肤组织的皮层厚度、血红蛋白浓度、黑色素含量等因素的影响,人体皮肤组织对特定波段范围的光线反射性能具有一定的特殊性,如图1所示,人体的皮肤在560nm左右的波段范围,980nm的波段范围具有特殊的光谱响应,这种特殊的光谱响应在纸张、模具等人工材料的光谱响应曲线上是不存在的。
据此,本申请提供了一种人脸活体防伪的方法,能够采集待识别目标对多个特定波段范围的光谱的响应,进一步基于该光谱响应进行活体防伪,有利于提升识别速度,同时还可以提升人脸识别的安全性。
应理解,本申请实施例中的待识别目标可以为人脸,或者也可以为人体的其他部位,例如手指,手掌,本申请实施例对此不作限定。
图2是本申请实施例提供的一种用于人脸识别的光学传感器20的示意性结构图,该光学传感器20包括:
像素阵列200,所述像素阵列中的第一像素单元集合21用于接收由光源依次发射的至少两种光信号从人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,其中,所述至少两种光信号对应不同的波段范围,所述至少两个人脸图像用于确定所述人脸的真假。
在本申请实施例中,不需要对像素阵列作特别的设计,只需基于至少两种不同波段范围的光信号,进行多次采集即可获取待识别目标对至少两种波段范围的光谱的响应,进一步可以基于所述至少两种光谱的响应进行活体识别,有利于提升人脸识别的安全性。
在本申请实施例中,所述光源的波段范围可以是特别设计的,作为一个可选的实现方式,可以根据人体皮肤的反射光谱曲线,设计所述光源的波段范围,例如,设计所述光源的波段范围为560nm左右的波段范围,或者其他可见光范围,或980nm左右的波段范围,或者也可以为人脸识别性能较优的红外波段范围,例如940nm左右的波段范围,或者,也可以根据活体的其他生物特征,选择合适的波段范围,只要能够与假体具有明显的区分度即可。
应理解,在该实施例中,所述至少两个人脸图像可以是光学传感器的像素阵列中的全部像素单元采集的,或者也可以是像素阵列中的部分像素单元采集的,也就是说,可以利用整个像素阵列做活体识别,或者也可以只利用部分像素单元做活体识别,后一种实现方式有利于降低模组的功耗,提高采集速度和识别速度。
当采用所述像素阵列的部分像素单元做活体识别时,所述像素阵列中除所述第一像素单元集合之外的其他像素单元采集的人脸图像用于人脸识别,例如,可以将所述其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定是否匹配成功,即所述第一像素单元集合中的像素单元采集的人脸图像可以用于活体防伪,所述像素阵列中的除所述第一像素单元集合之外的其他像素单元可以用于人脸识别,以确定采集的人脸图像和注册的人脸图像模板是否匹配。
当采用所述像素阵列的部分像素单元做活体识别时,传感器驱动模块可以关闭所述像素阵列中除所述第一像素单元集合之外的其他像素单元,从而能够降低模组的功耗,提升采集速度和识别速度。
可选地,当所述第一像素单元集合包括所述像素阵列中的部分像素单元时,所述第一像素单元集合中的像素单元的数量与所述像素阵列中的像素单元的总数量的比例小于第一比值,例如5%,以避免影响人脸识别性能。
可选地,所述第一像素单元集合中的像素单元离散分布在所述像素阵列中。能够保证活体识别功能不影响人脸识别性能。
可选地,在本申请实施例中,可以设置所述第一像素单元集合中连续的像素单元的数量小于或等于特定阈值,例如,6个,通过设置所述第一像素单元集合中连续的像素单元的个数小于一定阈值,能够避免影响人脸识别性能。
作为一个实施例,所述第一像素单元集合中的像素单元可以不设置滤光片,例如,做透明处理,或者设置透明材料,有利于保证所述第一像素单元集合可以采集到所述光源发射的波段范围的光信号。在其他实施例中,所述第一像素单元集合中的像素单元也可以设置特定波段的滤光片,例如,用于滤除影响人脸识别和活体识别的杂散光。
可选地,所述像素阵列中除所述第一像素单元集合之外的其他像素单元不设置滤光片,或设置特定波段范围的滤光片,例如,红外波段的滤光片,比如940nm左右波段范围的滤光片。
可选地,在一些实施例中,所述至少两种光信号包括第一光信号,第二光信号和第二光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围,所述第一波段范围,所述第二波段范围和所述第三波段范围两两不同,所述第一像素单元集合21具体用于:
在所述光源依次发射所述第一光信号,所述第二光信号和所述第三光信号时,依次接收所述第一光信号,所述第二光信号和所述第三光信号从所述人脸反射的反射光信号,并根据所述反射光信号分别对应获取第一人脸图像,第二人脸图像和第三人脸图像,其中,所述第一人脸图像,所述第二人脸图像和第三人脸图像用于确定人脸的真假,即这三个人脸图像可以用作活体识别。
可选地,在本申请一个实施例中,所述第一波段范围,所述第二波段范围和所述第三波段范围分别为以下三种波段范围中的一种:
包括560nm的波段范围,包括980nm的波段范围,包括940nm的波段 范围。
可选地,在其他实施例中,可以在可见光波段中确定上述三个波段范围,所述第一波段范围,所述第二波段范围和所述第三波段范围可以分别为红光波段,蓝光波段和绿光波段的一种。例如,蓝光的波段范围可以是中心波段为440nm~475nm,上截止波段约为550nm;绿光的波段范围可以是中心波段为520nm~550nm,上截止波段约为620nm,下截止波段约为460nm;红光的波段范围可以是下截止波段约为550nm。
因此,在本申请实施例中,光电传感器可以基于三种波段范围的光信号进行人脸图像的采集,从而获得人脸对三种波段范围的光谱的响应,进一步可以基于该三种光谱的响应进行活体识别,有利于提升人脸识别的安全性。
图3是根据本申请实施例的用于人脸识别的装置的示意性结构图,如图3所示,该装置30可以包括:光源31,用于依次发射至少两种光信号,所述至少两种光信号对应不同的波段范围;
光学传感器32,所述光学传感器32的第一像素单元集合用于接收所述光源依次发射的所述至少两种光信号从人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,所述至少两个人脸图像用于确定所述人脸的真假。
应理解,所述光学传感器32可以对应于图2所示实施例中的光学传感器20,具体实现可以参考前述实施例的相关描述,这里不再赘述。
可选地,在一些实施例中,所述至少两种光信号可以包括第一光信号,第二光信号和第三光信号,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围。其中,所述第一波段范围,所述第二波段范围和所述第三波段范围的实现可以参考前述实施例的相关描述,这里不再赘述。
可选地,在一些实施例中,所述光源31为一个光源,所述光源31能够发射的光信号的波段范围包括所述第一波段范围,所述第二波段范围和所述第三波段范围。即上述三种光信号是由同一光源发射的,例如,可以通过控制光源的频率使其发射不同波段范围的光信号。
可选地,在其他实施例中,所述光源31为包括第一子光源,第二子光源和第三子光源的光源组合,其中,所述第一子光源,所述第二子光源和所述第三子光源分别用于发射所述第一波段范围,所述第二波段范围和所述第 三波段范围的光信号,即上述三种光信号是由不同的光源发射的,每个光源可以用于发射一个波段范围的光信号。
可选地,在本申请一个实施例中,所述装置30还包括:
光源控制模块,用于控制所述光源31向所述人脸依次发射所述第一光信号,所述第二光信号和所述第三光信号。
例如,若所述光源31为一个光源,所述光源控制模块可以控制该光源31依次发射不同波段范围的光信号,或者,若所述光源包括多个光源,所述光源控制模块可以控制所述多个光源依次发光,以发射不同波段范围的光信号。
可选地,光源31可以是所述装置30的内置光源,或者也可以是所述装置30的外置光源,或者也可以复用所述装置30所安装的电子设备中的光源来发射用于活体识别的光信号,此情况下,所述装置30可以不包括所述光源31,本申请实施例对此不作限定。
可选地,在一个实施例中,所述光学传感器32具体用于:
在所述光源31向所述人脸依次发射第一光信号,第二光信号和第三光信号时,通过所述第一像素单元集合依次接收所述第一光信号,第二光信号和第三光信号分别从所述人脸反射的反射光信号,并根据所述反射光信号分别获取第一人脸图像,第二人脸图像和第三人脸图像。
可选地,在一些实施例中,所述装置30还可以包括:
处理器33,用于根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
应理解,若所述第一像素单元集合只包括像素阵列的部分像素单元,所述第一人脸图像,所述第二人脸图像和所述第三人脸图像可以为局部人脸图像,或者若所述第一像素单元集合包括像素阵列的全部像素单元,所述第一人脸图像,所述第二人脸图像和所述第三人脸图像可以为完整的人脸图像。
其中,所述第一人脸图像为人脸对第一波段范围的光谱的响应,所述第二人脸图像为人脸对所述第二波段范围的光谱的响应,所述第三人脸图像为所述人脸对第三波段范围的光谱的响应。由于该人脸对该三种光谱具有特殊的需要,因此,基于三种光谱的响应,可以有效区分人脸的真假。
作为一个可选的实现方式,可以根据上述三种光谱的响应中的一个光谱响应,确定人脸的真假,这样,在人脸图像采集时也可以只使用一个波段范 围的光信号进行人脸采集,进一步基于该一个光谱响应做活体识别。
但是,人工材质(例如纸张)对某个波段范围的光谱的响应可能与活人对该波段范围的光谱的响应存在部分重叠,如果基于该波段范围的光谱的响应进行活体识别,可能导致误识别。在本申请实施例中,可以通过第一像素单元集合采集待识别目标对于多个波段范围的光谱的响应,该多个波段范围的光谱的响应可以包括该待识别目标对第一波段范围的光谱的响应,该待识别目标对第二波段范围的光谱的响应,以及该待识别目标对第三波段范围的光谱的响应,也就是说,所述第一像素单元集合中的每个像素单元对应的光谱响应都可以包括人脸对三种不同波段范围的光谱的响应,这样,即使人工材质与活人的某个波段范围的光谱的响应存在重叠,也可以通过其他波段范围的光谱的响应进行区分,从而能够提升活体识别的准确度。
在本申请实施例中,对于一个纯色的测试对象,比如,白色纸张,对于不同的波段范围的光谱而言,我们期望该第一像素单元集合中的像素单元采集到的光谱响应的差别不大,也就是说,这第一像素单元集合对于三种波段范围的光信号的响应相同或相近,但是,实际应用中,所述第一像素单元集合所采集的三种光谱响应的大小可能具有一定的差别,这样,合成三种光谱响应时,如果某个光谱响应过大,可能导致其他光谱响应不能被有效区分,为了提升活体和假体的区分度,在本申请实施例中,还可以对第一像素单元集合采集的人脸图像进行校准。
可选地,在一些实施例中,所述处理器33还用于:
根据校准参数,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
以所述参考对象为测试对象,确定所述校准参数,该参考对象可以为纯色物体,例如白色纸张,肉色物体等,期望所述第一像素单元集合采集的所述参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应在同一水平,基于此目的,对所述第一像素单元集合实际采集的所述参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应进行校准,确定所述校准参数。
具体地,在所述光源向所述参考对象依次发射所述第一波段范围,所述 第二波段范围和所述第三波段范围的光信号时,通过所述第一像素单元集合分别采集所述参考对象的三幅图像,记为第一图像,第二图像和第三图像。进一步地,所述处理器33根据所述三幅图像中的对应像素单元的像素值的关系,确定所述校准参数,所述校准参数用于使得所述三幅图像的像素值在同一范围内。
假设,在第一图像中第一像素单元P1的像素值为200,在第二图像中所述第一像素单元P1采集的像素值为100,在第三图像中第一像素单元P1采集的像素值为50,上述三个像素值分别表示三种光谱的响应,比例关系为4:2:1,为了使得该第一像素单元P1采集的所述参考对象对三种光谱的响应在同一水平,可以对其进行校准,例如可以将第二图像中该第一像素单元P1采集的像素值乘以2,将第三图像中该第一像素单元P1采集的像素值乘以4,这样,在后续基于该像素单元采集的三个像素值合成包括三种光谱响应的彩色人脸图像时,可以更好的区分每种光谱响应。
可选地,在其他实施例中,所述校准参数为预设值,例如,可以根据所述三种光谱响应的经验值确定,或者也可以根据上述校准步骤确定,例如,在通过上述步骤确定校准参数之后,可以预存该校准参数,用于后续采集的人脸图像的校准,即后续的人脸图像的校准都采用该校准参数,本申请实施例对此不作限定。
应理解,在本申请实施例中,三种光谱的响应在同一范围内可以指三种光谱响应的差值小于特定阈值,或者三种光谱响应的像素值相当,或者说,处在同一水平。
可选地,在本申请实施例中,可以确定第一像素单元集合中的每个像素单元的校准参数,然后根据每个像素单元的校准参数对后续采集的像素值进行校准,或者,也可以将每个像素单元的校准参数进行平均,得到统一的校准参数,根据该统一的校准参数对第一像素单元集合中的所有像素单元采集的像素值进行校准,本申请实施例对于具体的校准方式不作限定。
可选地,在一些实施例中,所述处理器33还用于:
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,其中,所述彩色人脸图像中的每个像素包括所述人脸对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的三个响应像素值;
将所述彩色人脸图像进行特征提取,得到所述彩色人脸图像的特征信息;
根据所述彩色人脸图像的特征信息,确定所述人脸的真假。
例如,所述处理器33可以根据第一像素单元集合中的第一像素单元在所述第一人脸图像中的采样像素值,确定所述彩色人脸图像中的第一像素对应的第一响应像素值,其中,所述第一响应像素值表示所述人脸对所述第一波段范围的光谱的响应,所述第一像素对应所述第一像素单元;
根据第一像素单元集合中的第一像素单元在所述第二人脸图像中的采样像素值,确定所述第一像素对应的第二响应像素值,其中,所述第二响应像素值表示所述人脸对所述第二波段范围的光谱的响应;以及
根据第一像素单元集合中的第一像素单元在所述第三人脸图像中的采样像素值,确定所述第一像素对应的第三响应像素值,其中,所述第三响应像素值表示所述人脸对所述第三波段范围的光谱的响应。
具体而言,所述彩色人脸图像可以包括三个颜色通道,例如RGB,一种光谱响应对应一个颜色通道,也就是说,彩色人脸图像中的每个像素包括三个像素值(即所述三个响应像素值),分别对应三种光谱响应。以确定第一像素单元集合中的第一像素单元P1的三种光谱响应为例进行说明,可以将校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像中所述第一像素单元P1对应的三个像素值,作为所述彩色人脸图像中对应像素的三个颜色通道的像素值。按照类似的方法可以确定该第一像素单元集合中的每个像素单元的三种光谱的响应,从而得到这个彩色人脸图像。
进一步地,该处理器33可以根据该彩色人脸进行活体识别,以识别人脸的真假,例如,所述处理器33可以提取该彩色人脸的特征信息,例如,色彩特征信息,具体可以为色度,饱和度和纯度(Hue,Saturation,Value,HSV)信息,然后将该彩色局部人脸的特征信息输入到深度学习网络进行分类,确定人脸的真假。
可选地,在本申请实施例中,该深度学习网络可以为卷积神经网络,或者其他深度学习网络。以卷积神经网络为例,说明具体的训练过程。
首先,构建卷积神经网络结构,例如可以采用二层卷积神经网络,或者也可以采用三层网络结构或更多层网络结构等。
其次,设置该卷积神经网络的初始训练参数和收敛条件。
该初始训练参数可以是随机生成的,或根据经验值获取的,或者也可以是根据大量的真假人脸数据预训练好的卷积神经网络模型的参数。
作为示例而非限定,该收敛条件可以包括以下中的至少一项:
1、将真实人脸的彩色人脸图像判定为来自真实人脸的概率大于第一概率,例如,98%;
2、将假人脸的彩色人脸图像判断为来自假人脸的概率大于第二概率,例如95%;
3、将真实人脸的彩色人脸图像判定为来自假人脸的概率小于第三概率,例如,2%;
4、将假人脸的彩色人脸图像判断为来自真实人脸的概率小于第四概率,例如3%。
然后,向该卷积神经网络输入大量的真实人脸和假人脸的彩色人脸图像的特征信息,该卷积神经网络可以基于初始训练参数对上述彩色人脸图像的特征信息进行处理,确定对每个彩色人脸图像的判定结果,进一步地,根据该判定结果,调整卷积神经网络的结构和/或各层的训练参数,直至判定结果满足收敛条件,至此,训练完成。之后,可以将后续需要识别的人脸的彩色人脸图像的特征信息输入到该卷积神经网络,该卷积神经网络可以使用训练好的参数对该彩色人脸图像的特征信息进行处理,确定该彩色人脸图像是否来自真实人脸。
可选地,在一些实施例中,所述处理器33还用于:
根据所述像素阵列中的除所述第一像素单元集合之外的其他像素单元采集的人脸图像进行人脸识别。
例如,该处理器33可以在所述其他像素单元采集的人脸图像与注册的该待识别目标的人脸图像模板匹配的情况下,进一步对该待识别目标进行活体识别,在该待识别目标为真实人脸的情况下确定人脸识别成功,从而执行触发该人脸识别的操作,例如,进行终端解锁或支付等操作。
在其他实施例中,该处理器33也可以在待识别目标为真实人脸的情况下,进一步判断像素阵列中的除所述第一像素单元集合以外的其他像素单元采集的人脸图像是否与注册的该待识别目标的人脸图像模板匹配,在匹配的情况下确定人脸识别成功,进一步执行触发该人脸识别的操作,例如,进行终端解锁或支付等操作。
应理解,根据本申请实施例的用于人脸识别的装置30也可以适用于其他生物特征识别场景,例如指纹识别场景,例如,在采集指纹图像时,基于部分像素单元采集的手指的至少两种光谱响应,进一步基于该至少两种光谱响应,确定手指的真假。
在本申请实施例中,所述用于人脸识别的装置30可以包括该处理器33,例如该处理器33可以为该用于人脸识别的装置30中的微控制单元(Micro Control Unit,MCU),或者,在其他实施例中,该用于人脸识别的装置30可以不包括该处理器33,此情况下,所述处理器33所执行的功能可以由所述用于人脸识别的装置30所安装的电子设备中的处理器,例如主控(Host)模块执行,本申请实施例对此不作限定。
上文结合图2至图3,详细描述了本申请的装置实施例,下文结合图4,详细描述本申请的方法实施例,应理解,方法实施例与装置实施例相互对应,类似的描述可以参照装置实施例。
图4是本申请另一实施例的用于人脸识别的方法40的示意性流程图,如图4所示,该方法40包括:
S41,在光源依次发射至少两种光信号时,通过光学传感器的第一像素单元集合依次接收所述至少两种光信号从所述人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,其中,所述至少两种光信号对应不同的波段范围;
S42,根据所述至少两个人脸图像确定所述人脸的真假。
应理解,该方法40可以由用于人脸识别的装置执行,例如前述实施例中的装置30,具体地,S41可以由该装置30中的光学传感器执行,S42可以由该装置30中的处理器,例如MCU执行;或者,该方法40也可以由该用于人脸识别的装置所安装的电子设备执行,例如,S42可以由电子设备中的处理器,例如Host模块执行,本申请实施例对此不作限定。
可选地,在一些实施例中,所述光源依次向人脸发射至少两种光信号,包括:所述光源依次向所述人脸发射第一光信号,第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围。
可选地,在一些实施例中,所述第一波段范围,所述第二波段范围和所述第三波段范围分别为以下三种波段范围中的一种:包括560nm的波段范 围,包括980nm的波段范围,包括940nm的波段范围。
可选地,在一些实施例中,所述光源为一个光源,所述光源能够发射的光信号的波段范围包括所述第一波段范围,所述第二波段范围和第三波段范围。
可选地,在一些实施例中,所述光源包括第一子光源,第二子光源和第三子光源,分别用于发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号。
可选地,在一些实施例中,所述方法40还包括:
控制所述光源向所述人脸依次发射所述第一光信号,所述第二光信号和所述第三光信号。
可选地,在一些实施例中,所述通过光学传感器的第一像素单元集合依次接收所述至少两种光信号从所述人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,包括:
通过所述第一像素单元集合中的像素单元依次接收所述第一光信号,第二光信号和第三光信号分别从所述人脸反射的反射光信号,并根据所述反射光信号分别获取第一人脸图像,第二人脸图像和第三人脸图像。
可选地,在一些实施例中,所述方法40还包括:
根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
可选地,在一些实施例中,所述根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假,包括:
根据校准参数,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像,确定所述人脸的真假。
可选地,在一些实施例中,所述方法40还包括:
根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数。
可选地,在一些实施例中,所述根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数,包括:
在所述光源向所述参考对象依次发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号时,通过所述第一像素单元集合中的像素单元分别采集所述参考对象的三幅图像;
根据所述三幅图像的像素值的关系,确定所述校准参数,所述校准参数用于使得所述三幅图像的像素值在同一范围内。
可选地,在一些实施例中,所述根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像,确定所述人脸的真假,包括:
根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,其中,所述彩色人脸图像中的每个像素包括所述人脸对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的三个响应像素值;
将所述彩色人脸图像进行特征提取,得到所述彩色人脸图像的特征信息;
根据所述彩色人脸图像的特征信息,确定所述人脸的真假。
可选地,在一些实施例中,所述根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,包括:
根据第一像素单元集合中的第一像素单元在所述第一人脸图像中的采样像素值,确定所述彩色人脸图像中的第一像素对应的第一响应像素值,其中,所述第一响应像素值表示所述人脸对所述第一波段范围的光谱的响应,所述第一像素对应所述第一像素单元;
根据第一像素单元集合中的第一像素单元在所述第二人脸图像中的采样像素值,确定所述第一像素对应的第二响应像素值,其中,所述第二响应像素值表示所述人脸对所述第二波段范围的光谱的响应;以及
根据第一像素单元集合中的第一像素单元在所述第三人脸图像中的采样像素值,确定所述第一像素对应的第三响应像素值,其中,所述第三响应像素值表示所述人脸对所述第三波段范围的光谱的响应。
可选地,在一些实施例中,所述根据所述彩色人脸图像的特征信息,确定所述人脸的真假,包括:通过深度学习网络对所述彩色人脸图像的特征信息进行处理,确定所述人脸的真假。
可选地,在一些实施例中,所述方法40还包括:
根据所述像素阵列中除所述第一像素单元集合以外的其他像素单元采 集的人脸图像进行人脸识别。
可选地,在一些实施例中,所述根据所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像进行人脸识别,包括:
在所述人脸图像与注册的人脸图像匹配且所述人脸为真实人脸的情况下,确定人脸识别成功。
以下,结合图5,说明根据本申请实施例的用于人脸识别的方法的整体流程,如图5所示,该方法可以包括如下内容:
S51,在光源发送第一光信号,第二光信号和第三光信号时,通过第一像素单元集合分别采集第一人脸图像,第二人脸图像和第三人脸图像;
在S52中,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
具体实现参考前述实施例的相关说明,这里不再赘述。
在S53中,根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像;
在S54中,提取所述彩色人脸图像的色彩特征信息,例如HSV信息;
在S55中,根据所述彩色人脸图像的色彩特征信息进行分类,确定人脸的真假。具体地,可以将该彩色人脸图像输入到深度学习网络,以确定人脸的真假。
如图6所示,本申请实施例还提供了一种电子设备60,所述电子设备60可以包括用于人脸识别的装置61,该用于人脸识别的装置61可以为前述装置实施例中的用于人脸识别的装置30,其能够用于执行图4至图5中所述方法实施例中的内容,为了简洁,这里不再赘述。
应理解,本申请实施例的处理器或处理单元可以是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法实施例的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。上述的处理器可以是通用处理器、数字信号处理器(Digital Signal Processor,DSP)、专用集成电路(Application Specific Integrated Circuit,ASIC)、现成可编程门阵列(Field Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合本申请实施例所公开的方法的步骤可以直接体 现为硬件译码处理器执行完成,或者用译码处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器,处理器读取存储器中的信息,结合其硬件完成上述方法的步骤。
可以理解,本申请实施例的人脸识别还可以包括存储器,存储器可以是易失性存储器或非易失性存储器,或可包括易失性和非易失性存储器两者。其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),其用作外部高速缓存。通过示例性但不是限制性说明,许多形式的RAM可用,例如静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDR SDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synchlink DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DR RAM)。应注意,本文描述的系统和方法的存储器旨在包括但不限于这些和任意其它适合类型的存储器。
本申请实施例还提出了一种计算机可读存储介质,该计算机可读存储介质存储一个或多个程序,该一个或多个程序包括指令,该指令当被包括多个应用程序的便携式电子设备执行时,能够使该便携式电子设备执行图4至图5所示实施例的方法。
本申请实施例还提出了一种计算机程序,该计算机程序包括指令,当该计算机程序被计算机执行时,使得计算机可以执行图4至图5所示实施例的方法。
本申请实施例还提供了一种芯片,该芯片包括输入输出接口、至少一个处理器、至少一个存储器和总线,该至少一个存储器用于存储指令,该至少一个处理器用于调用该至少一个存储器中的指令,以执行图4至图5所示实施例的方法。
本领域普通技术人员可以意识到,结合本文中所公开的实施例描述的各示例的单元及算法步骤,能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行,取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能,但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到,为描述的方便和简洁,上述描述的系统、装置和单元的具体工作过程,可以参考前述方法实施例中的对应过程,在此不再赘述。
在本申请所提供的几个实施例中,应所述理解到,所揭露的系统、装置和方法,可以通过其它的方式实现。例如,以上所描述的装置实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。另一点,所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口,装置或单元的间接耦合或通信连接,可以是电性,机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本申请各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理存在,也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时,可以存储在一个计算机可读取存储介质中。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分或者所述技术方案的部分可以以软件产品的形式体现出来,所述计算机软件产品存储在一个存储介质中,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括:U盘、移动硬盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述,仅为本申请的具体实施方式,但本申请的保护范围并不局限于此,任何熟悉本技术领域的技术人员在本申请揭露的技术范围内,可轻易想到变化或替换,都应涵盖在本申请的保护范围之内。因此,本申请的保护范围应所述以权利要求的保护范围为准。

Claims (43)

  1. 一种用于人脸识别的光学传感器,其特征在于,包括:
    像素阵列,所述像素阵列中的第一像素单元集合用于接收由光源依次发射的至少两种光信号从人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,其中,所述至少两种光信号对应不同的波段范围,所述至少两个人脸图像用于确定所述人脸的真假。
  2. 根据权利要求1所述的光学传感器,其特征在于,所述至少两种光信号包括第一光信号、第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围,所述第一波段范围、所述第二波段范围和所述第三波段范围两两不同;
    所述至少两个人脸图像包括第一人脸图像、第二人脸图像和第三人脸图像,所述第一人脸图像、第二人脸图像和第三人脸图像分别由所述第一光信号、所述第二光信号和所述第三光信号经所述人脸反射后获得,所述第一人脸图像、所述第二人脸图像和所述第三人脸图像用于确定所述人脸的真假。
  3. 根据权利要求2所述的光学传感器,其特征在于,所述第一波段范围,所述第二波段范围和所述第三波段范围分别为以下三种波段范围中的一种:
    包括560nm的波段范围,包括980nm的波段范围,包括940nm的波段范围。
  4. 根据权利要求1至3中任一项所述的光学传感器,其特征在于,所述第一像素单元集合中的像素单元不设置滤光片。
  5. 根据权利要求1至4中任一项所述的光学传感器,其特征在于,所述像素阵列中除所述第一像素单元集合之外的其他像素单元不设置滤光片,或设置特定波段范围的滤光片。
  6. 根据权利要求5所述的光学传感器,其特征在于,所述特定波段范围的滤光片为包括940nm波段范围的滤光片。
  7. 根据权利要求1至6中任一项所述的光学传感器,其特征在于,所述第一像素单元集合包括所述像素阵列中的部分像素单元。
  8. 根据权利要求1至7中任一项所述的光学传感器,其特征在于,所述第一像素单元集合中的像素单元的数量与所述像素阵列中的像素单元的 总数量的比例小于第一比值。
  9. 根据权利要求1至8中任一项所述的光学传感器,其特征在于,所述第一像素单元集合中连续的像素单元的数量小于或等于特定阈值。
  10. 根据权利要求1至9中任一项所述的光学传感器,其特征在于,所述第一像素单元集合中的像素单元离散分布在所述像素阵列中。
  11. 根据权利要求7至10中任一项所述的光学传感器,其特征在于,所述像素阵列中除所述第一像素单元集合之外的其他像素单元采集的人脸图像用于人脸识别。
  12. 一种用于人脸识别的装置,其特征在于,包括:
    如权利要求1至11中任一项所述的光学传感器;
    光源,用于依次发射至少两种光信号,所述至少两种光信号对应不同的波段范围;
    其中,所述光学传感器的第一像素单元集合用于接收所述光源依次发射的所述至少两种光信号从人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,所述至少两个人脸图像用于确定所述人脸的真假。
  13. 根据权利要求12所述的装置,其特征在于,所述光源具体用于:
    向所述人脸依次发射第一光信号,第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围,所述第一波段范围,所述第二波段范围和所述第三波段范围两两不同。
  14. 根据权利要求13所述的装置,其特征在于,所述光源为一个光源,所述光源能够发射的光信号的波段范围包括所述第一波段范围,所述第二波段范围和第三波段范围。
  15. 根据权利要求13所述的装置,其特征在于,所述光源包括第一子光源,第二子光源和第三子光源,分别用于发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号。
  16. 根据权利要求13至15中任一项所述的装置,其特征在于,所述装置还包括:光源控制模块,用于控制所述光源向所述人脸依次发射所述第一光信号,所述第二光信号和所述第三光信号。
  17. 根据权利要求13至16中任一项所述的装置,其特征在于,所述光学传感器具体用于:
    通过所述第一像素单元集合依次接收所述第一光信号,第二光信号和第三光信号分别从所述人脸反射的反射光信号,并根据所述反射光信号分别获取第一人脸图像,第二人脸图像和第三人脸图像。
  18. 根据权利要求17所述的装置,其特征在于,所述装置还包括:
    处理器,用于根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
  19. 根据权利要求18所述的装置,其特征在于,所述处理器还用于:
    根据校准参数,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
    根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
  20. 根据权利要求19所述的装置,其特征在于,所述处理器还用于:
    根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数。
  21. 根据权利要求20所述的装置,其特征在于,所述光学传感器还用于:在所述光源向所述参考对象依次发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号时,分别采集所述参考对象的三幅图像;
    所述处理器还用于:
    根据所述三幅图像的像素值的关系,确定所述校准参数,所述校准参数用于使得所述三幅图像的像素值在同一范围内。
  22. 根据权利要求20或21所述的装置,其特征在于,所述处理器还用于:
    根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,其中,所述彩色人脸图像中的每个像素包括所述人脸对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的三个响应像素值;
    将所述彩色人脸图像进行特征提取,得到所述彩色人脸图像的特征信息;
    根据所述彩色人脸图像的特征信息,确定所述人脸的真假。
  23. 根据权利要求22所述的装置,其特征在于,所述处理器具体用于:
    根据第一像素单元集合中的第一像素单元在所述第一人脸图像中的采 样像素值,确定所述彩色人脸图像中的第一像素对应的第一响应像素值,其中,所述第一响应像素值表示所述人脸对所述第一波段范围的光谱的响应,所述第一像素对应所述第一像素单元;
    根据第一像素单元集合中的第一像素单元在所述第二人脸图像中的采样像素值,确定所述第一像素对应的第二响应像素值,其中,所述第二响应像素值表示所述人脸对所述第二波段范围的光谱的响应;以及
    根据第一像素单元集合中的第一像素单元在所述第三人脸图像中的采样像素值,确定所述第一像素对应的第三响应像素值,其中,所述第三响应像素值表示所述人脸对所述第三波段范围的光谱的响应。
  24. 根据权利要求22或23所述的装置,其特征在于,所述处理器具体用于:通过深度学习网络对所述彩色人脸图像的特征信息进行处理,确定所述人脸的真假。
  25. 根据权利要求18至24中任一项所述的装置,其特征在于,所述处理器还用于:
    将所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定人脸识别是否成功。
  26. 根据权利要求25所述的装置,其特征在于,所述处理器还用于:
    在所述人脸图像与注册的所述人脸图像模板匹配且所述人脸为真实人脸的情况下,确定人脸识别成功。
  27. 一种用于人脸识别的方法,其特征在于,所述方法包括:
    在光源依次发射至少两种光信号时,通过光学传感器的第一像素单元集合依次接收所述至少两种光信号从所述人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,其中,所述至少两种光信号对应不同的波段范围;
    根据所述至少两个人脸图像确定所述人脸的真假。
  28. 根据权利要求27所述的方法,其特征在于,所述光源依次向人脸发射至少两种光信号,包括:
    所述光源依次向所述人脸发射第一光信号,第二光信号和第三光信号,其中,所述第一光信号对应第一波段范围,所述第二光信号对应第二波段范围,所述第三光信号对应第三波段范围,所述第一波段范围,所述第二波段范围和所述第三波段范围两两不同。
  29. 根据权利要求28所述的方法,其特征在于,所述第一波段范围,所述第二波段范围和所述第三波段范围分别为以下三种波段范围中的一种:包括560nm的波段范围,包括980nm的波段范围,包括940nm的波段范围。
  30. 根据权利要求28或29所述的方法,其特征在于,所述光源为一个光源,所述光源能够发射的光信号的波段范围包括所述第一波段范围,所述第二波段范围和第三波段范围。
  31. 根据权利要求28至30中任一项所述的方法,其特征在于,所述光源包括第一子光源,第二子光源和第三子光源,分别用于发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号。
  32. 根据权利要求28至31中任一项所述的方法,其特征在于,所述方法还包括:控制所述光源向所述人脸依次发射所述第一光信号,所述第二光信号和所述第三光信号。
  33. 根据权利要求28至32中任一项所述的方法,其特征在于,所述通过光学传感器的第一像素单元集合依次接收所述至少两种光信号从所述人脸反射的反射光信号,并根据所述反射光信号获取至少两个人脸图像,包括:
    通过所述第一像素单元集合中的像素单元依次接收所述第一光信号,第二光信号和第三光信号分别从所述人脸反射的反射光信号,并根据所述反射光信号分别获取第一人脸图像,第二人脸图像和第三人脸图像。
  34. 根据权利要求33所述的方法,其特征在于,所述方法还包括:
    根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假。
  35. 根据权利要求34所述的方法,其特征在于,所述根据所述第一人脸图像,所述第二人脸图像和所述第三人脸图像确定所述人脸的真假,包括:
    根据校准参数,对所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行校准;
    根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像,确定所述人脸的真假。
  36. 根据权利要求35所述的方法,其特征在于,所述方法还包括:
    根据参考对象对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数。
  37. 根据权利要求36所述的方法,其特征在于,所述根据参考对象对 所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的响应之间的关系,确定所述校准参数,包括:
    在所述光源向所述参考对象依次发射所述第一波段范围,所述第二波段范围和所述第三波段范围的光信号时,通过所述第一像素单元集合中的像素单元分别采集所述参考对象的三幅图像;
    根据所述三幅图像的像素值的关系,确定所述校准参数,所述校准参数用于使得所述三幅图像的像素值在同一范围内。
  38. 根据权利要求36或37所述的方法,其特征在于,所述根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像,确定所述人脸的真假,包括:
    根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,其中,所述彩色人脸图像中的每个像素包括所述人脸对所述第一波段范围,所述第二波段范围和所述第三波段范围的光谱的三个响应像素值;
    将所述彩色人脸图像进行特征提取,得到所述彩色人脸图像的特征信息;
    根据所述彩色人脸图像的特征信息,确定所述人脸的真假。
  39. 根据权利要求38所述的方法,其特征在于,所述根据校准后的所述第一人脸图像,所述第二人脸图像和所述第三人脸图像进行合成,得到彩色人脸图像,包括:
    根据第一像素单元集合中的第一像素单元在所述第一人脸图像中的采样像素值,确定所述彩色人脸图像中的第一像素对应的第一响应像素值,其中,所述第一响应像素值表示所述人脸对所述第一波段范围的光谱的响应,所述第一像素对应所述第一像素单元;
    根据第一像素单元集合中的第一像素单元在所述第二人脸图像中的采样像素值,确定所述第一像素对应的第二响应像素值,其中,所述第二响应像素值表示所述人脸对所述第二波段范围的光谱的响应;以及
    根据第一像素单元集合中的第一像素单元在所述第三人脸图像中的采样像素值,确定所述第一像素对应的第三响应像素值,其中,所述第三响应像素值表示所述人脸对所述第三波段范围的光谱的响应。
  40. 根据权利要求38或39所述的方法,其特征在于,所述根据所述彩 色人脸图像的特征信息,确定所述人脸的真假,包括:
    通过深度学习网络对所述彩色人脸图像的特征信息进行处理,确定所述人脸的真假。
  41. 根据权利要求38至40中任一项所述的方法,其特征在于,所述方法还包括:
    将所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定人脸识别是否成功。
  42. 根据权利要求41所述的方法,其特征在于,所述将所述像素阵列中除所述第一像素单元集合以外的其他像素单元采集的人脸图像与注册的人脸图像模板进行匹配,确定人脸识别是否成功,包括:
    在所述人脸图像与注册的所述人脸图像模板匹配且所述人脸为真实人脸的情况下,确定人脸识别成功。
  43. 一种电子设备,其特征在于,包括:
    如权利要求12至26中任一项所述的用于人脸识别的装置。
PCT/CN2019/088654 2019-05-27 2019-05-27 用于人脸识别的光学传感器、装置、方法和电子设备 WO2020237483A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201980000833.9A CN110337656A (zh) 2019-05-27 2019-05-27 用于人脸识别的光学传感器、装置、方法和电子设备
PCT/CN2019/088654 WO2020237483A1 (zh) 2019-05-27 2019-05-27 用于人脸识别的光学传感器、装置、方法和电子设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/088654 WO2020237483A1 (zh) 2019-05-27 2019-05-27 用于人脸识别的光学传感器、装置、方法和电子设备

Publications (1)

Publication Number Publication Date
WO2020237483A1 true WO2020237483A1 (zh) 2020-12-03

Family

ID=68150222

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/088654 WO2020237483A1 (zh) 2019-05-27 2019-05-27 用于人脸识别的光学传感器、装置、方法和电子设备

Country Status (2)

Country Link
CN (1) CN110337656A (zh)
WO (1) WO2020237483A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112800956A (zh) * 2021-01-27 2021-05-14 杭州海康威视数字技术股份有限公司 一种指纹活体检测系统、方法及装置
TWI794696B (zh) * 2020-12-14 2023-03-01 晶元光電股份有限公司 光學感測裝置
WO2023036131A1 (zh) * 2021-09-10 2023-03-16 维沃移动通信有限公司 图像信息采集模组、信息处理方法、装置和电子设备

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112101194A (zh) * 2020-01-21 2020-12-18 神盾股份有限公司 电子装置及其操作方法
CN111611977B (zh) * 2020-06-05 2021-10-15 吉林求是光谱数据科技有限公司 基于光谱与多波段融合的人脸识别监控系统及识别方法
CN113297977B (zh) * 2021-05-26 2023-12-22 奥比中光科技集团股份有限公司 活体检测方法、装置及电子设备
CN113609924A (zh) * 2021-07-15 2021-11-05 奥比中光科技集团股份有限公司 一种光谱数据的确定方法、装置、终端和存储介质

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102622588A (zh) * 2012-03-08 2012-08-01 无锡数字奥森科技有限公司 双验证人脸防伪方法及装置
CN102831400A (zh) * 2012-07-31 2012-12-19 西北工业大学 一种多光谱人脸识别方法及其系统
CN105160289A (zh) * 2015-07-03 2015-12-16 深圳市金立通信设备有限公司 一种人脸识别方法及终端
CN107169405A (zh) * 2017-03-17 2017-09-15 上海云从企业发展有限公司 基于双目摄像机活体识别的方法及装置
CN107408298A (zh) * 2015-03-13 2017-11-28 日本电气株式会社 生命体检测设备、生命体检测方法和记录介质
CN107808115A (zh) * 2017-09-27 2018-03-16 联想(北京)有限公司 一种活体检测方法、装置及存储介质
WO2018079031A1 (ja) * 2016-10-31 2018-05-03 日本電気株式会社 画像処理装置、画像処理方法、顔認証システム、プログラム及び記録媒体
US20180196998A1 (en) * 2017-01-11 2018-07-12 Microsoft Technology Licensing, Llc Infrared imaging recognition enhanced by 3d verification
CN108710844A (zh) * 2018-05-14 2018-10-26 安徽质在智能科技有限公司 对面部进行检测的认证方法和装置
KR20180134280A (ko) * 2017-06-08 2018-12-18 (주)이더블유비엠 3차원 깊이정보 및 적외선정보에 기반하여 생체여부의 확인을 행하는 얼굴인식 장치 및 방법

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5988093B2 (ja) * 2012-07-30 2016-09-07 株式会社豊田中央研究所 画像処理装置、物体識別装置、及びプログラム
CN103268499B (zh) * 2013-01-23 2016-06-29 北京交通大学 基于多光谱成像的人体皮肤检测方法
CN107609459B (zh) * 2016-12-15 2018-09-11 平安科技(深圳)有限公司 一种基于深度学习的人脸识别方法及装置
CN107205139A (zh) * 2017-06-28 2017-09-26 重庆中科云丛科技有限公司 多通道采集的图像传感器及采集方法

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102622588A (zh) * 2012-03-08 2012-08-01 无锡数字奥森科技有限公司 双验证人脸防伪方法及装置
CN102831400A (zh) * 2012-07-31 2012-12-19 西北工业大学 一种多光谱人脸识别方法及其系统
CN107408298A (zh) * 2015-03-13 2017-11-28 日本电气株式会社 生命体检测设备、生命体检测方法和记录介质
CN105160289A (zh) * 2015-07-03 2015-12-16 深圳市金立通信设备有限公司 一种人脸识别方法及终端
WO2018079031A1 (ja) * 2016-10-31 2018-05-03 日本電気株式会社 画像処理装置、画像処理方法、顔認証システム、プログラム及び記録媒体
US20180196998A1 (en) * 2017-01-11 2018-07-12 Microsoft Technology Licensing, Llc Infrared imaging recognition enhanced by 3d verification
CN107169405A (zh) * 2017-03-17 2017-09-15 上海云从企业发展有限公司 基于双目摄像机活体识别的方法及装置
KR20180134280A (ko) * 2017-06-08 2018-12-18 (주)이더블유비엠 3차원 깊이정보 및 적외선정보에 기반하여 생체여부의 확인을 행하는 얼굴인식 장치 및 방법
CN107808115A (zh) * 2017-09-27 2018-03-16 联想(北京)有限公司 一种活体检测方法、装置及存储介质
CN108710844A (zh) * 2018-05-14 2018-10-26 安徽质在智能科技有限公司 对面部进行检测的认证方法和装置

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI794696B (zh) * 2020-12-14 2023-03-01 晶元光電股份有限公司 光學感測裝置
CN112800956A (zh) * 2021-01-27 2021-05-14 杭州海康威视数字技术股份有限公司 一种指纹活体检测系统、方法及装置
WO2023036131A1 (zh) * 2021-09-10 2023-03-16 维沃移动通信有限公司 图像信息采集模组、信息处理方法、装置和电子设备

Also Published As

Publication number Publication date
CN110337656A (zh) 2019-10-15

Similar Documents

Publication Publication Date Title
WO2020237483A1 (zh) 用于人脸识别的光学传感器、装置、方法和电子设备
US10943083B2 (en) Fingerprint identification apparatus and method and terminal device
US11475706B2 (en) Fingerprint identification device, method and electronic device
US11869255B2 (en) Anti-counterfeiting face detection method, device and multi-lens camera
US9848113B2 (en) Multi-band biometric camera system having iris color recognition
US10984267B2 (en) Method and system for imaging acquisition
CN110462633B (zh) 一种人脸识别的方法、装置和电子设备
WO2020237482A1 (zh) 用于人脸识别的光学传感器、装置、方法和电子设备
US8861847B2 (en) System and method for adaptive skin tone detection
US8705815B2 (en) Systems and methods for automatic skin-based identification of people using digital images
WO2018095293A1 (zh) 人脸图像处理方法、终端及存储介质
US8831296B2 (en) Biometric authentication apparatus, biometric authentication method, and program
TWI727219B (zh) 用於產生影像之表示之方法、成像系統、及機器可讀儲存裝置
CN107742274A (zh) 图像处理方法、装置、计算机可读存储介质和电子设备
WO2022110846A1 (zh) 一种活体检测的方法及设备
US10853624B2 (en) Apparatus and method
CN107820005A (zh) 图像处理方法、装置和电子装置
US20200311404A1 (en) Biometric identification using composite hand images
WO2021027155A1 (zh) 基于指静脉图像的验证方法、装置、存储介质及计算机设备
WO2019000294A1 (zh) 指纹识别装置和终端设备
CN107578372A (zh) 图像处理方法、装置、计算机可读存储介质和电子设备
WO2018014851A1 (zh) 一种生物特征识别的方法,装置及存储介质
US20210201492A1 (en) Image-based skin diagnostics
CN107770446A (zh) 图像处理方法、装置、计算机可读存储介质和电子设备
CN111062333B (zh) 活体的面部动态识别方法、系统、存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19930907

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19930907

Country of ref document: EP

Kind code of ref document: A1