WO2023109874A1 - 识别系统及其工作方法 - Google Patents

识别系统及其工作方法 Download PDF

Info

Publication number
WO2023109874A1
WO2023109874A1 PCT/CN2022/139150 CN2022139150W WO2023109874A1 WO 2023109874 A1 WO2023109874 A1 WO 2023109874A1 CN 2022139150 W CN2022139150 W CN 2022139150W WO 2023109874 A1 WO2023109874 A1 WO 2023109874A1
Authority
WO
WIPO (PCT)
Prior art keywords
spectral
spectral response
information
subject
response result
Prior art date
Application number
PCT/CN2022/139150
Other languages
English (en)
French (fr)
Inventor
李丽
黄志雷
汪舟
武振华
Original Assignee
北京与光科技有限公司
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 北京与光科技有限公司 filed Critical 北京与光科技有限公司
Publication of WO2023109874A1 publication Critical patent/WO2023109874A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/117—Identification of persons
    • A61B5/1171—Identification of persons based on the shapes or appearances of their bodies or parts thereof
    • A61B5/1172—Identification of persons based on the shapes or appearances of their bodies or parts thereof using fingerprinting
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
    • G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
    • G01J3/28—Investigating the spectrum
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
    • G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
    • 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
    • G06V40/12—Fingerprints or palmprints
    • 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
    • G06V40/12—Fingerprints or palmprints
    • G06V40/13—Sensors therefor

Definitions

  • the present application relates to the field of spectrum application, and more specifically, relates to an identification system and its working method.
  • biometric systems are increasingly used to provide greater security and/or enhanced user convenience.
  • fingerprint sensing systems have been widely used in various terminal devices, such as consumer smartphones, due to their small size, high performance, and high user acceptance.
  • fingerprint sensing systems there are many kinds of fingerprint sensing systems in the market, such as sensing systems based on capacitive fingerprint modules, sensing systems based on optical fingerprint modules, etc.
  • sensing systems based on capacitive fingerprint modules such as sensing systems based on optical fingerprint modules, etc.
  • fingerprint sensing systems can realize unlocking, they are still being used After the fingerprint identification of the mobile terminal is unlocked, criminals can steal the user's fingerprint to make a fake fingerprint to crack the user's security system, which instead increases the probability of the mobile terminal's fingerprint password being discovered, and has caused a great impact on the information security of the mobile terminal. threat.
  • liveness detection to protect biometric systems from attacks that exploit spoofed body parts, such as spoofed fingerprints.
  • the existing biometric systems with liveness detection function also have certain defects, which will fail in some scenarios.
  • a liveness detection scheme based on image processing its image-based software method clearly checks A sweating effect that introduces differences between image frames, but not all fingers exhibit the desired amount of sweating, especially in winter conditions.
  • Embodiments of the present application provide an identification system and its working method, wherein, the identification system obtains the spectral information and image information of the measured target through multiple projections of detection light, and uses the spectral information and image information obtained by multiple projections image information to improve the accuracy of liveness detection and object recognition.
  • the detection lights projected multiple times are different types of detection lights.
  • a working method of an identification system which includes:
  • Live body detection and object recognition are performed based on the first spectral information, the image information, and the second spectral information.
  • the first detection light and the second detection light are different types of light signals.
  • the first detection light is a mixed light containing at least two monochromatic lights of different wavelength bands
  • the second detection light is monochromatic light
  • the first detection light is white light
  • the second detection light is blue light
  • receiving the first detection light reflected back by the subject and generating first spectral information and image information of the subject based on the first detection light comprising: modulating at least a part of the first detection light through a light filtering structure of a receiving module to generate a modulated light signal; and receiving the modulated light signal through a modulation area of an image sensor of the receiving module to generate the first spectral information and receive the part of the first detection light not modulated by the filter structure through the non-modulation area of the image sensor to generate the image information.
  • the image sensor includes a non-modulation area for generating the image information and a modulation area for generating the first spectral information, and the filter structure corresponds to the the modulation area.
  • the ratio of the area of the modulation area to the area of the effective area of the image sensor is 12%-25%.
  • performing living body detection and object recognition based on the first spectral information, the image information and the second spectral information includes: performing the first spectral information and the processing the second spectral information to generate a first spectral response result and a second spectral response result; processing the image information to generate an image of the subject; comparing the image of the subject with a prestored reference comparing images; and, in response to successful matching between the image of the subject and the reference image, judging that the subject is based on the first spectral response result and/or the second spectral response result Whether the object is alive.
  • processing the image information to generate the image of the subject includes: optimizing the image of the subject with the first spectral information.
  • processing the first spectral information and the second spectral information to generate a first spectral response result and a second spectral response result includes: converting the first spectral converting the information into a one-dimensional spectral feature vector as the first spectral response result; and converting the second spectral information into a one-dimensional spectral feature vector as the second spectral response result.
  • processing the first spectral information and the second spectral information to generate a first spectral response result and a second spectral response result includes: The image information of the pixel points in the adjacent area of each pixel point in the spectral information is used to adjust the spectral information of the pixel point to obtain the first adjusted spectral information; the first adjusted spectral information is converted into a one-dimensional spectral feature vector as the obtained the first spectral response result; and converting the second spectral information into a one-dimensional spectral feature vector as the second spectral response result.
  • judging whether the subject is a living body based on the first spectral response result and/or the second spectral response result includes: using the first spectral response comparing the result to a reference spectral response; comparing the second spectral response result to the reference spectral response; and, in response to a successful match between the first spectral response result and the reference spectral response and the second The matching between the spectral response result and the reference spectral response is successful, and it is determined that the subject is a living body.
  • judging whether the subject is a living body based on the first spectral response result and/or the second spectral response result includes: fusing the first spectral response and the second spectral response result to obtain a fusion spectral response result; and, based on a comparison between the fusion spectral response result and a reference spectral response, determine whether the subject is a living body.
  • comparing the first spectral response result with a reference spectral response includes: calculating a first average value of values at various positions in the first spectral response result; and, In response to the first average value being within a preset range, it is determined that the matching between the first spectral response result and the reference spectral response is successful, wherein the lower limit of the preset range is the reference spectral response
  • the reference mean value minus half of the standard deviation of the reference spectral response, and the upper limit of the preset range is the reference mean value plus half of the standard deviation of the reference spectral response.
  • an identification system is also provided, wherein the identification system can execute the above-mentioned working method.
  • the recognition system obtains the spectral information and image information of the measured target through multiple projections of detection light, and uses the spectrum obtained by multiple projections Information and image information to improve the accuracy of liveness detection and object recognition.
  • the detection lights projected multiple times are different types of detection lights.
  • Fig. 1 illustrates a block diagram of an identification system according to an embodiment of the present application.
  • Fig. 2 illustrates a block diagram of a receiving module of the identification system according to an embodiment of the present application.
  • Fig. 3 illustrates a block diagram of an image sensor in the receiving module according to an embodiment of the present application.
  • Fig. 4 illustrates a flow chart of the working method of the identification system according to the embodiment of the present application.
  • Fig. 5 illustrates a flow chart of living body detection and object recognition based on the first spectral information, the image information and the second spectral information in the working method of the recognition system according to the embodiment of the present application.
  • Fig. 6 illustrates a schematic diagram of an image sensor in the recognition system according to an embodiment of the present application.
  • Fig. 7 illustrates a schematic diagram of the system architecture of the receiving module in the identification system according to the embodiment of the present application.
  • the existing biometric systems with liveness detection function also have certain defects, which will fail in some scenarios.
  • a liveness detection scheme based on image processing its image-based software method . Hence look at the effect of sweating that introduces differences between image frames, but not all fingers show the desired amount of sweating, especially in winter conditions. Therefore, an optimized biometric system is desired.
  • an identification system which includes a light source assembly 110 and a receiving module 120 (for example, the receiving module 120 is implemented as a fingerprint module assembly), wherein the light source assembly 110 is suitable for projecting
  • the detection light is sent to the subject (for example, a finger of a human hand), and the detection light reflected from the subject is received by the receiving module 120 to generate spectral information and/or image information of the subject, so that based on the spectral information and image information for liveness detection and object recognition.
  • the recognition system obtains the spectral information and image information of the measured target by projecting the detection light multiple times, and uses multiple The spectral information and image information obtained by the first projection are used to improve the accuracy of liveness detection and object recognition.
  • the detection lights projected multiple times are different types of detection light, so as to improve the security level of the living body detection.
  • the recognition system belongs to the biometric system, which performs liveness detection and object recognition by collecting spectral information and image information of the subject .
  • the identification system includes a light source assembly 110 and a receiving module 120, wherein the light source assembly 110 is suitable for projecting detection light to the subject, and the detection light reflected from the subject is received by the receiving module 120 Spectral information and/or image information of the subject is received and generated, so as to perform living body detection and object recognition based on the spectral information and image information.
  • the identification system when the identification system is implemented as a fingerprint identification system, the light source assembly 110 of the identification system is used to project detection light to the subject (under normal use, the subject is a real finger, in the scene of intrusion , the subject may be a finger model, a finger pattern, etc.), the receiving module 120 of the recognition system is used to receive the detection light reflected back from the subject to obtain the spectral information and image information of the subject, that is , under normal use, the spectral information is the spectral information of a real finger (which characterizes the physiological characteristics of a real finger), and the image information is an image of a real finger.
  • the identification system can perform living body detection and fingerprint identification based on the image information and the spectral information.
  • the recognition system obtains the spectral information of the subject based on the computational spectrum technology, that is, in the embodiment of the present application, the receiving module 120 obtains the spectral information of the subject based on the computational spectral technology.
  • the spectral information of the subject is, in the embodiment of the present application, the recognition system obtains the spectral information of the subject based on the computational spectrum technology.
  • the image sensor 122 of the above-mentioned circuit board 121 and the filter structure 123 held on the photosensitive path of the image sensor 122 .
  • the circuit board 121 may be a flexible board (FPC), a rigid board (PCB), a rigid-flex board (F-PCB), a ceramic substrate, and the like.
  • the image sensor 122 and the filter structure 123 constitute a spectrum chip 1000 , wherein the filter structure 123 is a broadband filter structure 123 in the frequency domain or wavelength domain.
  • the filter structure 123 can be metasurface, photonic crystal, nanocolumn, multilayer film, dye, quantum dot, MEMS (micro-electromechanical system), FP etalon (FP etalon), cavity layer (resonant cavity layer), waveguide Layer (waveguide layer), diffraction elements and other structures or materials with filter properties.
  • the light filtering structure 123 may be the light modulation layer in Chinese patent CN201921223201.2.
  • the receiving module 120 can also include an optical system 124, for example, it can be an optical system 124 such as a lens assembly, a uniform light assembly, etc., and the optical system 124 is located at the front end of the filter structure 123 , the detection light from the outside enters after being adjusted by the optical system 124 and is modulated by the filter structure 123, and then received by the image sensor 122 to obtain a spectral response.
  • an optical system 124 for example, it can be an optical system 124 such as a lens assembly, a uniform light assembly, etc.
  • the optical system 124 is located at the front end of the filter structure 123 , the detection light from the outside enters after being adjusted by the optical system 124 and is modulated by the filter structure 123, and then received by the image sensor 122 to obtain a spectral response.
  • the image sensor 122 may be a CMOS image sensor (CIS), a CCD, an array photodetector, or the like.
  • the receiving module 120 may also include a data processing unit 125, and the data processing unit 125 may be a processing unit such as MCU, CPU, GPU, FPGA, NPU, ASIC, etc., which may export the data generated by the image sensor 122 to processed externally. For example, the spectral response measured by the image sensor 122 is transmitted to the data processing unit 125 for spectral recovery calculation.
  • the intensity signals of incident light at different wavelengths ⁇ are denoted as x( ⁇ )
  • the transmission spectrum curve of the filter structure 123 is denoted as T( ⁇ )
  • the filter structure 123 has m groups of structural units, each The transmission spectra of the group structural units are different from each other.
  • the physical pixel refers to a pixel of the image sensor 122 ), which detects the light intensity bi modulated by the filtering light structure 123 .
  • a physical pixel is used, that is, a physical pixel corresponds to a group of structural units, but it is not limited thereto. In other embodiments, a group of multiple physical pixels may also correspond to a group Structural units. Therefore, in the receiving module 120 according to the embodiment of the present application, at least two groups of structural units form a "spectral pixel".
  • the effective transmission spectrum of the filter structure 123 (the transmission spectrum used for spectral recovery, called the effective transmission spectrum) Ti( ⁇ ) number and the number of structural units may be inconsistent, and the filter structure 123
  • the transmission spectrum is artificially set, tested, or calculated according to certain rules according to the requirements of identification or recovery (for example, the transmission spectrum of each structural unit above is an effective transmission spectrum), so the filter structure 123
  • the number of effective transmission spectra may be less than the number of structural units, or may even be more than the number of structural units; in this modified embodiment, a certain transmission spectrum curve is not necessarily determined by a group of structural units.
  • the present invention can use at least one spectral pixel to restore an image.
  • R( ⁇ ) is the response of the image sensor, recorded as:
  • light intensity measurement values corresponding to m physical pixels
  • A is the light response of the system to different wavelengths, which is determined by two factors: the transmittance of the filter structure and the quantum efficiency of the image sensor.
  • A is a matrix, and each row vector corresponds to the response of a group of structural units to incident light of different wavelengths.
  • the incident light is discretely and uniformly sampled, and there are n sampling points in total.
  • the number of columns of A is the same as the number of sampling points of the incident light.
  • x( ⁇ ) is the light intensity of the incident light at different wavelengths ⁇ , that is, the spectrum of the incident light to be measured.
  • the filter structure 123 can be directly formed on the upper surface of the image sensor 122, such as quantum dots, nanowires, etc., which are directly on the photosensitive area of the image sensor 122 Forming the filter structure 123 or materials (nanowires, quantum dots, etc.), taking the filter structure 123 as an example, at this time, it can be understood that when the raw material of the image sensor 122 is processed to form the image sensor 122, on the raw material The surface is processed to form a filter structure 123, and the transmission spectrum and the response of the image sensor 122 are integrated, that is, it can be understood that the response of the detector and the transmission spectrum are on the same curve, and the spectral distribution of the incident light at this time and the relationship between the light intensity measurement value of the image sensor 122 can be expressed by the following formula:
  • the relationship between the spectral distribution of the incident light and the light intensity measurement value of the image sensor 122 can be expressed by the following formula:
  • multiple physical pixels may also correspond to a group of structural units. It can be further defined that a group of structural units and at least one corresponding physical pixel constitute a unit pixel, and in principle, at least one unit pixel constitutes a spectral pixel.
  • a snapshot spectral imaging device can be realized by arraying the spectral pixels, that is, the receiving device can be used not only for spectral recovery but also for spectral imaging.
  • the recognition system obtains the spectral information and image information of the measured target through multiple projections of detection light, and uses multiple The acquired spectral information and image information are projected to improve the accuracy of liveness detection and object recognition.
  • the detection lights projected multiple times are different types of detection light, so as to improve the security level of the living body detection.
  • the light source assembly 110 includes a controller 111 and at least one light source 123 electrically connected to the controller 111, wherein the controller 111 can control the light source 123 to project Different types of detection light, such as mixed light and monochromatic light, enable the receiving module 120 to collect spectral information and/or image information of the subject.
  • the light source 123 can be implemented as a screen light source, which includes at least one red light emitting point, at least one green light emitting point and at least one blue light emitting point, wherein, in a possible working mode , the light source 123 can be mixed and projected by at least two light-emitting points, for example, the red light-emitting point and the green light-emitting point work simultaneously to project mixed light; in another possible working mode, the light source 123 can be composed of A luminescent point works to cast monochromatic light.
  • the selection of the light source 123 can be adjusted based on the requirements of the application scenario. For example, when the identification system is a fingerprint identification system, the light source used should make the real skin and fake skin as far as possible. The absorption and reflection properties of the projected detection light are different, thereby improving the recognition accuracy.
  • Fig. 4 illustrates a flow chart of the working method of the identification system according to the embodiment of the present application.
  • the working method of the recognition system includes the steps of: S110, projecting the first detection light to the subject; S120, receiving the first detection light reflected back by the subject Detecting light and generating first spectral information and image information of the object based on the first detection light; S130, projecting second detection light to the object; S140, receiving the light reflected by the object Generate the second spectral information of the subject based on the returned second detected light; and, S150, based on the first spectral information, the image information and the second spectral information for liveness detection and object recognition.
  • S110 projecting the first detection light to the subject
  • S120 receiving the first detection light reflected back by the subject Detecting light and generating first spectral information and image information of the object based on the first detection light
  • S130 projecting second detection light to the object
  • S140 receiving the light reflected by the object Generate the second spectral
  • the first detection light and the second detection light are different types of light signals, so that there is a difference between the first spectral information and the second spectral information, through In this way, the accuracy and security of liveness detection can be improved.
  • the first detection light and the second detection light may also be the same type of detection light. Through two spectral measurements and comparisons, it is also possible to improve the detection of living body to a certain extent.
  • the accuracy and security for example, can be simply understood as a process of secondary detection or secondary confirmation.
  • the first detection light is a mixed light containing at least two monochromatic lights of different wavelength bands, that is, the light source assembly 110 projects the mixed light to the subject object.
  • the mixed light may be white mixed light mixed with red light, green light and blue light emitted by the screen light source.
  • the receiving module 120 receives the first detection light reflected back by the subject and generates first spectral information and image information of the subject based on the first detection light.
  • the first detection light is modulated by the filter structure 123 of the receiving module 120 to generate a modulated light signal, that is, the filter structure 123 of the receiving module 120 At least a part of the first detection light is modulated to generate the modulated light signal, while the rest of the first detection light is not modulated.
  • the modulated light information is received in the modulation area 1221 of the image sensor 122 to generate the first spectral information (for this process, please refer to the description part of the computational spectral technology, which will not be repeated here) and, in the The non-modulation area 1222 of the image sensor 122 receives the unmodulated part of the first detection light to generate the image information.
  • the image sensor 122 includes a non-modulation area 1222 for generating the image information and a modulation area 1221 for generating the first spectral information
  • the filter structure 123 corresponds to the modulation area 1221 . That is to say, the receiving module 120 can not only collect the spectral information of the subject but also collect the image information of the subject.
  • the ratio of the area of the modulation region 1221 to the area of the effective region of the image sensor 122 is 10%-50%.
  • the area of the modulation region 1221 accounts for the area of the effective region of the image sensor 122 The ratio is 12%-25%.
  • the second detection light is monochromatic light, for example, red light, blue light or green light.
  • the second detection light is blue light or green light, which has a relatively good signal-to-noise ratio, and at the same time, the skin is more sensitive to blue light and green light, which is beneficial for living body recognition.
  • step S140 the receiving module 120 receives the second detection light reflected back by the subject and generates second spectral information of the subject based on the second detection light. That is, in the embodiment of the present application, the second detection light is only used to detect the spectral information of the subject, or in other words, during the second data collection process, only the spectral information of the subject is collected without Collect image information of the subject.
  • step S150 live body detection and object recognition are performed based on the first spectral information, the image information and the second spectral information. That is, the key to liveness detection and object recognition using the data obtained in multiple data collection processes is how to apply various data to improve the safety of liveness detection and object recognition.
  • the image information is first processed to generate an image of the subject, and then it is determined by comparing the image of the subject with a pre-stored reference image Whether the two match. Further, living body judgment is performed by using the first spectral information and the second spectral information. Correspondingly, when a living body is recognized and the images match, the living body detection is true, and the object recognition is true.
  • the process of performing living body detection based on the first spectral information and the second spectral information includes: first processing the first spectral information and the second spectral information to generate A first spectral response result and a second spectral response result; then, comparing the first spectral response result with a reference spectral response, and comparing the second spectral response result with the reference spectral response; finally, In response to the successful matching between the first spectral response result and the reference spectral response and the successful matching between the second spectral response result and the reference spectral response, it is determined that the subject is a living body.
  • live body detection and object recognition are performed, including: S210, the Process the first spectral information and the second spectral information to generate a first spectral response result and a second spectral response result; S220, process the image information to generate an image of the subject; S230, convert the obtained comparing the image of the subject with a pre-stored reference image; and, S240, in response to a successful match between the image of the subject and the reference image, based on the first spectral response result and/or the The second spectral response result is used to determine whether the subject is a living body. It should be understood that the present invention does not limit whether to determine whether the living body of the subject is determined first or to perform image matching determination first.
  • the first spectral information and the second spectral information may not be restored to spectral curves but only rely on spectral response data to perform living body judgment.
  • the process of processing the first spectral information and the second spectral information to generate the first spectral response result and the second spectral response result includes: converting the first spectral information into a one-dimensional spectrum a feature vector as the first spectral response result; and converting the second spectral information into a one-dimensional spectral feature vector as the second spectral response result.
  • the first spectral information and the second spectral information are arranged into a one-dimensional vector in a preset order, and the physical meaning of the characteristic value of each position in the one-dimensional spectral characteristic vector is reflected light Features in the light intensity data array received by the image sensor 122 after being modulated by the filter structure 123 .
  • the spectral curve can also be restored based on the first spectral information and/or the second spectral information, and then compared with the reference spectral curve to determine whether it is a living body.
  • the process of comparing the first spectral response result with a reference spectral response includes: calculating a first average value of values at various positions in the first spectral response result; and, in response to the first average value Within a preset range, it is determined that the match between the first spectral response result and the reference spectral response is successful, wherein the lower limit of the preset range is the reference average value of the reference spectral response minus the reference half of the standard deviation of the spectral response, and the upper limit of the preset range is the reference average plus half of the standard deviation of the reference spectral response.
  • Mean(M1) the average value of the feature values of each position in the first spectral response result, for example, denoted as Mean(M1)
  • Mean(M'1) the representation vector of the biometric feature in the recorded image
  • Mean(M'1) the representation vector of the biometric feature in the recorded image
  • the reference average value of the reference response spectrum only needs to be calculated and pre-stored at the time of the first entry, and further the reference average value will be calculated according to the actual identification situation of the user. Proactively update calibration.
  • the comparison between the second spectral response result and the reference spectral response may also be performed in a manner of comparing the first spectral response result with the reference spectral response , which will not be repeated here.
  • the comparison between the second spectral response result and the reference spectral response may also be performed in other ways, which is not limited by the present application.
  • the first spectral information can also realize living body identification to a certain extent, that is, there is no need to project the detection light twice or more, but only one projection can also realize living body and image information recognition.
  • living body detection and object recognition may also be performed based on the first spectral information, the image information, and the second spectral information in other manners.
  • the first spectral information and the second spectral information may be fused to perform living body detection. It should be understood that by fusing the first spectral information and the second spectral information, the spectral response information of the region to be measured under different light sources can be obtained, thereby improving the accuracy of living body detection.
  • the process of judging whether the subject is a living body includes: first fusing the first A spectral response result and the second spectral response result to obtain a fusion spectral response result; and, then, based on a comparison between the fusion spectral response result and a reference spectral response, judging whether the subject is a living body
  • the method of fusing the first spectral response result and the second spectral response result can be implemented as taking the union between the two as the fusion spectral response result, and using it as the living body detection result. basis. It should be understood that in other examples of the present application, other manners may also be used to fuse the first spectral information and the second spectral information, which is not limited by the present application.
  • step S120 the image information and spectral information of the subject are acquired based on the same beam of detection light (the first detection light). Therefore, in some other specific examples of this application, the image information may be used to assist the first spectral information to make the first spectral information more accurate, for example, the image information may be used to assist the first spectral information to achieve noise reduction.
  • the identification system when applied as a fingerprint identification system, there may be irregularities in the fingerprint, for example, valleys and ridges of the fingerprint.
  • irregularities in the fingerprint for example, valleys and ridges of the fingerprint.
  • the image information and spectral information of the same subject when the angle and position of the subject in the measured area change, it will cause the image information and spectral information of the same subject to change, for example, the change of valleys and ridges will cause the image Changes in the light intensity information received by the sensor 122 cause large changes in the current response value, thereby generating more noise.
  • the image information may be used to perform noise reduction on the first spectral information.
  • the implementation method is as follows: firstly adjust the spectral information of each pixel point in the region adjacent to each pixel point in the first spectral information to obtain the first adjusted spectral information; then, the first spectral information An adjustment spectral information is converted into a one-dimensional spectral feature vector as the first spectral response result; then, the second spectral information is converted into a one-dimensional spectral feature vector as the second spectral response result.
  • the modulation area 1221 is a pixel whose spectral information value is M real
  • the non-modulation area 1222 is 8 pixels, and its values are respectively is I 1 , I 2 ... I 8
  • the processed first spectral information Mi constitutes a spectral vector M1
  • features are extracted.
  • the average value removes the background noise of the spectral information, making the spectral information more accurate and not affected by the angle and displacement changes of the object to be measured.
  • the image information can also be optimized through the first spectral information to improve the accuracy of image matching, for example, information such as color and color temperature can be extracted from the first spectral information , and then use auxiliary information such as color and color temperature to assist the image information in imaging, so as to effectively improve the imaging quality of the imaged image and improve the recognition accuracy.
  • processing the image information to generate the image of the subject includes: optimizing the image of the subject with the first spectrum information.
  • the recognition system may also provide more than two projections to further improve the recognition accuracy.
  • three projections are mixed light, green light and blue light respectively.
  • mixed light is used to obtain image information and spectral information
  • green light and blue light are used to obtain spectral information, so as to improve the ability of living body judgment through more spectral information. precision and safety.
  • at least one is mixed light and at least one is monochromatic light.
  • the identification system based on the embodiment of the present application is clarified, wherein the light source of the identification system projects at least two different lights, wherein the detection light projected at least once is a mixed light, so that the receiving module 120
  • the image information and spectral information of the subject can be collected, wherein at least another projection of monochromatic light or corresponding mixed light allows the receiving module 120 to obtain spectral information, so that the image information obtained by multiple data collections can And spectral information to judge whether the image matches (fingerprint image) and whether it is a living body.
  • the structure of the receiving module 120 can be adjusted.
  • the spectrum chip 1000 of the receiving module 120 may not be configured with non-modulating The area 1222, that is, the receiving module 120 can only collect the spectral information of the subject, and instead, process the spectral information through an algorithm to restore the image of the subject. That is, in this embodiment, the spectral information of the receiving module 120 can only collect the spectral information of the subject and cannot directly collect the image information of the subject, but the image of the subject can be obtained through the spectral information generate.
  • the spectrum chip 1000 of the receiving module 120 can be configured as a non-modulation area 1222 and a modulation area 1221, and the modulation area 1221 is composed of color filters, such as R, G, One or more color filters in B. That is, a color filter is provided in the modulation area 1221 corresponding to the image sensor 122 , and the color filter allows light of a specific wavelength band to pass through, so that the spectrum chip 1000 can obtain corresponding spectral information in the modulation area 1221 .
  • color filters such as R, G, One or more color filters in B. That is, a color filter is provided in the modulation area 1221 corresponding to the image sensor 122 , and the color filter allows light of a specific wavelength band to pass through, so that the spectrum chip 1000 can obtain corresponding spectral information in the modulation area 1221 .
  • the receiving module 120 includes two chips, one of which is the spectrum chip 1000 and the other is the imaging chip 2000 . More specifically, in this embodiment, the receiving module 120 includes a spectrum chip 1000, an imaging chip 2000, and a beam splitter 3000, and the beam splitter 3000 is located on the path of the spectrum chip 1000 and the imaging chip 2000, wherein, after the incident light reaches the light splitter 3000, the first part of the light is deflected and the second part of the light is transmitted, and then received by the spectrum chip 1000 and the imaging chip 2000 respectively, so that the spectrum chip 1000 can obtain all Spectral information of the subject and image information of the subject obtained through the imaging chip 2000 .
  • the receiving module 120 also includes a homogenizing element 5000, which is arranged between the light splitting element 3000 and the spectrum chip 1000 to homogenize the light, and the spectrum chip 1000 obtains spectral information Carry out living body discrimination.
  • the receiving module 120 further includes a lens group 4000, the lens group 4000 is located between the imaging chip 2000 and the beam splitter 3000, the light is adjusted and received by the imaging chip 2000, which is beneficial to improve the imaging quality, For example more clarity.
  • the imaging chip 2000 is matched with the lens group 4000 , there is generally a requirement for focal length, and the size of the lens group 4000 is generally large, so it is suitable for horizontal placement, and the spectral chip 1000 with a homogenizing film generally has lower requirements for optics, and generally has no back focus Requirements, its size can be relatively smaller, so this part can be placed along the height direction, that is, after the incident light enters the light splitter 3000 along the height direction, the transmitted part enters the light homogenizing member 5000, and after homogenization reaches the spectrum chip 1000 ; while the turning part enters the lens group 4000 along the horizontal direction to be adjusted, and then is received by the imaging chip 2000 .
  • the identification system is implemented as a fingerprint identification system as an example, it should be understood that in other application examples of this application, the identification system can also be applied to other biometrics
  • the system for example, a live face detection system, etc., is not limited by this application.
  • the application terminal of the identification system is not limited by this application, and it can be applied to consumer electronic devices, wearable devices, and the like.

Abstract

一种识别系统及其工作方法，其中，所述识别系统通过多次投射检测光来获得被测目标的光谱信息和图像信息,并利用多次投射获得的光谱信息和图像信息来提高活体检测和对象识别的精度。多次投射的检测光可为不同类型的检测光。

Description

识别系统及其工作方法 技术领域
本申请涉及光谱应用领域，更为具体地说，涉及一种识别系统及其工作方法。
背景技术
各种类型的生物计量系统被越来越多地使用，以提供更高的安全性和/或增强的用户便利性。例如，指纹感测系统由于其尺寸小、性能高和用户接受度高已经被广泛地应用于各类终端设备中，例如消费者的智能手机中。目前市面上流通多种指纹感测系统，例如基于电容式指纹模组的感测系统、基于光学指纹模组的感测系统等，上述类型的指纹感测系统虽然可以实现解锁，但是在被应用于移动终端的指纹识别解锁后，不法分子可以通过窃取用户指纹制作出假指纹来破解用户的安全系统，这反而增加了移动终端指纹密码被识破的概率，对移动终端的信息安全造成了较大的威胁。
因此，需要对传统的生物计量系统进行安全加强，例如，通过活性检测来保护生物计量系统免受利用欺骗的身体部位的攻击，例如欺骗指纹。存在许多活体检测的解决方案，例如，寻找材料属性的基于硬件的方法、通过血氧定量法的脉冲检测、寻找所获得的指纹图像中的伪造的伪像欺骗并查看精细尺度的纹理的基于软件的方法。
但是，现有的具有活体检测功能的生物计量系统也存在一定的缺陷，其在一些场景下会失效，例如，在一种基于图像处理的活体检测方案中，其基于图像的软件方法明确地查看引入图像帧之间的差异的出汗效果，但并非所有手指都表现出所需的出汗量，尤其是在冬季条件下。
因此，期待一种优化的生物计量系统。
发明内容
为了解决上述技术问题，提出了本申请。本申请的实施例提供了一种识别系统及其工作方法，其中，所述识别系统通过多次投射检测光来获得被测目标的光谱信息和图像信息，并利用多次投射获得的光谱信息和图像信息来 提高活体检测和对象识别的精度。特别地，在本申请的一些示例中，多次投射的检测光为不同类型的检测光。
根据本申请的一方面，提供了一种识别系统的工作方法，其包括：
投射第一检测光至被摄对象；
接收被所述被摄对象反射回来的所述第一检测光并基于所述第一检测光生成所述被摄目标的第一光谱信息和图像信息；
投射第二检测光至所述被摄对象；
接收被所述被摄对象反射回来的所述第二检测光并基于所述第二检测光生成所述被摄目标的第二光谱信息；以及
基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别。
在根据本申请的识别系统的工作方法中，所述第一检测光与所述第二检测光为不同类型的光信号。
在根据本申请的识别系统的工作方法中，所述第一检测光为包含至少两个不同波段的单色光的混合光，所述第二检测光为单色光。
在根据本申请的识别系统的工作方法中，所述第一检测光为白光，所述第二检测光为蓝光。
在根据本申请的识别系统的工作方法中，接收被所述被摄对象反射回来的所述第一检测光并基于所述第一检测光生成所述被摄目标的第一光谱信息和图像信息，包括：通过接收模组的滤光结构对所述第一检测光中的至少一部分进行调制以生成调制光信号；以及，通过所述接收模组的图像传感器的调制区域接收所述调制光信号以生成所述第一光谱信息以及通过所述图像传感器的非调制区域接收所述第一检测光中未被所述滤光结构所调制的部分以生成所述图像信息。
在根据本申请的识别系统的工作方法中，所述图像传感器包括用于生成所述图像信息的非调制区域和用于生成所述第一光谱信息的调制区域，所述滤光结构对应于所述调制区域。
在根据本申请的识别系统的工作方法中，所述调制区域的面积占所述图像传感器的有效区域的面积的比例为12％-25％。
在根据本申请的识别系统的工作方法中，基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别，包括：对所述第 一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果；对所述图像信息进行处理以生成所述被摄对象的图像；将所述被摄对象的图像与预存的基准图像进行比较；以及，响应于所述被摄对象的图像与所述基准图像之间的匹配成功，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体。
在根据本申请的识别系统的工作方法中，对所述图像信息进行处理以生成所述被摄对象的图像，包括：以所述第一光谱信息对所述被摄对象的图像进行优化。
在根据本申请的识别系统的工作方法中，对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果，包括：将所述第一光谱信息转化为一维的光谱特征向量作为所述第一光谱响应结果；以及，将所述第二光谱信息转化为一维的光谱特征向量作为所述第二光谱响应结果。
在根据本申请的识别系统的工作方法中，对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果，包括：以与所述第一光谱信息中各个像素点相邻区域内的像素点的图像信息来调整该像素点的光谱信息以获得第一调整光谱信息；将所述第一调整光谱信息转化为一维的光谱特征向量作为所述第一光谱响应结果；以及，将所述第二光谱信息转化为一维的光谱特征向量作为所述第二光谱响应结果。
在根据本申请的识别系统的工作方法中，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体，包括：将所述第一光谱响应结果与基准光谱响应进行比较；将所述第二光谱响应结果与所述基准光谱响应进行比较；以及，响应于所述第一光谱响应结果与基准光谱响应之间的匹配成功以及所述第二光谱响应结果与所述基准光谱响应之间的匹配成功，确定所述被摄对象为活体。
在根据本申请的识别系统的工作方法中，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体，包括：融合所述第一光谱响应结果和所述第二光谱响应结果以获得融合光谱响应结果；以及，基于所述融合光谱响应结果与基准光谱响应之间的比较，判断所述被摄对象是否为活体。
在根据本申请的识别系统的工作方法中，将所述第一光谱响应结果与基准光谱响应进行比较，包括：计算所述第一光谱响应结果中各个位置的值的第一平均值；以及，响应于所述第一平均值在预设范围内，确定所述第一光谱响应结果与所述基准光谱响应之间的匹配成功，其中，所述预设范围的下限为所述基准光谱响应的基准平均值减去所述基准光谱响应的标准差的一半，所述预设范围的上限为所述基准平均值加上所述基准光谱响应的标准差的一半。
根据本申请的另一方面，还提供了一种识别系统，其中，所述识别系统能够执行如上所述的工作方法。
与现有技术相比，根据本申请的识别系统及其工作方法，其中，所述识别系统通过多次投射检测光来获得被测目标的光谱信息和图像信息，并利用多次投射获得的光谱信息和图像信息来提高活体检测和对象识别的精度。特别地，在本申请的一些示例中，多次投射的检测光为不同类型的检测光。
附图说明
通过阅读下文优选的具体实施方式中的详细描述，本申请各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。说明书附图仅用于示出优选实施方式的目的，而并不认为是对本申请的限制。显而易见地，下面描述的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。而且在整个附图中，用相同的附图标记表示相同的部件。
图1图示了根据本申请实施例的识别系统的框图。
图2图示了根据本申请实施例的所述识别系统的接收模组的框图。
图3图示了根据本申请实施例的所述接收模组中图像传感器的框图。
图4图示了根据本申请实施例的所述识别系统的工作方法的流程图。
图5图示了根据本申请实施例的所述识别系统的工作方法中基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别的流程图。
图6图示了根据本申请实施例的所述识别系统中图像传感器的示意图。
图7图示了根据本申请实施例额所述识别系统中接收模组的系统架构示意图。
具体实施方式
下面，将参考附图详细地描述根据本申请的示例实施例。显然，所描述的实施例仅仅是本申请的一部分实施例，而不是本申请的全部实施例，应理解，本申请不受这里描述的示例实施例的限制。
申请概述
如前所述，现有的具有活体检测功能的生物计量系统也存在一定的缺陷，其在一些场景下会失效，例如，在一种基于图像处理的活体检测方案中，其基于图像的软件方法明确地查看引入图像帧之间的差异的出汗效果，但并非所有手指都表现出所需的出汗量，尤其是在冬季条件下。因此，期待一种优化的生物计量系统。
由于人皮肤中存在毛细血管(血液)、汗孔等生理特征，相对指纹纹路来讲难以被伪造。相应地，由于人手指纹存在生理特征，其会导致皮肤对不同波段的光谱吸收/反射的程度不同，因此，可根据由皮肤反射后的光谱信息来进行活体判断，从而实现对指纹的活体检测。具体地，本申请发明人通过对真人手指和指模材料进行反射光谱测试可知，在300nm-1100nm波长下，真人手指反射光谱和指模材料的反射光谱存在差异，通过该实验表明，可以通过接收到的反射光谱进行活体检测。
基于此，本申请提出了一种识别系统，其包括光源组件110和接收模组120(例如，所述接收模组120被实施为指纹模组组件)，其中，所述光源组件110适于投射检测光至被摄对象(例如，人手的手指)，从被摄对象反射回来的检测光被所述接收模组120所接收并生成被摄对象的光谱信息和/或图像信息，从而基于光谱信息和图像信息来进行活体检测和对象识别。特别地，在本申请的一些示例中，为了加强活体检测和对象识别的安全性和准确性，所述识别系统通过多次投射检测光来获得被测目标的光谱信息和图像信息，并利用多次投射获得的光谱信息和图像信息来提高活体检测和对象识别的精度。优选地，多次投射的检测光为不同类型的检测光，以此来提高活体检测的安全级别。
在介绍了本申请的基本原理之后，下面将参考附图来具体介绍本申请的各种非限制性实施例。
示意性识别系统及其工作方法
如图1和图2所示，根据本申请实施例的识别系统被阐明，其中，所述识别系统属于生物计量系统，其通过采集被摄对象的光谱信息和图像信息来进行活体检测和对象识别。具体地，所述识别系统包括光源组件110和接收模组120，其中，所述光源组件110适于投射检测光至被摄对象，从被摄对象反射回来的检测光被所述接收模组120所接收并生成被摄对象的光谱信息和/或图像信息，从而基于光谱信息和图像信息来进行活体检测和对象识别。
例如，当所述识别系统被实施为指纹识别系统时，所述识别系统的光源组件110用以投射检测光至被摄对象(正常使用下，被摄对象为真人手指，在被入侵的场景下，被摄对象可能是手指模型、手指图案等)，所述识别系统的接收模组120用以接收自被摄对象反射回来的检测光以获得所述被摄对象的光谱信息和图像信息，即，在正常使用下，所述光谱信息为真人手指的光谱信息(其表征真人手指的生理特征)，所述图像信息为真人手指的图像。进而，所述识别系统能够基于所述图像信息和所述光谱信息来进行活体检测和指纹识别。
特别地，在本申请实施例中，所述识别系统基于计算光谱技术来获取被摄对象的光谱信息，也就是，在本申请实施例中，所述接收模组120基于计算光谱技术来获得被摄对象的光谱信息。
为了便于理解，对计算光谱技术从结构和原理进行介绍，如图2所示，在本申请实施例中，被实施为计算光谱装置的所述接收模组120包括线路板121、电连接于所述线路板121的图像传感器122，以及，被保持于图像传感器122的感光路径上的滤光结构123。所述线路板121可以为软板(FPC)、硬板(PCB)或软硬结合板(F-PCB)、陶瓷基板等。
所述图像传感器122和滤光结构123构成光谱芯片1000，其中，所述滤光结构123为频域或者波长域上的宽带滤光结构123。所述滤光结构123可以是超表面、光子晶体、纳米柱、多层膜、染料、量子点、MEMS(微机电系统)、FP etalon(FP标准具)、cavity layer(谐振腔层)、waveguide layer(波导层)、衍射元件等具有滤光特性的结构或者材料。例如，在本申请实施例中，所述滤光结构123可以是中国专利CN201921223201.2中的光调制层。进一步，在一些实施例中，所述接收模组120还可以包括光学系统124，例 如，可以是透镜组件、匀光组件等光学系统124，所述光学系统124位于所述滤光结构123的前端，来自外界的检测光经由所述光学系统124调整后进入并被所述滤光结构123所调制，再被所述图像传感器122接收，获得光谱响应。
所述图像传感器122可以是CMOS图像传感器(CIS)、CCD、阵列光探测器等。另外，所述接收模组120还可以包括数据处理单元125，所述数据处理单元125可以是MCU、CPU、GPU、FPGA、NPU、ASIC等处理单元，其可以将图像传感器122生成的数据导出到外部进行处理。例如，所述图像传感器122测得光谱响应后，传入所述数据处理单元125进行光谱恢复计算。
具体地，光谱恢复计算的过程具体描述如下：
将入射光在不同波长λ下的强度信号记为x(λ)，所述滤光结构123的透射谱曲线记为T(λ)，所述滤光结构123具有m组的结构单元，每一组结构单元的透射谱互不相同，整体来讲，滤光结构123可记为Ti(λ)(i＝1,2,3,…,m)。每一组结构单元下方都有相应的物理像素(这里，所述物理像素指的是所述图像传感器122的一个像素点)，探测经过滤光结构123调制的光强bi。在本申请的特定实施例中，以一个物理像素，即一个物理像素对应一组结构单元，但是不限定于此，在其它实施例中，也可以是多个物理像素为一组对应于一组结构单元。因此，在根据本申请实施例的接收模组120中，至少二组结构单元构成一个“光谱像素”。需要注意的是，所述滤光结构123的有效的透射谱(用以光谱恢复的透射谱，叫做有效的透射谱)Ti(λ)数量与结构单元数量可以不一致，所述滤光结构123的透射谱根据识别或恢复的需求人为的按照一定规则去设置、测试、或计算获得(例如上述每个结构单元通过测试出来的透射谱就为有效的透射谱)，因此所述滤光结构123的有效透射谱的数量可以比结构单元数量少，甚至也可能比结构单元数量多；该变形实施例中，某一个所述透射谱曲线并不一定是一组结构单元所决定。进一步，本发明可以用至少一个光谱像素去还原图像。
入射光的频谱分布和图像传感器122的测量值之间的关系可以由下式表示：
bi＝∫x(λ)*Ti(λ)*R(λ)dλ
再进行离散化，得到
bi＝Σ(x(λ)*Ti(λ)*R(λ))
其中R(λ)为图像传感器的响应，记为：
Ai(λ)＝Ti(λ)*R(λ)，
则上式可以扩展为矩阵形式：
Figure PCTCN2022139150-appb-000001
其中，bi(i＝1,2,3,…,m)是待测光透过滤光结构后图像传感器的响应，分别对应m个结构单元对应的图像传感器的光强测量值，当一个物理像素对应一个结构单元时，可以理解为m个‘物理像素’对应的光强测量值，其是一个长度为m的向量。A是系统对于不同波长的光响应，由滤光结构透射率和图像传感器的量子效率两个因素决定。A是矩阵，每一个行向量对应一组结构单元对不同波长入射光的响应，这里，对入射光进行离散、均匀的采样，共有n个采样点。A的列数与入射光的采样点数相同。这里，x(λ)即是入射光在不同波长λ的光强，也就是待测量的入射光光谱。
在一些实施例中，与上述实施例不同之处在于，所述滤光结构123可直接形成于所述图像传感器122上表面，例如量子点、纳米线等，其直接在图像传感器122的感光区域形成滤光结构123或材料(纳米线、量子点等)，以滤光结构123为例，此时，可以理解为所述图像传感器122的原材料在加工形成所述图像传感器122时，在原材料上表面加工形成滤光结构123，所述透射谱和所述图像传感器122的响应是一体的，即可以理解为所述探测器的响应和所述透射谱为同一曲线，此时入射光的频谱分布和图像传感器122的光强测量值之间的关系可以由下式表示：
bi＝Σ(x(λ)*Ri(λ))
即本实施例中，透射谱Ai(λ)＝Ri(λ)
此时，入射光的频谱分布和图像传感器122的光强测量值之间的关系可以由下式表示：
bi＝∫x(λ)*Ti(λ)*Ri(λ)dλ
再进行离散化，得到
bi＝Σ(x(λ)*Ti(λ)*Ri(λ))
本实施例中，Ai(λ)＝Ti(λ)*Ri(λ)
以上以一个物理像素对应一组结构单元为例，讲述了如何利用m组物理像素以及其对应的m组结构单元(调制层上相同结构界定为结构单元)恢复出一个光谱信息，又称为“光谱像素”。
值得注意的是，在本申请实施例中，也可以是多个物理像素对应一组结构单元。可以进一步定义，一组结构单元和对应的至少一物理像素构成一单元像素，原则上，至少一单元像素构成一所述光谱像素。在上述实现方式的基础上，将光谱像素进行阵列化处理，则可以实现快照式的光谱成像设备，也就是，所述接收装置即可用以光谱恢复又可以用以光谱成像。
特别地，在本申请实施例中，为了加强活体检测和对象识别的安全性和准确性，所述识别系统通过多次投射检测光来获得被测目标的光谱信息和图像信息，并利用多次投射获得的光谱信息和图像信息来提高活体检测和对象识别的精度。优选地，多次投射的检测光为不同类型的检测光，以此来提高活体检测的安全级别。
相应地，在本申请实施例中，所述光源组件110包括控制器111和与所述控制器111电连接的至少一光源123，其中，通过所述控制器111可以控制所述光源123投射出不同类型的检测光，例如，混合光、单色光，以使得所述接收模组120可以采集被摄目标的光谱信息和/或图像信息。在一个具体的示例中，所述光源123可以被实施为屏幕光源，其包括至少一红光发光点、至少一绿光发光点和至少一蓝光发光点，其中，在一种可能的工作模式下，所述光源123可以由至少两个发光点进行混合投射，例如红光发光点和绿光发光点同时工作，以投射混合光；在另一种可能的工作模式下，所述光源123可以由一个发光点进行工作以投射单色光。
值得一提的是，对于所述光源123的选择可基于应用场景的要求做出调整，例如，在所述识别系统为指纹识别系统时，使用的光源应尽可能地使得真皮肤和伪造皮肤对其投射的检测光的吸收和反射性能不同，从而提高识别精度。
图4图示了根据本申请实施例的所述识别系统的工作方法的流程图。如图4所示，根据本申请实施例的所述识别系统的工作方法，包括步骤：S110，投射第一检测光至被摄对象；S120，接收被所述被摄对象反射回来的所述第一检测光并基于所述第一检测光生成所述被摄目标的第一光谱信息和图像信息；S130，投射第二检测光至所述被摄对象；S140，接收被所述被摄对象 反射回来的所述第二检测光并基于所述第二检测光生成所述被摄目标的第二光谱信息；以及，S150，基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别。值得一提的是，本发明投射何种检测光并没有顺序限定，可以根据需求和设定自由实现投射不同的光源。
优选地，在本申请实施例中，所述第一检测光与所述第二检测光为不同类型的光信号，从而所述第一光谱信息和所述第二光谱信息之间具有差异，通过这样的方式，来提高活体检测的准确度和安全性。当然，在本申请的一些示例中，所述第一检测光和所述第二检测光也可以是相同类型的检测光，通过两次的光谱测量和比较，在一定程度上也可能提升活体检测的准确度和安全性，例如，可以简单地理解为有二次检测或者二次确认的过程。
相应地，在一个具体的示例中，在步骤S110中，所述第一检测光为包含至少两个不同波段的单色光的混合光，也就是，所述光源组件110投射混合光于被摄对象。更具体地，所述混合光可以是屏幕光源发出的混杂有红光、绿光和蓝光的白色混合光。
在步骤S120中，所述接收模组120接收被所述被摄对象反射回来的所述第一检测光并基于所述第一检测光生成所述被摄目标的第一光谱信息和图像信息。其具体过程中，首先通过所述接收模组120的滤光结构123对所述第一检测光进行调制以生成调制光信号，也就是，在所述接收模组120的滤光结构123对所述第一检测光中的至少一部分进行调制以生成所述调制光信号，而所述第一检测光中剩余部分的检测光没有被调制。然后，在所述图像传感器122的调制区域1221接收所述调制光信息以生成所述第一光谱信息(此过程可参见对于计算光谱技术的描述部分，在此不再赘述)以及，在所述图像传感器122的非调制区域1222接收所述第一检测光中未被调制的部分以生成所述图像信息。
也就是说，在本申请实施例中，如图3所示，所述图像传感器122包括用于生成所述图像信息的非调制区域1222和用于生成所述第一光谱信息的调制区域1221，所述滤光结构123对应于所述调制区域1221。也就是说，所述接收模组120不仅可以采集被摄对象的光谱信息也可以采集被摄目标的图像信息。并且，所述调制区域1221的面积占所述图像传感器122的有效区域的面积的比例为10％-50％，优选地，所述调制区域1221的面积占所述图像传感器122的有效区域的面积的比例为12％-25％，可以理解的是所述调 制区域1221面积占比过大的话会很有可能会影响图像信息的完备性，但是面积占比过小则获得的光谱信息过少，导致活体实现难度增加。
进一步地，在一个具体的示例中，在步骤S130中，所述第二检测光为单色光，例如，红光、蓝光或者绿光。优选地，所述第二检测光为蓝光或者绿光，相对来讲有较好的信噪比，同时皮肤对蓝光和绿光较为敏感，有利于活体识别。
在步骤S140中，所述接收模组120接收被所述被摄对象反射回来的所述第二检测光并基于所述第二检测光生成所述被摄目标的第二光谱信息。也就是，在本申请实施例中，所述第二检测光仅用于探测被摄对象的光谱信息，或者说，在第二次数据采集的过程中，仅采集被摄对象的光谱信息而没有采集被摄对象的图像信息。
在步骤S150中，基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别。也就是，利用多次数据采集过程所获得的数据来进行活体检测和对象识别，其关键在于各项数据如何应用来提高活体检测和对象识别的安全性。
具体地，在本申请一个具体的示例中，首先对所述图像信息进行处理以生成所述被摄对象的图像，然后，通过将所述被摄对象的图像与预存的基准图像进行比较来判断两者是否匹配。进一步地，利用所述第一光谱信息和所述第二光谱进行来进行活体判断。相应地，当识别为活体且图像匹配时，则活体检测为真，对象识别为真。
由于不同的材料、以及真人皮肤对不同光的吸收、反射情况不同，本方案中两次用不同的光源去照射，可以提高仿造假皮肤的难度，即当只用一种检测光的时候，仿造时只需要找到一种材料在该光源下光谱响应与真人皮肤一致即可，这也可以理解为使得识别的精度得以提高。
具体地，在该具体示例中，基于所述第一光谱信息和所述第二光谱信息进行活体检测的过程，包括：首先对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果；然后，将所述第一光谱响应结果与基准光谱响应进行比较，以及，将所述第二光谱响应结果与所述基准光谱响应进行比较；最终，响应于所述第一光谱响应结果与基准光谱响应之间的匹配成功以及所述第二光谱响应结果与所述基准光谱响应之间的匹配成功，确定所述被摄对象为活体。
也就是，在本申请实施例中，如图5所示，基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别，包括：S210，对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果；S220，对所述图像信息进行处理以生成所述被摄对象的图像；S230，将所述被摄对象的图像与预存的基准图像进行比较；以及，S240，响应于所述被摄对象的图像与所述基准图像之间的匹配成功，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体。需要理解的，本发明并不限定先对被摄对象活体判断还是先进行图像匹配判断。
相应地，在该具体示例中，所述第一光谱信息和所述第二光谱信息可以不被还原为光谱曲线而仅依靠于光谱响应数据来进行活体判断。具体地，对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果的过程，包括：将所述第一光谱信息转化为一维的光谱特征向量作为所述第一光谱响应结果；以及，将所述第二光谱信息转化为一维的光谱特征向量作为所述第二光谱响应结果。也就是说，将所述第一光谱信息和所述第二光谱信息按照预设的顺序排列为一维向量，其所述一维的光谱特征向量中各个位置的特征值的物理意义为反射光经所述滤光结构123调制后被所述图像传感器122所接收的光强数据阵列中的特征。需要注意的是，一定条件下，也可以基于第一光谱信息和/或第二光谱信息进行光谱曲线的恢复，再去跟基准光谱曲线比较，判断是否为活体。
然后，将所述第一光谱响应结果与基准光谱响应进行比较的过程，包括：计算所述第一光谱响应结果中各个位置的值的第一平均值；以及，响应于所述第一平均值在预设范围内，确定所述第一光谱响应结果与所述基准光谱响应之间的匹配成功，其中，所述预设范围的下限为所述基准光谱响应的基准平均值减去所述基准光谱响应的标准差的一半，所述预设范围的上限为所述基准平均值加上所述基准光谱响应的标准差的一半。也就是，首先计算所述第一光谱响应结果中各个位置的特征值的平均值，例如，记为Mean(M1)，以及，计算录入的图像中生物特征的表示向量(即，所述基准光谱响应)的平均值，例如，记为Mean(M'1)。然后，判断所述Mean(M1)的值是否在{Mean(M'1)-std(M'1)/2，Mean(M'1)+std(M'1)/2}的范围内，如果在此范围内，则匹配成功，若不在此范围内，则匹配失败。
值得一提的是，在本申请实施例中，所述基准响应光谱的基准平均值只需要在第一次录入时进行计算并预存即可，更进一步该基准平均值会根据用户实际识别情况进行主动更新校准。当然，在本申请其他实施例中，也可以预存所述基准响应光谱中各个位置的基准响应值，并计算各个基准响应值之间的均值来生成所述基准平均值，对此，并不为本申请所局限。
应可以理解，在该具体示例中，所述第二光谱响应结果与所述基准光谱响应之间的比较也可以以所述第一光谱响应结果与所述基准光谱响应之间的比较方式来进行，在此不再赘述。当然，为了提高活体检测的准确率和安全性，所述第二光谱响应结果与所述基准光谱响应之间的比较还可以以其他方式来进行，对此，并不为本申请所局限。可以理解，在本申请一些实施例中，所述第一光谱信息一定程度上也可以实现活体识别，即无需两次或多次投射检测光，仅投射一次也可以实现活体和图像信息识别。
值得一提的是，在本申请其他实施例中，还可以以其他方式来基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别。例如，在本申情另外一个具体的示例中，可以将所述第一光谱信息和所述第二光谱信息进行融合以进行活体检测。应可以理解，通过融合所述第一光谱信息和所述第二光谱信息可以获取在光源不同的情况下，待测区域的光谱响应信息情况，从而提升提高活体检测的准确率。
也就是，在本申请另外的示例中，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体的过程，包括：首先融合所述第一光谱响应结果和所述第二光谱响应结果以获得融合光谱响应结果；以及，然后，基于所述融合光谱响应结果与基准光谱响应之间的比较，判断所述被摄对象是否为活体
在具体实施例中，融合所述第一光谱响应结果和所述第二光谱响应结果的方式可被实施为取两者之间的并集作为所述融合光谱响应结果，并以此作为活体检测的依据。应可以理解，在本申请其他示例中，还可以采用其他方式来融合所述第一光谱信息和所述第二光谱信息，对此，并不为本申请所局限。
应注意到，在本申请实施例中，在步骤S120中，基于同一束检测光(所述第一检测光)获取被摄对象的图像信息和光谱信息，因此，在本申请另外一些具体的示例中，可以用所述图像信息去辅助所述第一光谱信息，使得所 述第一光谱信息更加准确，例如用所述图像信息帮助所述第一光谱信息去实现降噪等。
应可以理解，当所述识别系统被应用为指纹识别系统时，由于指纹会存在不平整之处，例如，指纹的谷和脊。相应地，在测试时，当被摄对象在被测区域的角度和位置发生变化时，会导致同一被摄对象的图像信息和光谱信息发生变化，例如，谷和脊的变化会导致所述图像传感器122接收到的光强信息发生变化使得电流响应值发生较大变化，从而产生较多噪声。相应地，在该应用示例中，可利用所述图像信息来对所述第一光谱信息进行降噪。其实施方式为：首先以与所述第一光谱信息中各个像素点相邻区域内的像素点的图像信息来调整该像素点的光谱信息以获得第一调整光谱信息；然后，将所述第一调整光谱信息转化为一维的光谱特征向量作为所述第一光谱响应结果；接着，将所述第二光谱信息转化为一维的光谱特征向量作为所述第二光谱响应结果。
在一个具体的示例中，以9个像素为一组为例，如图6所示，调制区域1221为一个像素其光谱信息值为M 实，而非调制区域1222为8个像素，其值分别为I 1、I 2……I 8，则处理所述第一光谱信息时，可以令Mi＝M 实-(I 1+I 2+……+I 8)/8(i为1至n的整数)，经过处理的第一光谱信息Mi构成光谱向量M1，再进行提取特征。应可以理解，以9个像素为一组只是举例，可以为任意个像素为一组，在对所述第一光谱信息和所述图像信息分别取平均再取差值，利用所述图像信息的均值去除光谱信息的底噪，使得光谱信息更准确，不受待测物的角度、位移变动影响。
进一步地，在本申请其他示例中，还可以通过所述第一光谱信息对所述图像信息进行优化以提高图像匹配的精度，例如，从所述第一光谱信息中提取出色彩、色温等信息，然后，利用色彩、色温等辅助信息来协助所述图像信息进行成像，以有效地提高成像后的图像的成像质量，提升识别准确率。也就是，在本申请的一些示例中，对所述图像信息进行处理以生成所述被摄对象的图像，包括：以所述第一光谱信息对所述被摄对象的图像进行优化。
值得一提的是，在本申请实施例中，所述识别系统也可以提供两次以上的投射，以进一步地提高识别精度。例如，投射三次，分别为混合光、绿光和蓝光，其中，混合光用于获取图像信息和光谱信息，而绿光和蓝光则为了 获取光谱信息，以通过更多光谱信息来提高活体判断的精度和安全性。多次投射中，至少一次为混合光，至少一次为单色光。
综上，基于本申请实施例的所述识别系统被阐明，其中，所述识别系统的光源投射至少二次不同的光，其中，至少一次投射的检测光为混合光，以使得接收模组120可以采集被摄对象的图像信息和光谱信息，其中，至少另一次投射单色光或对应的混合光，使得所述接收模组120可以获得光谱信息，从而根据多次数据采集所获得的图像信息和光谱信息去判断图像是否匹配(指纹图像)和是否为活体。
值得一提的是，在本申请一些实施例中，所述接收模组120的结构可做一些调整，例如，在一个具体的示例中，所述接收模组120的光谱芯片1000可不配置非调制区域1222，也就是，所述接收模组120只能够采集被摄目标的光谱信息，而取而代之的是，通过算法来处理所述光谱信息来还原所述被摄对象的图像。也就是，在该实施例中，所述接收模组120的光谱信息只能够采集被摄目标的光谱信息不能够直接采集被摄对象的图像信息，而被摄对象的图像可通过所述光谱信息生成。
在另一个实施例中，所述接收模组120的光谱芯片1000可以配置为非调制区域1222和调制区域1221构成，所述调制区域1221由滤色器(color filter)构成，例如R、G、B中一种或多种滤色器构成。即在图像传感器122对应的调制区域1221设置有滤色器，该滤色器允许特定波段的光透过，从而使得光谱芯片1000可以在该调制区域1221可以获得对应的光谱信息。
在本申请另外一个具体的示例中，如图7所示，所述接收模组120包括两个芯片，其中一个为所述光谱芯片1000而另外一个为成像芯片2000。更具体地，在该实施例中，所述所述接收模组120包括光谱芯片1000、成像芯片2000和分光件3000，所述分光件3000位于所述光谱芯片1000和成像芯片2000的路径上，其中，入射光在抵达所述分光件3000后使得第一部分光转折和第二部分光透过，再分别由所述光谱芯片1000和所述成像芯片2000接收，以通过所述光谱芯片1000获得所述被摄对象的光谱信息和通过所述成像芯片2000获得所述被摄对象的图像信息。优选地，所述接收模组120还包括匀光件5000，所述匀光件5000设置于所述分光件3000和光谱芯片1000之间，对光进行匀化，再有光谱芯片1000获得光谱信息进行活体判别。
需要说明的是，由于所述被摄对象往往是不平整的，例如指纹有谷和脊，当测试时对应的区域的变动会导致不同区域产生的光谱响应不同，使得活体判断难度增加，而使用匀光后，即使测试时区域有所变动，其整体的光谱信息是不变的。例如指纹活体判别过程中，测试者在手指放置产生一定角度的偏转，光谱芯片1000对应的指纹谷和脊也会变动，会导致到达光谱芯片1000的光谱信息发生变动，此时需要进行额外处理才能实现准确判断，而匀光后，由于光源不动，手指的谷和脊整体是不变的，因此偏转不会导致光谱信息发生较大变动，从而可以较为简单、高效实现活体判断。优选地，所述接收模组120还包括透镜组4000，所述透镜组4000位于所述成像芯片2000和分光件3000之间，对光进行调整后被成像芯片2000接收，有利于提高成像质量，例如更加清晰。
鉴于在实际应用中会对尺寸有要求，例如在手机、穿戴设备之类的，需要将某一方向上的尺寸进行限制，以高度方向为例，由于所述成像芯片2000搭配透镜组4000的情况下，一般会有焦距的要求，再加上透镜组4000的尺寸一般会较大，因此其适合水平放置，而所述光谱芯片1000搭配匀光片一般对光学的要求较低，一般无后焦距的需求，其尺寸相对来讲也可以做的更小，故该部分可以沿着高度方向放置，即入射光沿着高度方向进入分光件3000后，透射部分进入所述匀光件5000，匀化后到达所述光谱芯片1000；而转折部分则沿着水平方向进入透镜组4000被调整，再由所述成像芯片2000接收。
应可以理解，虽然在上述实施例中，以所述识别系统被实施为指纹识别系统为示例，但应可以理解，在本申请其他应用示例中，所述识别系统还可以被应用于其他生物计量系统，例如，活体人脸检测系统等，对此，并不为本申请所局限。并且，所述识别系统的应用终端也并不为本申请所局限，其可以被应用于消费电子设备、可穿戴设备等。
以上结合具体实施例描述了本申请的基本原理，但是，需要指出的是，在本申请中提及的优点、优势、效果等仅是示例而非限制，不能认为这些优点、优势、效果等是本申请的各个实施例必须具备的。另外，上述公开的具体细节仅是为了示例的作用和便于理解的作用，而非限制，上述细节并不限制本申请为必须采用上述具体的细节来实现。

Claims (16)

  1. 一种识别系统的工作方法，其特征在于，包括：
    投射第一检测光至被摄对象；
    接收被所述被摄对象反射回来的所述第一检测光并基于所述第一检测光生成所述被摄目标的第一光谱信息和图像信息；
    投射第二检测光至所述被摄对象；
    接收被所述被摄对象反射回来的所述第二检测光并基于所述第二检测光生成所述被摄目标的第二光谱信息；以及
    基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别。
  2. 根据权利要求1所述的识别系统的工作方法，其中，所述第一检测光与所述第二检测光为不同类型的光信号。
  3. 根据权利要求2所述的识别系统的工作方法，其中，所述第一检测光为包含至少两个不同波段的单色光的混合光，所述第二检测光为单色光。
  4. 根据权利要求3所述的识别系统的工作方法，其中，所述第一检测光为白光，所述第二检测光为蓝光或绿光。
  5. 根据权利要求1所述的识别系统的工作方法，其中，接收被所述被摄对象反射回来的所述第一检测光并基于所述第一检测光生成所述被摄目标的第一光谱信息和图像信息，包括：
    通过接收模组的滤光结构对所述第一检测光中的至少一部分进行调制以生成调制光信号；以及
    通过所述接收模组的图像传感器的调制区域接收所述调制光信号以生成所述第一光谱信息以及通过所述图像传感器的非调制区域接收所述第一检测光中未被所述滤光结构所调制的部分以生成所述图像信息。
  6. 根据权利要求5所述的识别系统的工作方法，其中，所述滤光结构对应于所述调制区域。
  7. 根据权利要求6所述的识别系统的工作方法，其中，所述调制区域的面积占所述图像传感器的有效区域的面积的比例为12％-25％。
  8. 根据权利要求1所述的识别系统的工作方法，其中，基于所述第一光谱信息、所述图像信息和所述第二光谱信息，进行活体检测和对象识别，包括：
    对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果；
    对所述图像信息进行处理以生成所述被摄对象的图像；
    将所述被摄对象的图像与预存的基准图像进行比较；
    响应于所述被摄对象的图像与所述基准图像之间的匹配成功，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体。
  9. 根据权利要求8所述的识别系统的工作方法，其中，对所述图像信息进行处理以生成所述被摄对象的图像，包括：
    以所述第一光谱信息对所述被摄对象的图像进行优化。
  10. 根据权利要求8所述的识别系统的工作方法，其中，对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果，包括：
    将所述第一光谱信息转化为一维的光谱特征向量作为所述第一光谱响应结果；以及
    将所述第二光谱信息转化为一维的光谱特征向量作为所述第二光谱响应结果。
  11. 根据权利要求8所述的识别系统的工作方法，其中，对所述第一光谱信息和所述第二光谱信息进行处理以生成第一光谱响应结果和第二光谱响应结果，包括：
    以与所述第一光谱信息中各个像素点相邻区域内的像素点的图像信息来调整该像素点的光谱信息以获得第一调整光谱信息；
    将所述第一调整光谱信息转化为一维的光谱特征向量作为所述第一光谱响应结果；以及
    将所述第二光谱信息转化为一维的光谱特征向量作为所述第二光谱响应结果。
  12. 根据权利要求10或11所述的识别系统的工作方法，其中，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体，包括：
    将所述第一光谱响应结果与基准光谱响应进行比较；
    将所述第二光谱响应结果与所述基准光谱响应进行比较；以及
    响应于所述第一光谱响应结果与基准光谱响应之间的匹配成功以及所述第二光谱响应结果与所述基准光谱响应之间的匹配成功，确定所述被摄对象为活体。
  13. 根据权利要求10或11所述的识别系统的工作方法，其中，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体，包括：
    融合所述第一光谱响应结果和所述第二光谱响应结果以获得融合光谱响应结果；以及
    基于所述融合光谱响应结果与基准光谱响应之间的比较，判断所述被摄对象是否为活体。
  14. 根据权利要求8所述的识别系统的工作方法，其中，响应于所述被摄对象的图像与所述基准图像之间的匹配成功，基于所述第一光谱响应结果和/或所述第二光谱响应结果，判断所述被摄对象是否为活体，包括：
    计算所述第一光谱响应结果中各个位置的值的第一平均值；以及
    响应于所述第一平均值在预设范围内，确定所述第一光谱响应结果与所述基准光谱响应之间的匹配成功，其中，所述预设范围的下限为所述基准光谱响应的基准平均值减去所述基准光谱响应的标准差的一半，所述预设范围的上限为所述基准平均值加上所述基准光谱响应的标准差的一半。
  15. 根据权利要求12所述的识别系统的工作方法，其中，将所述第一光谱响应结果与基准光谱响应进行比较，包括：
    计算所述第一光谱响应结果中各个位置的值的第一平均值；以及
    响应于所述第一平均值在预设范围内，确定所述第一光谱响应结果与所述基准光谱响应之间的匹配成功，其中，所述预设范围的下限为所述基准光谱响应的基准平均值减去所述基准光谱响应的标准差的一半，所述预设范围的上限为所述基准平均值加上所述基准光谱响应的标准差的一半。
  16. 一种识别系统，其特征在于，所述识别系统能够执行如权利要求1至15任一所述的工作方法。
PCT/CN2022/139150 2021-12-16 2022-12-14 识别系统及其工作方法 WO2023109874A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111544309.3 2021-12-16
CN202111544309.3A CN116266409A (zh) 2021-12-16 2021-12-16 识别系统及其工作方法

Publications (1)

Publication Number Publication Date
WO2023109874A1 true WO2023109874A1 (zh) 2023-06-22

Family

ID=86743224

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/139150 WO2023109874A1 (zh) 2021-12-16 2022-12-14 识别系统及其工作方法

Country Status (2)

Country Link
CN (1) CN116266409A (zh)
WO (1) WO2023109874A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160338592A1 (en) * 2015-05-20 2016-11-24 Canon Kabushiki Kaisha Control apparatus, measurement apparatus, control method, and storage medium
US20180239944A1 (en) * 2015-09-17 2018-08-23 Nec Corporation Living body determination device, living body determination method, and program
CN112446271A (zh) * 2019-08-30 2021-03-05 指纹卡有限公司 生物特征光学反欺骗
CN113468915A (zh) * 2020-03-31 2021-10-01 吉林求是光谱数据科技有限公司 具有真假指纹识别功能的手机指纹识别系统和识别方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160338592A1 (en) * 2015-05-20 2016-11-24 Canon Kabushiki Kaisha Control apparatus, measurement apparatus, control method, and storage medium
US20180239944A1 (en) * 2015-09-17 2018-08-23 Nec Corporation Living body determination device, living body determination method, and program
CN112446271A (zh) * 2019-08-30 2021-03-05 指纹卡有限公司 生物特征光学反欺骗
CN113468915A (zh) * 2020-03-31 2021-10-01 吉林求是光谱数据科技有限公司 具有真假指纹识别功能的手机指纹识别系统和识别方法

Also Published As

Publication number Publication date
CN116266409A (zh) 2023-06-20

Similar Documents

Publication Publication Date Title
US20220114834A1 (en) Face recognition monitoring system based on spectrum and multi-band fusion and recognition method using same
US11294996B2 (en) Systems and methods for using machine learning for image-based spoof detection
RU2328035C2 (ru) Устройство обработки информации
Du et al. A prism-based system for multispectral video acquisition
KR101080643B1 (ko) 생체지문 판단장치 및 그 판단방법
KR101672541B1 (ko) 지문과 지정맥을 동시에 인증하는 3d 입체 촬영장치
KR102146089B1 (ko) 지문 센서를 포함하는 디바이스의 안티-스푸핑 방법 및 시스템
KR102317180B1 (ko) 3차원 깊이정보 및 적외선정보에 기반하여 생체여부의 확인을 행하는 얼굴인식 장치 및 방법
KR102009000B1 (ko) 지문 센서를 포함하는 디바이스의 안티-스푸핑 방법 및 시스템
JP2010035560A (ja) 生体認証装置
US20230196842A1 (en) Spoof-Resistant Facial Recognition Through Illumination and Imaging Engineering
JP5763160B2 (ja) 生体認証装置及び個人認証システム
CN111222472A (zh) 一种基于结构光频域特征的人脸识别方法
CN111428673A (zh) 一种用于指纹增强显示的偏振矢量彩色图像融合方法
WO2023109874A1 (zh) 识别系统及其工作方法
Ngo et al. Design and implementation of a multispectral iris capture system
TWI592882B (zh) 偽裝影像檢測方法及系統
CN111699678B (zh) 使用面部id感测的安全检查系统和安全检查方法
WO2023143242A1 (zh) 指纹检测模组、活体指纹检测装置以及检测方法
KR20100123812A (ko) 생체지문 판단장치 및 그 판단방법
CN216901720U (zh) 活体指纹检测装置
US11741748B2 (en) Passive image depth sensing for object verification based on chromatic differentiation
CN116563955A (zh) 活体指纹检测装置和检测方法
WO2021045140A1 (ja) 情報処理装置、脈波計測システム及び脈波計測プログラム
CN117011897A (zh) 指纹检测模组和活体指纹识别系统

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: 22906619

Country of ref document: EP

Kind code of ref document: A1