Nothing Special   »   [go: up one dir, main page]

WO2021218823A1 - 指纹活体检测方法、设备及存储介质 - Google Patents

指纹活体检测方法、设备及存储介质 Download PDF

Info

Publication number
WO2021218823A1
WO2021218823A1 PCT/CN2021/089361 CN2021089361W WO2021218823A1 WO 2021218823 A1 WO2021218823 A1 WO 2021218823A1 CN 2021089361 W CN2021089361 W CN 2021089361W WO 2021218823 A1 WO2021218823 A1 WO 2021218823A1
Authority
WO
WIPO (PCT)
Prior art keywords
fingerprint
image
gray
feature vector
estimated
Prior art date
Application number
PCT/CN2021/089361
Other languages
English (en)
French (fr)
Inventor
王升国
贺岳明
任志浩
张云生
Original Assignee
杭州海康威视数字技术股份有限公司
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 杭州海康威视数字技术股份有限公司 filed Critical 杭州海康威视数字技术股份有限公司
Priority to KR1020227030315A priority Critical patent/KR20220130804A/ko
Priority to EP21797544.0A priority patent/EP4145343A4/en
Publication of WO2021218823A1 publication Critical patent/WO2021218823A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/40—Spoof detection, e.g. liveness detection
    • G06V40/45—Detection of the body part being alive
    • 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
    • G06V40/1382—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
    • G06V40/1388—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger using image processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/12—Details of acquisition arrangements; Constructional details thereof
    • G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
    • G06V10/143—Sensing or illuminating at different wavelengths
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • 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
    • G06V10/72—Data preparation, e.g. statistical preprocessing of image or video features
    • 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
    • G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
    • 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
    • G06V10/778—Active pattern-learning, e.g. online learning of image or video features
    • 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
    • G06V40/1347—Preprocessing; Feature extraction
    • 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
    • G06V40/1382—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
    • 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
    • G06V40/1365—Matching; Classification

Definitions

  • the embodiments of the present application relate to the field of security technologies, and in particular, to a method, device, and storage medium for detecting a fingerprint in vivo.
  • fingerprint recognition has very important applications in many scenarios, such as attendance, access control, computer unlocking, mobile phone unlocking, payment and other scenarios.
  • Fingerprint recognition is mainly to extract the characteristics of fingerprint images, such as lines, break points, intersections, etc. Determine whether it is a valid fingerprint.
  • the current fingerprint forgery technology has also been developed, for example, using silica gel to make a fingerprint film to forge other people's fingerprints, and successfully perform some illegal actions through fingerprint recognition.
  • silica gel to make a fingerprint film to forge other people's fingerprints, and successfully perform some illegal actions through fingerprint recognition.
  • the embodiments of the present application provide a fingerprint living detection method, device, and storage medium, which can improve the security of the fingerprint living detection method.
  • the technical solution is as follows:
  • a method for fingerprint living detection includes:
  • the first fingerprint image and the second fingerprint image refer to images of the same fingerprint, the first fingerprint image is a red channel image, and the second fingerprint image is blue Color channel image
  • the first fingerprint image and the second fingerprint image are respectively divided into regions, wherein the divided first fingerprint image includes a plurality of first estimated bright areas and a plurality of first estimated dark areas.
  • the second fingerprint image includes multiple second predicted bright areas and multiple second predicted dark areas;
  • the plurality of first estimated bright areas determines a multi-zone gray scale Distribution feature vector
  • the multi-partition gray distribution feature vector is input to a fingerprint living detection model to obtain a fingerprint living detection result.
  • the fingerprint living detection model is obtained through supervised learning training based on a living fingerprint sample image and a non- living fingerprint sample image.
  • the acquiring the first fingerprint image and the second fingerprint image includes:
  • the first original image is an image of the fingerprint collected when the green light is off, and the red light and blue light are on;
  • the acquiring the first fingerprint image and the second fingerprint image includes:
  • the second original image is an image of the fingerprint collected when the green light and blue light are off, and the red light is on
  • the third original image is the green light and the red light is off , The image of the fingerprint collected when the blue light is on;
  • the division of regions of the first fingerprint image and the second fingerprint image respectively includes:
  • partition parameter Determine the partition parameter according to the distribution law of the bright area and the dark area in the red channel image and the blue channel image of the living fingerprint, and the partition parameter includes a plurality of partition width ratios;
  • the first fingerprint image and the second fingerprint image are respectively divided into regions.
  • the image size of the first fingerprint image and the second fingerprint image are the same, and the plurality of first predicted bright areas, the plurality of first predicted dark areas, and the multiple The second predicted bright area and the multiple second predicted dark areas to determine the multi-region gray distribution feature vector include:
  • the multi-partition gray-scale distribution feature vector is generated.
  • the determining the first gray-scale feature vector according to the plurality of first estimated bright areas and the plurality of second estimated dark areas includes:
  • the first grayscale feature vector is generated according to the number of pixels corresponding to each grayscale value in the plurality of first estimated bright areas and the plurality of second estimated dark areas, respectively.
  • the determining a second gray-scale feature vector according to the plurality of first estimated dark areas and the plurality of second estimated bright areas includes:
  • the second grayscale feature vector is generated according to the number of pixels corresponding to each grayscale value in the plurality of first predicted dark areas and the plurality of second predicted bright areas, respectively.
  • the method before determining the first gray-scale feature vector according to the plurality of first estimated bright areas and the plurality of second estimated dark areas, the method further includes:
  • the gray value statistical range being determined according to an exposure parameter of an image capture device that captures the image of the fingerprint
  • the determining a first gray-scale feature vector according to the plurality of first estimated bright areas and the plurality of second estimated dark areas includes:
  • the gray value statistical range determine the number of pixels corresponding to each gray value within the gray value statistical range in the plurality of first estimated bright areas and the plurality of second estimated dark areas number;
  • the first grayscale is generated according to the number of pixels corresponding to each grayscale value in the grayscale value statistical range in the plurality of first estimated bright areas and the plurality of second estimated dark areas Feature vector.
  • the generating the multi-partition gray-scale distribution feature vector according to the first gray-scale feature vector and the second gray-scale feature vector includes:
  • the first gray-scale feature vector and the second gray-scale feature vector are spliced to obtain the multi-partition gray-scale distribution feature vector.
  • the step of inputting the multi-partition gray distribution feature vector into a fingerprint living body detection model to obtain a fingerprint living body detection result further includes:
  • the live fingerprint sample image includes images of multiple live fingerprint samples collected when the green light is off, the red light and the blue light are on
  • the non-living fingerprint Fingerprint sample images include images of multiple non-living fingerprint samples collected when the green light is off, the red light and the blue light are on
  • the fingerprint live detection model is obtained through supervised learning training.
  • the method further includes:
  • the security verification result is determined.
  • a fingerprint biometric detection device in another aspect, includes:
  • the first acquisition module is used to acquire a first fingerprint image and a second fingerprint image.
  • the first fingerprint image and the second fingerprint image refer to images of the same fingerprint, and the first fingerprint image is a red channel image.
  • the second fingerprint image is a blue channel image;
  • the partition module is used to partition the first fingerprint image and the second fingerprint image respectively, wherein the divided first fingerprint image includes a plurality of first estimated bright areas and a plurality of first estimated areas Dark area, the divided second fingerprint image includes multiple second predicted bright areas and multiple second predicted dark areas;
  • the first determining module is used to determine the number of dark areas according to the plurality of first estimated bright areas, the plurality of first estimated dark areas, the plurality of second estimated bright areas, and the plurality of second estimated dark areas. Zone, determine the multi-zone gray distribution feature vector;
  • the detection module is used to input the multi-partition gray distribution feature vector into the fingerprint live detection model to obtain the fingerprint live detection result.
  • the fingerprint live detection model is obtained through supervised learning training based on the live fingerprint sample image and the non-live fingerprint sample image of.
  • the first obtaining module includes:
  • the first acquiring unit is configured to acquire a first original image, where the first original image is an image of the fingerprint collected when the green light is off, and the red light and blue light are on;
  • the first extraction unit is configured to extract the image data of the red channel of the first original image to generate the first fingerprint image
  • the second extraction unit is used to extract the image data of the blue channel of the first original image to generate the second fingerprint image.
  • the first obtaining module includes:
  • the second acquisition unit is configured to acquire a second original image and a third original image
  • the second original image is an image of the fingerprint collected with the green light and blue light off and the red light on
  • the third original image The image is an image of the fingerprint collected when the green light and the red light are off, and the blue light is on;
  • the third extraction unit is configured to extract the image data of the red channel of the second original image to generate the first fingerprint image
  • the fourth extraction unit is used to extract the image data of the blue channel of the third original image to generate the second fingerprint image.
  • the partition module includes:
  • the first determining unit is configured to determine the partition parameter according to the distribution law of the bright area and the dark area in the red channel image and the blue channel image of the living fingerprint, and the partition parameter includes a plurality of partition width ratios;
  • the partition unit is configured to partition the first fingerprint image and the second fingerprint image according to the width ratios of the plurality of partitions.
  • the image sizes of the first fingerprint image and the second fingerprint image are the same, and the first determining module includes:
  • a second determining unit configured to determine a first grayscale feature vector according to the plurality of first estimated bright areas and the plurality of second estimated dark areas;
  • a third determining unit configured to determine a second grayscale feature vector according to the plurality of first estimated dark areas and the plurality of second estimated bright areas;
  • the generating unit is configured to generate the multi-partition gray-scale distribution feature vector according to the first gray-scale feature vector and the second gray-scale feature vector.
  • the second determining unit includes:
  • the first determining subunit is configured to determine the number of pixels corresponding to each gray value in the plurality of first estimated bright areas and the plurality of second estimated dark areas;
  • the first generating subunit is configured to generate the first gray scale according to the number of pixels corresponding to each gray value in the plurality of first estimated bright areas and the plurality of second estimated dark areas, respectively Feature vector.
  • the third determining unit includes:
  • the second determining subunit is configured to determine the number of pixels corresponding to each gray value of the plurality of first estimated dark areas and the plurality of second estimated bright areas respectively;
  • the second generating subunit is configured to generate the second gray scale according to the number of pixels corresponding to each gray value in the plurality of first estimated dark areas and the plurality of second estimated bright areas, respectively Feature vector.
  • the first determining module further includes:
  • the third acquiring unit is configured to acquire a gray value statistical range, the gray value statistical range being determined according to an exposure parameter of an image capture device that captures the image of the fingerprint;
  • the second determining unit includes:
  • the third determining subunit is configured to determine each of the plurality of first predicted bright areas and the plurality of second predicted dark areas within the gray value statistical range according to the gray value statistical range The number of pixels corresponding to the gray value;
  • the third generation subunit is configured to calculate the number of pixels corresponding to each gray value within the gray value statistical range in the plurality of first estimated bright areas and the plurality of second estimated dark areas , Generating the first gray-scale feature vector.
  • the generating unit includes:
  • the splicing subunit is used to splice the first gray-scale feature vector and the second gray-scale feature vector to obtain the multi-partition gray-scale distribution feature vector.
  • the device further includes:
  • the second acquisition module is used to acquire the live fingerprint sample image and the non-live fingerprint sample image
  • the live fingerprint sample image includes a plurality of live fingerprint samples collected when the green light is off, the red light and the blue light are on
  • the image of the non-living fingerprint sample includes images of multiple non-living fingerprint samples collected when the green light is off, and the red light and blue light are on;
  • the training module is used to obtain the fingerprint live detection model through supervised learning training according to the live fingerprint sample image and the non-live fingerprint sample image.
  • the device further includes:
  • the third acquisition module is configured to acquire a fourth original image, where the fourth original image is an image of the fingerprint collected when the green light is on;
  • An identification module configured to determine a fingerprint identification result according to the fourth original image
  • the second determining module is configured to determine the security verification result according to the fingerprint identification result and the fingerprint live detection result.
  • a fingerprint living body detection device in another aspect, includes an image collector and a processor, and the image collector includes a fill light and an image sensor;
  • the image collector is used to collect a first original image
  • the supplemental light is used to turn off the green light, turn on the red light and the blue light to supplement light for the image sensor during the process of collecting the first original image;
  • the processor is configured to process the first original image to obtain a first fingerprint image and a second fingerprint image, where the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image
  • the first fingerprint image and the second fingerprint image are respectively divided into regions, wherein the divided first fingerprint image includes a plurality of first estimated bright areas and a plurality of first estimated dark areas, divided
  • the second fingerprint image includes multiple second predicted bright areas and multiple second predicted dark areas; according to the multiple first predicted bright areas, the multiple first predicted dark areas, and The plurality of second predicted bright areas and the plurality of second predicted dark areas determine a multi-division gray distribution feature vector; input the multi-division gray distribution feature vector into a fingerprint living detection model to obtain a fingerprint living detection
  • the fingerprint live detection model is obtained through supervised learning training based on live fingerprint sample images and non-live fingerprint sample images.
  • the image collector is also used to collect a fourth original image
  • the supplementary light is also used to turn on the green light to supplement the light for the image sensor during the process of collecting the fourth original image;
  • the processor is further configured to determine a fingerprint recognition result according to the fourth original image, and determine a security verification result according to the fingerprint recognition result and the fingerprint live detection result.
  • a computer-readable storage medium is provided, and a computer program is stored in the storage medium, and when the computer program is executed by a processor, the steps of the above-mentioned fingerprint living detection method are realized.
  • a computer program product containing instructions which when run on a computer, causes the computer to execute the steps of the fingerprint living detection method described above.
  • the first fingerprint image and the second fingerprint image can be divided into regions, and then the multi-partition gray distribution feature vector can be determined according to the divided regions, so as to perform fingerprint living detection through the fingerprint living detection model, where:
  • the first fingerprint image is a red channel image
  • the second fingerprint image is a blue channel image. Since the distribution of gray values in each area of the two channel images of non-living fingerprints is different from that of living fingerprints, the multi-partition gray distribution feature vector can be used to identify whether the fingerprint image has the gray distribution characteristics of living fingerprints. Then it is determined whether the fingerprint is a living fingerprint, and the security is relatively high.
  • FIG. 1 is a schematic structural diagram of a fingerprint living body detection device provided by an embodiment of the present application
  • FIG. 2 is a flowchart of a fingerprint living body detection method provided by an embodiment of the present application
  • FIG. 3 is a schematic diagram of an image sensor used in an embodiment of the present application.
  • FIG. 4 is a schematic diagram of extracted image data of the red channel provided by an embodiment of the present application.
  • FIG. 5 is a schematic diagram of a first fingerprint image provided by an embodiment of the present application.
  • Fig. 6 is a schematic diagram of extracted image data of the blue channel provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram of a second fingerprint image provided by an embodiment of the present application.
  • FIG. 8 is a schematic diagram of determining each area of an image provided by an embodiment of the present application.
  • FIG. 9 is a flowchart of another fingerprint living body detection method provided by an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a fingerprint living body detection device provided by an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of another fingerprint living body detection device provided by an embodiment of the present application.
  • FIG. 12 is a schematic structural diagram of a computer device provided by an embodiment of the present application.
  • fingerprint recognition has very important applications in many scenarios, such as attendance, access control, computer unlocking, mobile phone unlocking, payment and other scenarios.
  • fingerprint forgery technology has also been developed, for example, using silica gel to make a fingerprint film to forge other people's fingerprints, and successfully perform some illegal actions through fingerprint recognition.
  • silica gel to make a fingerprint film to forge other people's fingerprints, and successfully perform some illegal actions through fingerprint recognition.
  • fingerprint live detection can be performed according to the method provided in the embodiments of this application to obtain fingerprint live detection results, and fingerprint recognition is performed on the collected fingerprint images on this basis to obtain fingerprint recognition results , In the case that both the fingerprint identification and the fingerprint live detection results are passed, the security verification is determined to be passed.
  • FIG. 1 is a schematic structural diagram of a fingerprint living body detection device 100 provided by an embodiment of the present application.
  • the fingerprint living detection device 100 includes an image collector 101 and a processor 102.
  • the image collector 101 includes an LED fill light, a lens, and an image sensor.
  • the LED fill light includes a green light, a red light, and a blue light.
  • the image collector 101 is used to receive optical signals or sense capacitance, convert the received optical signals or sensed capacitances into electrical signals and send them to the processor 102.
  • the processor 102 is used to determine the current light intensity according to the received electrical signals. The amount of change or capacitance value. When the amount of change in light intensity indicates that the current light is obviously dimmed or the capacitance value is the capacitance value of a living finger, send a control instruction for collecting fingerprint images to the image collector 101, that is, wake up the image collector 101 performs fingerprint collection.
  • the image collector 101 When the image collector 101 receives the control instruction, it controls the LED supplementary light to turn on the green light, or turn on the red and blue lights to supplement light for the image sensor according to the fingerprint image acquisition method provided in the embodiment of the application, through the lens and The image sensor collects the original image, and sends the collected original image to the processor 102.
  • a fingerprint living detection model is deployed in the processor 102.
  • the original image can be processed according to the fingerprint living detection method provided in the embodiment of the application to obtain the fingerprint living body. Test results.
  • the fingerprint live detection model is obtained through supervised learning training based on live fingerprint sample images and non-live fingerprint sample images.
  • the image collector 101 can collect the first original image, and the LED supplement light can turn off the green light, turn on the red light and the blue light during the process of collecting the first original image to supplement light for the image sensor, and process
  • the device 102 may process the first original image to obtain a first fingerprint image and a second fingerprint image.
  • the first fingerprint image is a red channel image and the second fingerprint image is a blue channel image.
  • the processor 102 may divide the first fingerprint image and the second fingerprint image respectively, where the divided first fingerprint image includes a plurality of first estimated bright areas and a plurality of first estimated dark areas,
  • the divided second fingerprint image includes multiple second predicted bright areas and multiple second predicted dark areas.
  • the processor 102 may also determine the multi-zone gray distribution characteristics according to the multiple first predicted bright areas, multiple first predicted dark areas, multiple second predicted bright areas, and multiple second predicted dark areas Vector, and then input the multi-partition gray distribution feature vector into the fingerprint living detection model to obtain the fingerprint living detection result.
  • the image collector 101 may also collect a fourth original image
  • the LED fill light may turn on the green light to fill the image sensor with light during the process of collecting the fourth original image
  • the processor 102 may use the fourth original image, Determine the fingerprint recognition result, and determine the security verification result based on the fingerprint recognition result and the fingerprint live detection result.
  • the fingerprint biometric detection device 100 may further include a memory 103, which is used to store data and computer programs required by the embodiment of the present application, for example, images of multiple valid fingerprints are stored for fingerprint identification.
  • the processor 102 can execute the computer program in the memory 103, and compare the collected original image with the stored images of multiple valid fingerprints to perform fingerprint recognition, and obtain the fingerprint recognition result. Finally, it can also be based on the fingerprint live detection result and The fingerprint recognition result is used to determine the security verification result.
  • the LED supplementary light supplements the light for the image sensor to collect each original image, and the processor 102 processes each original image to obtain the fingerprint living body detection result or the specific implementation of the fingerprint recognition result. You can refer to the following figure. 2 Relevant introduction of method embodiment.
  • the LED supplementary light includes red, green, and blue lights.
  • the wavelength ranges of the red, green, and blue lights corresponding to these three lights can be based on the QE (Quantum Efficiency) of the image sensor. , Quantum efficiency) curve to determine. For example, when only the red light is turned on, when the peak of the QE curve of the image sensor reaches the maximum, the wavelength corresponding to the peak position can be used as the center wavelength of the red light, and a section of wavelengths can be taken from the center wavelength to the left and to the right. Range, as the wavelength range of red light.
  • the wavelength range of green light determined according to this method can be 500nm-550nm
  • the wavelength range of red light can be 600nm-650nm
  • the wavelength range of blue light can be 420nm-480nm.
  • the fingerprint living detection device 100 shown in FIG. 1 may refer to a front-end image acquisition device, such as a camera, that is, the fingerprint living detection method provided in the embodiments of the present application can be applied to the front-end image acquisition device .
  • the function of the fingerprint biometric detection device 100 can also be implemented by two separate devices, a front-end device and a back-end device, where the back-end device can be connected to the front-end device through a wired or wireless connection to perform data. transmission.
  • the front-end device may include the image collector 101 in the foregoing embodiment to collect the original image.
  • the front-end device can send the collected original image to the back-end device, and the back-end device can use this
  • the fingerprint living detection method provided in the application embodiment detects and recognizes the collected original images, and finally determines the security verification result.
  • the security verification result can be sent to the front-end device or other devices to prompt the user to verify the security in text or voice. Pass or fail.
  • the first fingerprint image and the second fingerprint image can be divided into regions, and then the multi-division gray distribution feature vector can be determined according to the divided regions, so as to perform fingerprint living body detection through the fingerprint living body detection model. Detection, where the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image. Since the gray value distribution of the bright and dark areas of the two channel images of non-living fingerprints is different from that of living fingerprints, the multi-partition gray distribution feature vector can identify whether the fingerprint image has the grays of living fingerprints. Degree distribution characteristics, and then determine whether the fingerprint is a living fingerprint, and the security is relatively high.
  • the fingerprint biometric detection device provided in the above embodiment detects fingerprints
  • only the division of the above functional modules or devices is used as an example for illustration.
  • the above functions can be allocated to different functional modules according to needs.
  • the device is completed, that is, the internal structure of the device is divided into different functional modules or devices to complete all or part of the functions described above.
  • the fingerprint living body detection device provided in the above embodiment belongs to the same concept as the following fingerprint living body detection method embodiments. For the specific implementation process, please refer to the following method embodiments, which will not be repeated here.
  • FIG. 2 is a flowchart of a fingerprint living body detection method provided by an embodiment of the present application, and the method is applied to the fingerprint living body detection device shown in FIG. 1 as an example for description. Please refer to Figure 2.
  • the method includes the following steps.
  • Step 201 Obtain a first fingerprint image and a second fingerprint image, where the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image.
  • the red channel image and blue channel image of the living fingerprint are alternately distributed in the bright and dark areas, and the gray value distribution of the pixels is relatively regular, the gray value of the pixels of the two channel images of the non-live fingerprint
  • the value distribution is rather chaotic, that is, the distribution law of gray values corresponding to non-living fingerprints is different from that of living fingerprints. Therefore, fingerprint live detection can be performed according to the distribution law of gray values. That is, in the embodiment of the present application, the acquired first fingerprint image is a red channel image, the second fingerprint image is a blue channel image, and the first fingerprint image and the second fingerprint image refer to images of the same fingerprint.
  • the image collector needs to be awakened. It can be determined whether the collection object is currently detected, such as a finger, according to the light intensity change or the capacitance value. When the collection object is detected, you can Wake up the image collector and start collecting fingerprint images.
  • the embodiment of the application needs to perform fingerprint live detection based on the red channel image and the blue channel image, based on this, the embodiment of the application provides two image acquisition methods, that is, two methods for acquiring the first fingerprint image and the first fingerprint image are provided. Second, the realization of the fingerprint image, the following two realizations are introduced.
  • the fingerprint living detection device can obtain the first original image, which is the image of the fingerprint collected when the green light is off, and the red light and blue light are on. After that, the fingerprint living detection device can extract the first original image.
  • the image data of the red channel of the original image is generated to generate a first fingerprint image
  • the image data of the blue channel of the first original image is extracted to generate a second fingerprint image.
  • the image sensor includes a red channel, a blue channel, and a green channel.
  • the image sensor may be a Bayer mode color image sensor as shown in FIG. 3, where R is a red (R) channel, and Gr And Gb are both green (G) channels, and B is blue (B) channels.
  • the image collector After waking up the image collector, you can turn on the red light and blue light at the same time, and collect the first original image through the lens and image sensor, that is, the first original image is collected when the green light is off, and the red light and blue light are on.
  • the image of the fingerprint After that, the image collector can send the collected original image to the processor.
  • the processor may extract the image data of the red channel of the first original image to generate the first fingerprint image, and extract the image data of the blue channel to generate the second fingerprint image.
  • the size of the first original image is w0*h0, and w0 and h0 represent the width and height of the first original image, respectively. Indicates the number of pixels in the width direction and height direction.
  • the processor may extract the image data of the R channel to obtain the red channel image data as shown in FIG. 4, and generate the first fingerprint image as shown in FIG. 5 according to the red channel image data.
  • the processor can extract the image data of the B channel to obtain the blue channel image data as shown in FIG. 6, and generate the second fingerprint image as shown in FIG. 7 according to the blue channel image data, the first fingerprint image and the second fingerprint image as shown in FIG.
  • the image size of the second fingerprint image is w0*h0/4.
  • the fingerprint living detection device can obtain a second original image and a third original image.
  • the second original image is an image of the fingerprint collected when the green and blue lights are off and the red light is on.
  • the third original image It is the image of the fingerprint collected when the green light and red light are off, and the blue light is on.
  • the fingerprint living detection device may extract the image data of the red channel of the second original image to generate the first fingerprint image, extract the image data of the blue channel of the third original image, and generate the second fingerprint image.
  • the second implementation method is to collect the first original image by only turning on the red light, and then only the blue light to collect the second original image, and then extract the image data of the red channel of the first original image to generate the first image.
  • the image data of the blue channel of the second original image is extracted to generate the second fingerprint image.
  • the method of extracting image data can refer to the related introduction in the first implementation manner, which will not be repeated here.
  • two original images are collected by turning on only the red light and only the blue light. This can avoid the red light versus blue light when the red light and the blue light are turned on at the same time. The influence of the color channel, and the influence of the blue light on the red channel.
  • an original image is collected by turning off the green light and turning on the red and blue lights at the same time. Compared with the second implementation, the time for image acquisition can be reduced.
  • the fingerprint living detection device can extract the image data of the red channel and the image data of the blue channel, and can also compare the red channel image and the blue channel image. Perform distortion correction.
  • the fingerprint living detection device can also cut the red channel image and the blue channel image separately according to the preset image size parameters to remove the unclear edges of the image, and obtain the first fingerprint image and the first fingerprint image with the same image size. 2. Fingerprint image.
  • Step 202 Divide the first fingerprint image and the second fingerprint image respectively, where the divided first fingerprint image includes a plurality of first estimated bright areas and a plurality of first estimated dark areas, and the divided first fingerprint image
  • the second fingerprint image includes a plurality of second estimated bright areas and a plurality of second estimated dark areas.
  • the fingerprint living detection device can divide the first fingerprint image and the second fingerprint image according to the distribution law of the bright area and the dark area in the red channel image and the blue channel image of the living fingerprint.
  • the first fingerprint image is divided into multiple first predicted bright areas and multiple first predicted dark areas
  • the second fingerprint image is divided into multiple second predicted bright areas and multiple second predicted dark areas.
  • the distribution of the bright and dark areas of the red channel image from top to bottom along the image height direction is: bright, dark, Bright, dark, bright, blue channel image along the image height direction from top to bottom, the distribution of bright and dark areas: dark, light, dark, light, dark, that is, red channel image and blue channel image
  • the fingerprint living body detection device can divide the first fingerprint image and the second fingerprint image according to the distribution law.
  • the fingerprint living detection device can determine the partition parameter according to the distribution law of the bright and dark areas in the red channel image and the blue channel image of the living fingerprint.
  • the partition parameter includes multiple partition width ratios, and then, According to a plurality of partition width ratios, the first fingerprint image and the second fingerprint image are respectively divided into regions.
  • the partition parameter can also be a parameter set in advance according to the distribution law of each region.
  • the light and dark distribution of red channel images and blue channel images of multiple live fingerprints can be counted, and the living body can be determined according to the light and dark distribution.
  • FIG. 8 is a schematic diagram of determining each area of an image provided in an embodiment of the present application. See FIG. 8.
  • the upper left image in FIG. 8 is the red channel image of the living fingerprint, that is, the first fingerprint image
  • the upper right image is The blue channel image of the living fingerprint, that is, the second fingerprint image, through the red channel image and the blue channel image both show the distribution of light and dark alternately, the red channel image and the blue channel image can be divided into 5 from top to bottom. From top to bottom, the 5 regions included in the red channel image are bright, dark, bright, dark, and bright. The 5 regions included in the blue channel image are dark and bright from top to bottom.
  • Area, dark area, bright area and dark area, and the 5 areas of the red channel image correspond to the 5 areas of the blue channel one-to-one, that is, the width ratio of the corresponding areas in the two images is the same, the difference is that the red channel In the bright area of the image, the corresponding area in the blue channel image at the same height range is the dark area.
  • both the blue channel image and the red channel image can be divided into five areas as shown in Figure 8 by the same division rule, namely area 1.
  • the corresponding partition width ratios of the 5 areas are shown in Figure 8. For example, from top to bottom, they can be 20/66, 10/66, 15/66, 9/66, and 12/66, other ratios are also possible.
  • the fingerprint living detection device can partition the first fingerprint image and the second fingerprint image according to the partition width ratio shown in FIG. 8.
  • the divided first fingerprint image includes 3 first predicted bright areas and 2 first preset Estimate the dark area, where the 3 first estimated bright areas correspond to area 1, area 3, and area 5 in Figure 8, and the two first estimated dark areas correspond to area 2 and area 4 in Fig. 8.
  • the divided second fingerprint image includes two second predicted bright areas and three second predicted dark areas, among which the two second predicted bright areas correspond to areas 2 and 4 in Fig. 8, and the third Second, the predicted dark area corresponds to area 1, area 3, and area 5 in Figure 8.
  • the partition parameter may include multiple width values.
  • the width values of the five regions shown in FIG. 8 may be 20, 10, 15, 9 and 12 pixels in order from top to bottom.
  • the fingerprint living body detection device can divide the first fingerprint image and the second fingerprint image into regions according to the multiple width values.
  • Step 203 According to the multiple first predicted bright areas, multiple first predicted dark areas, multiple second predicted bright areas, and multiple second predicted dark areas, determine a multi-zone gray distribution feature vector.
  • the image size of the first fingerprint image and the second fingerprint image are the same.
  • the fingerprint living body detection device can divide the regions according to statistics.
  • the fingerprint living detection device may determine the first gray-scale feature vector according to the plurality of first estimated bright areas and the plurality of second estimated dark areas, and according to the plurality of first estimated dark areas Area and the plurality of second predicted bright areas to determine the second gray-scale feature vector. Afterwards, the fingerprint living detection device can generate a multi-partition gray distribution feature vector according to the first gray feature vector and the second gray feature vector.
  • the fingerprint living detection device can determine the first grayscale feature vector and the second grayscale feature vector by counting the grayscale values of the respective regions of the first fingerprint image and the second fingerprint image.
  • the fingerprint living detection device can determine the number of pixels corresponding to each gray value in the plurality of first estimated bright areas and the plurality of second estimated dark areas, and according to the plurality of first estimated dark areas One predicts the number of pixels corresponding to each gray value in the bright area and the plurality of second estimated dark areas, and generates the first gray level feature vector. That is, the fingerprint living detection device can count the number of pixels corresponding to each gray value in the multiple first predicted bright areas of the first fingerprint image and the multiple second predicted dark areas of the second fingerprint image. , To determine the first gray-scale feature vector.
  • the plurality of first estimated bright areas correspond to the plurality of second estimated dark areas in a one-to-one correspondence, and the corresponding areas are areas in the same height range of the two images.
  • the fingerprint living detection device can count the grayscale values in area 1, area 3, and area 5 of the first fingerprint image.
  • the number of pixels corresponding to the degree value, the number of pixels corresponding to each gray value obtained by statistics, are arranged in order from 0-255 in order of gray value, so as to obtain the vector hisr1.
  • the number of points is arranged in sequence from 0-255 according to the gray value, and the vector histb1 is obtained.
  • the length of the vector hisr1 and the vector histb1 are both 256.
  • the vector hisr1 and the vector histb1 can be added to obtain the first gray-scale feature vector hisfp1, and the length of the vector hisfp1 is also 256.
  • the fingerprint living detection device can determine the number of pixels corresponding to each gray value of the plurality of first estimated dark areas and the plurality of second estimated bright areas, and calculate the number of pixels according to the plurality of first estimated dark areas.
  • the number of pixels corresponding to each gray value in the dark area and the plurality of second estimated bright areas, respectively, generates a second gray level feature vector. That is, the fingerprint living detection device can count the number of pixels corresponding to each gray value in the multiple first predicted dark areas of the first fingerprint image and the multiple second predicted bright areas of the second fingerprint image. , To determine the second gray-scale feature vector.
  • the fingerprint living detection device can count the gray values of each of the gray values from “0" to "255" in area 2 and area 4 of the first fingerprint image.
  • the number of pixels corresponding to each gray value obtained by the statistics is arranged in the order of gray value from 0-255 to obtain the vector hisr2.
  • Count the number of pixels of each gray value from gray value "0" to "255” in area 2 and area 4 of the second fingerprint image, and count the number of pixels corresponding to each gray value obtained Arranged in the order of gray value from 0 to 255 to get the vector histb2.
  • the length of the vector hisr2 and the vector histb2 are both 256.
  • the vector hisr2 and the vector histb2 can be added to obtain the second gray-scale feature vector hisfp2, and the length of the vector hisfp2 is also 256.
  • the exposure of the fingerprint in vivo detection device will affect the captured image, the exposure is too high, the captured image is too bright, the exposure is too low, the captured image is too dark, in order to reduce the exposure Influencing, retaining effective gray information, can reduce the statistical range of gray values, that is, a statistical range of gray values can be determined according to the exposure parameters of the image acquisition device that collects the image of the fingerprint.
  • the exposure parameters of different image acquisition devices may be different, and the corresponding gray value statistical range can be set according to the exposure parameters of each device.
  • the fingerprint living body detection device is an image acquisition device.
  • the amount of calculation can be reduced and the detection speed can be accelerated.
  • the fingerprint biometric detection device can obtain the gray value statistical range, and according to the gray value statistical range, determine the plurality of first estimated bright areas and the plurality of second estimated areas The number of pixels corresponding to each gray value in the dark area in the gray value statistical range is based on the plurality of first estimated bright areas and the plurality of second estimated dark areas in the gray value statistical range The number of pixels corresponding to each gray value generates the first gray level feature vector.
  • the fingerprint living detection device can determine the corresponding gray values of the plurality of first predicted dark areas and the plurality of second predicted bright areas in the gray value statistical range according to the gray value statistical range According to the number of pixels in the plurality of first predicted dark areas and the plurality of second predicted bright areas, the number of pixels corresponding to each gray value within the gray value statistical range is used to generate the second gray Feature vector.
  • the fingerprint living detection device can only count the number of pixels corresponding to each gray value within the gray value statistical range in each area of the first fingerprint image and the second fingerprint image, and follow the aforementioned method To generate the first gray-scale feature vector and the second gray-scale feature vector.
  • the gray value statistical range may be from '11' to '230', so that the length of the first gray feature vector and the second gray feature vector obtained according to the foregoing method are both 220, and the length of the vector is reduced , which can reduce the amount of subsequent calculations.
  • the fingerprint living detection device can splice the first gray-scale feature vector and the second gray-scale feature vector to obtain a multi-partition gray-scale distribution feature vector.
  • the first gray-scale feature vector is hisfp1
  • the second gray-scale feature vector is hisfp2.
  • the obtained multi-partition gray-scale distribution feature vector may be [hisfp1, hisfp2].
  • the steps of the fingerprint living detection device described above to obtain the multi-zone gray distribution feature vector are to first count the multiple first predicted bright areas of the first fingerprint image and the multiple second predicted dark areas of the second fingerprint image, respectively Obtain two vectors, add the two vectors to obtain the first gray-scale feature vector, and then count multiple first predicted dark areas of the first fingerprint image and multiple second predicted bright areas of the second fingerprint image, respectively Two vectors are obtained, and the two vectors are added to obtain a second gray-scale feature vector, and then the first gray-scale feature vector and the second gray-scale feature vector are spliced to obtain a multi-partition gray-scale distribution feature vector.
  • the fingerprint living detection device may also first count multiple first predicted bright areas of the first fingerprint image to obtain the vector hisr1, and count multiple first predicted dark areas of the first fingerprint image. Area, get the vector hisr2, stitch the vector hisr1 and the vector hisr2 to get the first gray-scale mosaic vector hisr, then count the multiple second predicted dark areas of the second fingerprint image, get the vector histb1, and count the second fingerprint image Multiple second predicted bright areas, get the vector histb2, join the vector histb1 and the vector histb2 to get the second gray mosaic vector histb, and then add the vector hisr and the vector histb to get the multi-partition gray distribution feature vector hisfp.
  • the two images are counted separately to obtain the gray-scale mosaic vectors of the bright and dark areas corresponding to the two images, and then the two gray-scale mosaic vectors are added to obtain the multi-region gray distribution feature vector.
  • the fingerprint living detection device can process one image first, and then process another image, without changing the image data of the image to be counted back and forth.
  • the statistical order of each region can be in no particular order, and a statistical order can be determined according to the actual situation.
  • the fingerprint living body detection device may also use the gray-scale histogram to represent each of the above-mentioned statistical results, and determine each vector according to the gray-scale histogram.
  • Step 204 Input the multi-zone gray distribution feature vector into the fingerprint living body detection model to obtain the fingerprint living body detection result.
  • the fingerprint living body detection device after obtaining the multi-zone gray distribution feature vector, can input the vector into the fingerprint living body detection model to obtain the fingerprint living body detection result.
  • the fingerprint living detection model may be a support vector machine (SVM) classifier, a neural network model, a random forest model, etc., which are not limited in the embodiment of the application.
  • SVM support vector machine
  • the result of fingerprint live detection can be ‘is a live fingerprint’ or ‘not a live fingerprint’, or ‘0’ or ‘1’, ‘0’ means not a live fingerprint, and ‘1’ means a live fingerprint.
  • the fingerprint live detection model can be obtained through supervised learning training based on live fingerprint sample images and non-live fingerprint sample images.
  • the fingerprint live detection model is a pre-trained model, which can be The fingerprint living detection model can be trained on the living body detection device, or the fingerprint living body detection model can also be trained by other equipment, such as background equipment, or other computer equipment, and then the fingerprint living body detection model can be deployed on the fingerprint living body detection device superior.
  • the embodiment of the present application is introduced by taking the fingerprint living detection model obtained by training on the fingerprint living detection device as an example.
  • the fingerprint live detection device can obtain live fingerprint sample images and non-live fingerprint sample images.
  • the live fingerprint sample images include multiple live fingerprint samples collected when the green light is off, the red light and the blue light are on.
  • Image, non-living fingerprint sample images include images of multiple non-living fingerprint samples collected when the green light is off, the red light and the blue light are on.
  • the fingerprint living detection device can obtain a fingerprint living detection model through supervised learning training based on the living fingerprint sample image and the non-living fingerprint sample image.
  • the method of acquiring the live fingerprint sample image and the non-live fingerprint sample image is the same as the method of acquiring the first original image described above, and will not be repeated here.
  • the fingerprint live detection device After acquiring the live fingerprint sample image and the non-live fingerprint sample image, the fingerprint live detection device can obtain the multi-region gray distribution feature vector corresponding to each sample image according to the foregoing method of obtaining the multi-region gray distribution feature vector. After that, the fingerprint living detection model can be trained according to the multi-region gray distribution feature vector corresponding to each sample image.
  • each sample image corresponds to a sample label
  • the sample label of the living fingerprint sample image can be '1'
  • the sample label of the non-living fingerprint sample image can be '0'
  • the fingerprint living detection device can correspond to each sample image
  • the multi-partition gray distribution feature vector and sample label of, the fingerprint living detection model is obtained through supervised learning training.
  • the fingerprint living detection device can obtain the SVM classifier through supervised learning training according to the multi-partition gray distribution feature vector and sample label corresponding to each sample image, the SVM classifier It is a two-classifier.
  • Fig. 9 is a flowchart of another fingerprint living detection method provided by an embodiment of the present application.
  • the fingerprint live detection model is an SVM classifier
  • the vector hisr and the vector histb are spliced together to obtain the multi-partition gray distribution feature vector hiafp.
  • the vector hisfp is input into the SVM classifier, and the fingerprint live detection result is output.
  • fingerprint identification and fingerprint living detection need to be combined to determine the final security verification result. That is, if the fingerprint recognition result indicates that the collected fingerprint is a valid fingerprint, and the fingerprint live detection result indicates that the collected fingerprint is a live fingerprint, it is determined that the security verification is passed. Based on this, the embodiment of the present application also provides a fingerprint security verification method, which will be introduced in the following.
  • the fingerprint living detection device may also obtain a fourth original image, which is an image of the fingerprint collected when the green light is turned on, and the fingerprint recognition result is determined based on the fourth original image. After that, the fingerprint living detection device can determine the security verification result based on the fingerprint identification result and the fingerprint living detection result.
  • the green channel image of the living fingerprint can present relatively clear fingerprint lines, that is, there is no feature of alternating light and dark, the green channel image is more suitable for fingerprint recognition than the red channel image and the blue channel image.
  • the image of the collected living fingerprint also presents a very clear fingerprint pattern, which is also suitable for fingerprint identification. Based on this, when the image collector is awakened for fingerprint collection, not only the original image of the fingerprint can be collected when the red and blue lights are on, but also the image of the fingerprint can be collected when the green light is on to obtain a fourth original image.
  • the state where the green light is turned on may refer to the state where the green light is turned on, the red light and the blue light are turned off, or the state where the green light, the red light and the blue light are all turned on. That is, after waking up the image collector, the first original image and the fourth original image can be collected to complete one collection, or the second original image, the third original image, and the fourth original image can be collected to complete one collection.
  • all the lights can be turned off, and the embodiment of the present application does not limit the order of the original images to be acquired.
  • only the green light may be turned on to collect the fourth original image, and then the green light may be turned off, and the red light and blue light may be turned on to collect the first original image.
  • the green light, red light, and blue light may be turned on to acquire the fourth original image, and then the green light may be turned off, and the red light and blue light may be kept on to acquire the first original image.
  • the fingerprint living detection device can extract the image data of the green channel of the fourth original image to generate a third fingerprint image, and then extract the fingerprint features of the third fingerprint image, such as the number of patterns, the shape, and Core points, break points, intersections, directions, curvatures, etc., are compared according to the extracted fingerprint features with the corresponding features of the stored images of each valid fingerprint to obtain the fingerprint recognition result.
  • the fingerprint identification result can be'is a valid fingerprint' or'not a valid fingerprint'.
  • the implementation of extracting the image data of the green channel of the fourth original image and generating the third fingerprint image can refer to the aforementioned related introduction about generating the first fingerprint image and the second fingerprint image.
  • the fingerprint living body detection device can extract the image data of the Gr channel, or extract the image data of the Gb channel, to obtain The third fingerprint image with an image size of w0*h0/4.
  • the fingerprint living detection device may also perform distortion correction on the image obtained after extraction to obtain a corrected third fingerprint image.
  • the fingerprint living body detection device may also crop the corrected image to obtain a clearer part in the middle according to a preset image size parameter, which is used as the third fingerprint image.
  • the image size of the third fingerprint image may be the same as or different from the image size of the first fingerprint image. If they are the same, just set a set of preset image size parameters.
  • the fingerprint living body detection device can also directly extract the fingerprint features of the fourth original image, and compare the extracted fingerprint features with the features corresponding to the stored images of each valid fingerprint to obtain the fingerprint Recognition results.
  • the fingerprint living body detection device can determine the final security verification result based on the two results. Exemplarily, if the fingerprint identification result indicates that the collected fingerprint is a valid fingerprint, and the fingerprint live detection result indicates that the collected fingerprint is a live fingerprint, it is determined that the security verification is passed. If the fingerprint recognition result indicates that the collected fingerprint is not a valid fingerprint, or the fingerprint live detection result indicates that the collected fingerprint is not a live fingerprint, it is determined that the security verification is not passed, and the fingerprint live detection device can prompt the user with text or voice.
  • the first fingerprint image and the second fingerprint image can be divided into regions, and then the multi-zone gray distribution feature vector can be determined according to the divided regions, so as to perform fingerprinting through the fingerprint living detection model.
  • Living body detection where the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image. Since the distribution of gray values in each area of a non-living fingerprint is different from that of a living fingerprint, the multi-division gray distribution feature vector can identify whether the fingerprint image has the gray distribution feature of the living fingerprint, and then determine whether the fingerprint is It is a living fingerprint with high security.
  • Fig. 10 is a schematic structural diagram of a fingerprint living body detection device 1000 provided by an embodiment of the present application.
  • the fingerprint living body detection device 1000 can be implemented as a part or all of a computer device by software, hardware, or a combination of the two. Please refer to FIG. 10, the device includes: a first obtaining module 1001, a first obtaining module 1002, a first obtaining module 1003, and a first obtaining module 1004.
  • the first acquisition module 1001 is used to acquire a first fingerprint image and a second fingerprint image.
  • the first fingerprint image and the second fingerprint image refer to images of the same fingerprint.
  • the first fingerprint image is a red channel image
  • the second fingerprint image is a blue image.
  • the partition module 1002 is used to partition the first fingerprint image and the second fingerprint image respectively, wherein the divided first fingerprint image includes a plurality of first estimated bright areas and a plurality of first estimated dark areas, The divided second fingerprint image includes multiple second predicted bright areas and multiple second predicted dark areas;
  • the first determining module 1003 is configured to determine the multi-zone gray scale according to multiple first predicted bright areas, multiple first predicted dark areas, multiple second predicted bright areas, and multiple second predicted dark areas Distribution feature vector;
  • the detection module 1004 is used to input the multi-division gray distribution feature vector into the fingerprint live detection model to obtain the fingerprint live detection result.
  • the fingerprint live detection model is obtained through supervised learning training based on the live fingerprint sample image and the non-live fingerprint sample image.
  • the first obtaining module 1001 includes:
  • the first acquiring unit is configured to acquire a first original image, where the first original image is an image of a fingerprint collected when the green light is off, and the red light and blue light are on;
  • the first extraction unit is configured to extract the image data of the red channel of the first original image to generate a first fingerprint image
  • the second extraction unit is used to extract the image data of the blue channel of the first original image to generate a second fingerprint image.
  • the first obtaining module 1001 includes:
  • the second acquisition unit is used to acquire a second original image and a third original image.
  • the second original image is an image of a fingerprint collected when the green light and blue light are off and the red light is on
  • the third original image is a green light and a red light. Fingerprint images collected with the blue light turned off and on;
  • the third extraction unit is used to extract the image data of the red channel of the second original image to generate the first fingerprint image
  • the fourth extraction unit is used to extract the image data of the blue channel of the third original image to generate a second fingerprint image.
  • the partition module 1002 includes:
  • the first determining unit is configured to determine the partition parameters according to the distribution law of the bright area and the dark area in the red channel image and the blue channel image of the living fingerprint, and the partition parameter includes a plurality of partition width ratios;
  • the partition unit is used to partition the first fingerprint image and the second fingerprint image according to multiple partition width ratios.
  • the image sizes of the first fingerprint image and the second fingerprint image are the same, and the first determining module 1003 includes:
  • the second determining unit is configured to determine the first grayscale feature vector according to the plurality of first estimated bright areas and the plurality of second estimated dark areas;
  • the third determining unit is configured to determine the second gray-scale feature vector according to the plurality of first estimated dark areas and the plurality of second estimated bright areas;
  • the generating unit is used to generate a multi-partition gray distribution feature vector according to the first gray feature vector and the second gray feature vector.
  • the second determining unit includes:
  • the first determining subunit is used to determine the number of pixels corresponding to each gray value in the plurality of first estimated bright areas and the plurality of second estimated dark areas;
  • the first generating subunit is configured to generate the first grayscale feature vector according to the number of pixels corresponding to each grayscale value in the plurality of first estimated bright areas and the plurality of second estimated dark areas.
  • the third determining unit includes:
  • the second determining subunit is used to determine the number of pixels corresponding to each gray value of the plurality of first estimated dark areas and the plurality of second estimated bright areas;
  • the second generation subunit is used to generate a second gray-scale feature vector according to the number of pixels corresponding to each gray-scale value in the plurality of first estimated dark areas and the plurality of second estimated bright areas.
  • the first determining module 1003 further includes:
  • the third acquiring unit is configured to acquire the gray value statistical range, which is determined according to the exposure parameter of the image capture device that captures the image of the fingerprint;
  • the second determining unit includes:
  • the third determining subunit is used to determine the number of pixels corresponding to each gray value within the gray value statistical range in the plurality of first estimated bright areas and the plurality of second estimated dark areas according to the gray value statistical range number;
  • the third generation subunit is used to generate the first gray scale according to the number of pixels corresponding to each gray value within the gray value statistical range in the plurality of first estimated bright areas and the plurality of second estimated dark areas Feature vector.
  • the generating unit includes:
  • the splicing subunit is used to splice the first gray-scale feature vector and the second gray-scale feature vector to obtain a multi-zone gray-scale distribution feature vector.
  • the device 1000 further includes:
  • the second acquisition module is used to acquire the live fingerprint sample image and the non-live fingerprint sample image.
  • the live fingerprint sample image includes the images of multiple live fingerprint samples collected when the green light is off, the red light and the blue light are on.
  • the non-live fingerprint Sample images include images of multiple non-living fingerprint samples collected when the green light is off, the red light and the blue light are on;
  • the training module is used to obtain a fingerprint live detection model through supervised learning training based on the live fingerprint sample image and the non-live fingerprint sample image.
  • the apparatus 1000 further includes:
  • the third acquisition module 1005 is configured to acquire a fourth original image, the fourth original image being an image of a fingerprint collected when the green light is on;
  • the recognition module 1006 is used to determine the fingerprint recognition result according to the fourth original image
  • the second determination module 1007 is used to determine the security verification result according to the fingerprint identification result and the fingerprint biometric detection result.
  • the first fingerprint image and the second fingerprint image can be divided into regions, and then the multi-division gray distribution feature vector can be determined according to the divided regions, so as to perform fingerprint living body detection through the fingerprint living body detection model. Detection, where the first fingerprint image is a red channel image, and the second fingerprint image is a blue channel image. Since the gray value distribution of the bright and dark areas of the two channel images of non-living fingerprints is different from that of living fingerprints, the multi-partition gray distribution feature vector can identify whether the fingerprint image has the grays of living fingerprints. Degree distribution characteristics, and then determine whether the fingerprint is a living fingerprint, and the security is relatively high.
  • the fingerprint biometric detection device detects fingerprints
  • only the division of the above-mentioned functional modules is used as an example for illustration.
  • the above-mentioned functions can be allocated by different functional modules as required. That is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above.
  • the fingerprint living body detection device provided in the above embodiment and the fingerprint living body detection method embodiment belong to the same concept. For the specific implementation process, please refer to the method embodiment, which will not be repeated here.
  • FIG. 12 is a structural block diagram of a computer device 1200 provided by an embodiment of the present application.
  • the computer device 1200 may be a fingerprint living detection device that can perform security verification by collecting fingerprints, such as a smart phone, a tablet computer, a notebook computer, or a desktop computer.
  • the computer device 1200 includes a processor 1201 and a memory 1202.
  • the processor 1201 may include one or more processing cores, such as a 4-core processor, a 12-core processor, and so on.
  • the processor 1201 may adopt at least one hardware form among DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array, Programmable Logic Array). accomplish.
  • the processor 1201 may also include a main processor and a coprocessor.
  • the main processor is a processor used to process data in the awake state, also called a CPU (Central Processing Unit, central processing unit); the coprocessor is A low-power processor used to process data in the standby state.
  • the processor 1201 may be integrated with a GPU (Graphics Processing Unit, image processor), and the GPU is used for rendering and drawing content that needs to be displayed on the display screen.
  • the processor 1201 may further include an AI (Artificial Intelligence) processor, and the AI processor is used to process computing operations related to machine learning.
  • AI Artificial Intelligence
  • the memory 1202 may include one or more computer-readable storage media, which may be non-transitory.
  • the memory 1202 may also include high-speed random access memory and non-volatile memory, such as one or more magnetic disk storage devices and flash memory storage devices.
  • the non-transitory computer-readable storage medium in the memory 1202 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 1201 to implement the method provided in the embodiment of the present application. Fingerprint live detection method.
  • the computer device 1200 may optionally further include: a peripheral device interface 1203 and at least one peripheral device.
  • the processor 1201, the memory 1202, and the peripheral device interface 1203 may be connected by a bus or a signal line.
  • Each peripheral device can be connected to the peripheral device interface 1203 through a bus, a signal line, or a circuit board.
  • the peripheral device includes: at least one of a radio frequency circuit 1204, a display screen 1205, a camera component 1206, an audio circuit 1207, a positioning component 1208, and a power supply 1209.
  • the peripheral device interface 1203 may be used to connect at least one peripheral device related to I/O (Input/Output) to the processor 1201 and the memory 1202.
  • the processor 1201, the memory 1202, and the peripheral device interface 1203 are integrated on the same chip or circuit board; in some other embodiments, any one of the processor 1201, the memory 1202, and the peripheral device interface 1203 or The two can be implemented on a separate chip or circuit board, which is not limited in this embodiment.
  • the radio frequency circuit 1204 is used for receiving and transmitting RF (Radio Frequency, radio frequency) signals, also called electromagnetic signals.
  • the radio frequency circuit 1204 communicates with a communication network and other communication devices through electromagnetic signals.
  • the radio frequency circuit 1204 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals into electrical signals.
  • the radio frequency circuit 1204 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user identity module card, and so on.
  • the radio frequency circuit 1204 can communicate with other computer devices through at least one wireless communication protocol.
  • the wireless communication protocol includes but is not limited to: World Wide Web, Metropolitan Area Network, Intranet, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area network and/or WiFi (Wireless Fidelity, wireless fidelity) network.
  • the radio frequency circuit 1204 may also include a circuit related to NFC (Near Field Communication), which is not limited in the embodiment of the present application.
  • the display screen 1205 is used to display a UI (User Interface, user interface).
  • the UI can include graphics, text, icons, videos, and any combination thereof.
  • the display screen 1205 also has the ability to collect touch signals on or above the surface of the display screen 1205.
  • the touch signal may be input to the processor 1201 as a control signal for processing.
  • the display screen 1205 may also be used to provide virtual buttons and/or virtual keyboards, also called soft buttons and/or soft keyboards.
  • the display screen 1205 may be a flexible display screen, which is arranged on the curved surface or the folding surface of the computer device 1200.
  • the display screen 1205 can also be set as a non-rectangular irregular pattern, that is, a special-shaped screen.
  • the display screen 1205 may be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).
  • the camera assembly 1206 is used to capture images or videos.
  • the camera assembly 1206 includes a front camera and a rear camera.
  • the front camera is set on the front panel of the computer device, and the rear camera is set on the back of the computer device.
  • the camera assembly 1206 may also include a flash.
  • the flash can be a single-color flash or a dual-color flash. Dual color temperature flash refers to a combination of warm light flash and cold light flash, which can be used for light compensation under different color temperatures.
  • the audio circuit 1207 may include a microphone and a speaker.
  • the microphone is used to collect sound waves of the user and the environment, and convert the sound waves into electrical signals and input them to the processor 1201 for processing, or input to the radio frequency circuit 1204 to implement voice communication.
  • the microphone can also be an array microphone or an omnidirectional acquisition microphone.
  • the speaker is used to convert the electrical signal from the processor 1201 or the radio frequency circuit 1204 into sound waves.
  • the speaker can be a traditional thin-film speaker or a piezoelectric ceramic speaker.
  • the audio circuit 1207 may also include a headphone jack.
  • the positioning component 1208 is used to locate the current geographic location of the computer device 1200 to implement navigation or LBS (Location Based Service, location-based service).
  • the positioning component 1208 may be a positioning component based on the GPS (Global Positioning System, Global Positioning System) of the United States, the Beidou system of China, or the Galileo system of Russia.
  • the power supply 1209 is used to supply power to various components in the computer device 1200.
  • the power source 1209 may be alternating current, direct current, disposable batteries, or rechargeable batteries.
  • the rechargeable battery may be a wired rechargeable battery or a wireless rechargeable battery.
  • a wired rechargeable battery is a battery charged through a wired line
  • a wireless rechargeable battery is a battery charged through a wireless coil.
  • the rechargeable battery can also be used to support fast charging technology.
  • the computer device 1200 further includes one or more sensors 1210.
  • the one or more sensors 1210 include, but are not limited to: an acceleration sensor 1211, a gyroscope sensor 1212, a pressure sensor 1213, a fingerprint sensor 1214, an optical sensor 1215, and a proximity sensor 1216.
  • the acceleration sensor 1211 can detect the magnitude of acceleration on the three coordinate axes of the coordinate system established by the computer device 1200.
  • the acceleration sensor 1211 can be used to detect the components of gravitational acceleration on three coordinate axes.
  • the processor 1201 may control the display screen 1205 to display the user interface in a horizontal view or a vertical view according to the gravity acceleration signal collected by the acceleration sensor 1211.
  • the acceleration sensor 1211 may also be used for the collection of game or user motion data.
  • the gyroscope sensor 1212 can detect the body direction and rotation angle of the computer device 1200, and the gyroscope sensor 1212 can cooperate with the acceleration sensor 1211 to collect the user's 3D actions on the computer device 1200. Based on the data collected by the gyroscope sensor 1212, the processor 1201 can implement the following functions: motion sensing (such as changing the UI according to the user's tilt operation), image stabilization during shooting, game control, and inertial navigation.
  • motion sensing such as changing the UI according to the user's tilt operation
  • image stabilization during shooting
  • game control game control
  • inertial navigation inertial navigation
  • the pressure sensor 1213 may be arranged on the side frame of the computer device 1200 and/or the lower layer of the display screen 1205.
  • the processor 1201 performs left and right hand recognition or quick operation according to the holding signal collected by the pressure sensor 1213.
  • the processor 1201 controls the operability controls on the UI interface according to the user's pressure operation on the display screen 1205.
  • the operability control includes at least one of a button control, a scroll bar control, an icon control, and a menu control.
  • the fingerprint sensor 1214 is used to collect the user's fingerprint.
  • the processor 1201 identifies the user's identity according to the fingerprint collected by the fingerprint sensor 1214, or the fingerprint sensor 1214 identifies the user's identity according to the collected fingerprint.
  • the processor 1201 authorizes the user to perform related sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, paying, and changing settings.
  • the fingerprint sensor 1214 may be provided on the front, back or side of the computer device 1200. When the computer device 1200 is provided with a physical button or a manufacturer logo, the fingerprint sensor 1214 can be integrated with the physical button or the manufacturer logo.
  • the optical sensor 1215 is used to collect the ambient light intensity.
  • the processor 1201 may control the display brightness of the display screen 1205 according to the ambient light intensity collected by the optical sensor 1215. Specifically, when the ambient light intensity is high, the display brightness of the display screen 1205 is increased; when the ambient light intensity is low, the display brightness of the display screen 1205 is decreased.
  • the processor 1201 may also dynamically adjust the shooting parameters of the camera assembly 1206 according to the ambient light intensity collected by the optical sensor 1215.
  • the proximity sensor 1216 also called a distance sensor, is usually arranged on the front panel of the computer device 1200.
  • the proximity sensor 1216 is used to collect the distance between the user and the front of the computer device 1200.
  • the processor 1201 controls the display screen 1205 to switch from the bright screen state to the off screen state; when the proximity sensor 1216 detects When the distance between the user and the front of the computer device 1200 gradually increases, the processor 1201 controls the display screen 1205 to switch from the rest screen state to the bright screen state.
  • FIG. 12 does not constitute a limitation on the computer device 1200, and may include more or fewer components than those shown in the figure, or combine certain components, or adopt different component arrangements.
  • a computer-readable storage medium is also provided, and a computer program is stored in the storage medium.
  • the computer program when executed by a processor, implements the steps of the fingerprint living detection method in the above-mentioned embodiments.
  • the computer-readable storage medium may be ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
  • the computer-readable storage medium mentioned in the embodiment of the present application may be a non-volatile storage medium, in other words, it may be a non-transitory storage medium.
  • a computer program product containing instructions is also provided, which when run on a computer, causes the computer to execute the steps of the fingerprint living detection method described above.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Collating Specific Patterns (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
  • Image Input (AREA)
  • Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)

Abstract

一种指纹活体检测方法、设备（100）及存储介质，属于安全技术领域。可以对第一指纹图像和第二指纹图像进行区域划分，进而根据划分的区域来确定多分区灰度分布特征向量，以通过指纹活体检测模型进行指纹活体检测，其中，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像。由于非活体指纹的这两个通道图像各个区域灰度值的分布规律与活体指纹的不同，因此，通过多分区灰度分布特征向量可以识别出指纹图像是否存在活体指纹的灰度分布特征，进而确定出该指纹是否为活体指纹，安全性较高。

Description

指纹活体检测方法、设备及存储介质
本申请实施例要求于2020年4月30日提交的申请号为202010364403.X、发明名称为“指纹活体检测方法、设备及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请实施例中。
技术领域
本申请实施例涉及安全技术领域，特别涉及一种指纹活体检测方法、设备及存储介质。
背景技术
目前，指纹识别在很多场景中有很重要的应用，例如考勤、门禁、电脑解锁、手机解锁、支付等场景，指纹识别主要是提取指纹图像的特征，例如纹路、断点、交叉点等，来确定是否是有效指纹。而当前指纹伪造技术也随之发展，例如用硅胶做成指纹膜来伪造他人指纹，成功通过指纹识别来进行一些非法行为。为了保证指纹识别的安全性，需要在指纹识别的基础上增加指纹活体检测，以有效防止使用指纹膜来冒充他人指纹。
相关技术中，可以根据皮肤温度或皮肤电容值来检测是否是指纹活体，而由于皮肤温度和皮肤电容值通常是相对稳定的，恶意分子可以很容易地伪造与皮肤温度或电容值相同的指纹膜，进而成功通过指纹活体检测，因此，这种方法的安全性仍较低。
发明内容
本申请实施例提供了一种指纹活体检测方法、设备及存储介质，可以提高指纹活体检测方法的安全性。所述技术方案如下：
一方面，提供了一种指纹活体检测方法，所述方法包括：
获取第一指纹图像和第二指纹图像，所述第一指纹图像和所述第二指纹图像是指同一指纹的图像，所述第一指纹图像为红色通道图像，所述第二指纹图像为蓝色通道图像；
对所述第一指纹图像和所述第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区；
根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量；
将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果，所述指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
可选地，所述获取第一指纹图像和第二指纹图像，包括：
获取第一原始图像，所述第一原始图像为绿灯关闭、红灯和蓝灯开启的状态下采集的所述指纹的图像；
提取所述第一原始图像的红色通道的图像数据，生成所述第一指纹图像；
提取所述第一原始图像的蓝色通道的图像数据，生成所述第二指纹图像。
可选地，所述获取第一指纹图像和第二指纹图像，包括：
获取第二原始图像和第三原始图像，所述第二原始图像为绿灯和蓝灯关闭、红灯开启的状态下采集的所述指纹的图像，所述第三原始图像为绿灯和红灯关闭、蓝灯开启的状态下采集的所述指纹的图像；
提取所述第二原始图像的红色通道的图像数据，生成所述第一指纹图像；
提取所述第三原始图像的蓝色通道的图像数据，生成所述第二指纹图像。
可选地，所述对所述第一指纹图像和所述第二指纹图像分别进行区域划分，包括：
根据活体指纹的红色通道图像和蓝色通道图像中明区和暗区的分布规律，确定分区参数，所述分区参数包括多个分区宽度比例；
根据所述多个分区宽度比例，对所述第一指纹图像和所述第二指纹图像分别进行区域划分。
可选地，所述第一指纹图像和所述第二指纹图像的图像大小相同，所述根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量，包括：
根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量；
根据所述多个第一预估暗区和所述多个第二预估明区，确定第二灰度特征 向量；
根据所述第一灰度特征向量和所述第二灰度特征向量，生成所述多分区灰度分布特征向量。
可选地，所述根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量，包括：
确定所述多个第一预估明区和所述多个第二预估暗区内各个灰度值分别对应的像素点个数；
根据所述多个第一预估明区和所述多个第二预估暗区内各个灰度值分别对应的像素点个数，生成所述第一灰度特征向量。
可选地，所述根据所述多个第一预估暗区和所述多个第二预估明区，确定第二灰度特征向量，包括：
确定所述多个第一预估暗区和所述多个第二预估明区内各个灰度值分别对应的像素点个数；
根据所述多个第一预估暗区和所述多个第二预估明区内各个灰度值分别对应的像素点个数，生成所述第二灰度特征向量。
可选地，所述根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量之前，还包括：
获取灰度值统计范围，所述灰度值统计范围根据采集所述指纹的图像的图像采集设备的曝光参数确定；
所述根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量，包括：
根据所述灰度值统计范围，确定所述多个第一预估明区和所述多个第二预估暗区内处于所述灰度值统计范围的各个灰度值对应的像素点个数；
根据所述多个第一预估明区和所述多个第二预估暗区内处于所述灰度值统计范围的各个灰度值对应的像素点个数，生成所述第一灰度特征向量。
可选地，所述根据所述第一灰度特征向量和所述第二灰度特征向量，生成所述多分区灰度分布特征向量，包括：
将所述第一灰度特征向量和所述第二灰度特征向量进行拼接，得到所述多分区灰度分布特征向量。
可选地，所述将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果之前，还包括：
获取所述活体指纹样本图像和所述非活体指纹样本图像，所述活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个活体指纹样本的图像，所述非活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个非活体指纹样本的图像；
根据所述活体指纹样本图像和所述非活体指纹样本图像，通过监督学习训练得到所述指纹活体检测模型。
可选地，所述方法还包括：
获取第四原始图像，所述第四原始图像为绿灯开启的状态下采集的所述指纹的图像；
根据所述第四原始图像，确定指纹识别结果；
根据所述指纹识别结果和所述指纹活体检测结果，确定安全验证结果。
另一方面，提供了一种指纹活体检测装置，所述装置包括：
第一获取模块，用于获取第一指纹图像和第二指纹图像，所述第一指纹图像和所述第二指纹图像是指同一指纹的图像，所述第一指纹图像为红色通道图像，所述第二指纹图像为蓝色通道图像；
分区模块，用于对所述第一指纹图像和所述第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区；
第一确定模块，用于根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量；
检测模块，用于将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果，所述指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
可选地，所述第一获取模块包括：
第一获取单元，用于获取第一原始图像，所述第一原始图像为绿灯关闭、红灯和蓝灯开启的状态下采集的所述指纹的图像；
第一提取单元，用于提取所述第一原始图像的红色通道的图像数据，生成所述第一指纹图像；
第二提取单元，用于提取所述第一原始图像的蓝色通道的图像数据，生成 所述第二指纹图像。
可选地，所述第一获取模块包括：
第二获取单元，用于获取第二原始图像和第三原始图像，所述第二原始图像为绿灯和蓝灯关闭、红灯开启的状态下采集的所述指纹的图像，所述第三原始图像为绿灯和红灯关闭、蓝灯开启的状态下采集的所述指纹的图像；
第三提取单元，用于提取所述第二原始图像的红色通道的图像数据，生成所述第一指纹图像；
第四提取单元，用于提取所述第三原始图像的蓝色通道的图像数据，生成所述第二指纹图像。
可选地，所述分区模块包括：
第一确定单元，用于根据活体指纹的红色通道图像和蓝色通道图像中明区和暗区的分布规律，确定分区参数，所述分区参数包括多个分区宽度比例；
分区单元，用于根据所述多个分区宽度比例，对所述第一指纹图像和所述第二指纹图像分别进行区域划分。
可选地，所述第一指纹图像和所述第二指纹图像的图像大小相同，所述第一确定模块包括：
第二确定单元，用于根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量；
第三确定单元，用于根据所述多个第一预估暗区和所述多个第二预估明区，确定第二灰度特征向量；
生成单元，用于根据所述第一灰度特征向量和所述第二灰度特征向量，生成所述多分区灰度分布特征向量。
可选地，所述第二确定单元包括：
第一确定子单元，用于确定所述多个第一预估明区和所述多个第二预估暗区内各个灰度值分别对应的像素点个数；
第一生成子单元，用于根据所述多个第一预估明区和所述多个第二预估暗区内各个灰度值分别对应的像素点个数，生成所述第一灰度特征向量。
可选地，所述第三确定单元包括：
第二确定子单元，用于确定所述多个第一预估暗区和所述多个第二预估明区内各个灰度值分别对应的像素点个数；
第二生成子单元，用于根据所述多个第一预估暗区和所述多个第二预估明 区内各个灰度值分别对应的像素点个数，生成所述第二灰度特征向量。
可选地，所述第一确定模块还包括：
第三获取单元，用于获取灰度值统计范围，所述灰度值统计范围根据采集所述指纹的图像的图像采集设备的曝光参数确定；
所述第二确定单元包括：
第三确定子单元，用于根据所述灰度值统计范围，确定所述多个第一预估明区和所述多个第二预估暗区内处于所述灰度值统计范围的各个灰度值对应的像素点个数；
第三生成子单元，用于根据所述多个第一预估明区和所述多个第二预估暗区内处于所述灰度值统计范围的各个灰度值对应的像素点个数，生成所述第一灰度特征向量。
可选地，所述生成单元包括：
拼接子单元，用于将所述第一灰度特征向量和所述第二灰度特征向量进行拼接，得到所述多分区灰度分布特征向量。
可选地，所述装置还包括：
第二获取模块，用于获取所述活体指纹样本图像和所述非活体指纹样本图像，所述活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个活体指纹样本的图像，所述非活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个非活体指纹样本的图像；
训练模块，用于根据所述活体指纹样本图像和所述非活体指纹样本图像，通过监督学习训练得到所述指纹活体检测模型。
可选地，所述装置还包括：
第三获取模块，用于获取第四原始图像，所述第四原始图像为绿灯开启的状态下采集的所述指纹的图像；
识别模块，用于根据所述第四原始图像，确定指纹识别结果；
第二确定模块，用于根据所述指纹识别结果和所述指纹活体检测结果，确定安全验证结果。
另一方面，提供了一种指纹活体检测设备，所述指纹活体检测设备包括图像采集器和处理器，所述图像采集器包括补光灯和图像传感器；
所述图像采集器，用于采集第一原始图像；
所述补光灯，用于在采集所述第一原始图像的过程中，关闭绿灯、开启红灯和蓝灯为所述图像传感器进行补光；
所述处理器，用于对所述第一原始图像进行处理，得到第一指纹图像和第二指纹图像，所述第一指纹图像为红色通道图像，所述第二指纹图像为蓝色通道图像；对所述第一指纹图像和所述第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区；根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量；将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果，所述指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
可选地，所述图像采集器，还用于采集第四原始图像；
所述补光灯，还用于在采集所述第四原始图像的过程中，开启绿灯为所述图像传感器进行补光；
所述处理器，还用于根据所述第四原始图像，确定指纹识别结果，以及根据所述指纹识别结果和所述指纹活体检测结果，确定安全验证结果。
另一方面，提供了一种计算机可读存储介质，所述存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现上述所述指纹活体检测方法的步骤。
另一方面，提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述所述的指纹活体检测方法的步骤。
本申请实施例提供的技术方案至少可以带来以下有益效果：
在本申请实施例中，可以对第一指纹图像和第二指纹图像进行区域划分，进而根据划分的区域来确定多分区灰度分布特征向量，以通过指纹活体检测模型进行指纹活体检测，其中，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像。由于非活体指纹的这两个通道图像的各个区域灰度值的分布规律与活体指纹的不同，因此，通过多分区灰度分布特征向量可以识别出指纹图像是否存在活体指纹的灰度分布特征，进而确定出该指纹是否为活体指纹，安 全性较高。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中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。
图1是本申请实施例提供的一种指纹活体检测设备的结构示意图；
图2是本申请实施例提供的一种指纹活体检测方法的流程图；
图3是本申请实施例采用的一种图像传感器的示意图；
图4是本申请实施例提供的一种提取出的红色通道的图像数据的示意图；
图5是本申请实施例提供的一种第一指纹图像的示意图；
图6是本申请实施例提供的一种提取出的蓝色通道的图像数据的示意图；
图7是本申请实施例提供的一种第二指纹图像的示意图；
图8是本申请实施例提供的一种确定图像各个区域的示意图；
图9是本申请实施例提供的另一种指纹活体检测方法的流程图；
图10是本申请实施例提供的一种指纹活体检测装置的结构示意图；
图11是本申请实施例提供的另一种指纹活体检测装置的结构示意图；
图12是本申请实施例提供的一种计算机设备的结构示意图。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚，下面将结合附图对本申请实施方式作进一步地详细描述。
目前，指纹识别在很多场景中有很重要的应用，例如考勤、门禁、电脑解锁、手机解锁、支付等场景。而当前指纹伪造技术也随之发展，例如用硅胶做成指纹膜来伪造他人指纹，成功通过指纹识别来进行一些非法行为。为了保证指纹识别的安全性，需要在指纹识别的基础上增加指纹活体检测，以有效防止使用指纹膜来冒充他人指纹。对于各种场景中所采集的指纹图像，可以根据本申请实施例提供的方法来进行指纹活体检测，得到指纹活体检测结果，并在此基础上对采集的指纹图像进行指纹识别，得到指纹识别结果，在指纹识别和指纹活体检测结果均通过的情况下，确定安全验证通过。
图1是本申请实施例提供的一种指纹活体检测设备100的结构示意图。参见图1，该指纹活体检测设备100包括图像采集器101、处理器102，图像采集器101包括LED补光灯、镜头和图像传感器，LED补光灯包括绿灯、红灯和蓝灯。
图像采集器101用于接收光信号或者感知电容，将接收到的光信号或感知到的电容转换为电信号发送给处理器102，处理器102用于根据接收到的电信号，确定当前光强变化量或者电容值，在光强变化量指示当前光线明显变暗或者电容值为活体手指的电容值时，向图像采集器101发送用于采集指纹图像的控制指令，也即是唤醒图像采集器101进行指纹采集。
图像采集器101在接收到该控制指令时，控制LED补光灯根据本申请实施例提供的指纹图像采集方法来开启绿灯，或者开启红灯和蓝灯来为图像传感器进行补光，通过镜头和图像传感器采集原始图像，并将采集的原始图像发送给处理器102，处理器102中部署有指纹活体检测模型，可以根据本申请实施例提供的指纹活体检测方法对原始图像进行处理，得到指纹活体检测结果。其中，指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
在本申请实施例中，图像采集器101可以采集第一原始图像，LED补光灯可以在采集第一原始图像的过程中，关闭绿灯、开启红灯和蓝灯为图像传感器进行补光，处理器102可以对第一原始图像进行处理，得到第一指纹图像和第二指纹图像，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像。之后，处理器102可以对第一指纹图像和第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区。处理器102还可以根据该多个第一预估明区、多个第一预估暗区、多个第二预估明区和多个第二预估暗区，确定多分区灰度分布特征向量，然后，将该多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果。
可选地，图像采集器101还可以采集第四原始图像，LED补光灯可以在采集第四原始图像的过程中，开启绿灯为图像传感器进行补光，处理器102可以根据第四原始图像，确定指纹识别结果，并根据指纹识别结果和指纹活体检测结果，确定安全验证结果。
在本申请实施例中，指纹活体检测设备100还可以包括存储器103，存储器103用于存储本申请实施例所需的数据以及计算机程序，例如存储有多个有效指纹的图像，用于指纹识别。处理器102可以执行存储器103中的计算机程序，以及对采集的原始图像与存储的多个有效指纹的图像进行比对，以进行指纹识别，得到指纹识别结果，最终还可以根据指纹活体检测结果和指纹识别结果来确定安全验证结果。
需要说明的是，LED补光灯为图像传感器补光来采集各个原始图像，以及处理器102对各个原始图像进行处理，得到指纹活体检测结果或者指纹识别结果的具体实现方式，可以参照下述图2方法实施例的相关介绍。
可选地，在本申请实施例中，LED补光灯包括红灯、绿灯和蓝灯，这三种灯分别对应的红光、绿光和蓝光的波长范围可以根据图像传感器的QE(Quantum Efficiency，量子效率)曲线来确定。例如，在仅开启红光的状态下，当图像传感器的QE曲线的峰值达到最大时，可以将该峰值位置对应的波长作为红光的中心波长，从中心波长分别向左和向右取一段波长范围，作为红光的波长范围。在本申请实施例中，根据该方法确定的绿光的波长范围可以为500nm-550nm，红光的波长范围可以为600nm-650nm，蓝光的波长范围可以为420nm-480nm，这几个范围仅作为一种示例，实际应用中，不同的设备最终确定的波长范围可能存在不同。
需要说明的是，图1所示的指纹活体检测设备100可以是指前端的图像采集设备，例如摄像头，也即是，本申请实施例所提供的指纹活体检测方法可以应用于前端的图像采集设备。在一种可能的实现方式中，上述指纹活体检测设备100的功能还可以通过前端设备和后端设备两个独立的设备来实现，其中，后台设备可以与前端设备通过有线或者无线连接以进行数据传输。在这种情况下，前端设备可以包括前述实施例中的图像采集器101，以进行原始图像的采集，之后，前端设备可以将采集到的原始图像发送至后端设备，后端设备可以通过本申请实施例提供的指纹活体检测方法对采集的原始图像进行检测以及识别，最终确定安全验证结果，之后，可以将安全验证结果发送给前端设备或者其他设备，以文字形式或者声音形式提示用户安全验证通过或者不通过。
综上所述，在本申请实施例中，可以对第一指纹图像和第二指纹图像进行区域划分，进而根据划分区域来确定多分区灰度分布特征向量，以通过指纹活体检测模型进行指纹活体检测，其中，第一指纹图像为红色通道图像，第二指 纹图像为蓝色通道图像。由于非活体指纹的这两个通道图像的明区和暗区的灰度值的分布规律与活体指纹的不同，因此，通过多分区灰度分布特征向量可以识别出指纹图像是否存在活体指纹的灰度分布特征，进而确定出该指纹是否为活体指纹，安全性较高。
需要说明的是：上述实施例提供的指纹活体检测设备在检测指纹时，仅以上述各功能模块或者器件的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块或者器件完成，即将设备的内部结构划分成不同的功能模块或者器件，以完成以上描述的全部或者部分功能。另外，上述实施例提供的指纹活体检测设备与下述指纹活体检测方法实施例属于同一构思，其具体实现过程详见下述方法实施例，这里不再赘述。
接下来对本申请实施例提供的指纹活体检测方法进行详细的解释说明。
图2是本申请实施例提供的一种指纹活体检测方法的流程图，以该方法应用于图1所示的指纹活体检测设备为例进行说明。请参考图2，该方法包括如下步骤。
步骤201：获取第一指纹图像和第二指纹图像，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像。
由于活体指纹的红色通道图像和蓝色通道图像均是明区和暗区交替分布的，且像素点的灰度值分布较为规律，而非活体指纹的这两个通道图像的像素点的灰度值分布较为混乱，也即是非活体指纹对应的灰度值的分布规律与活体指纹的不同，因此，可以根据灰度值的分布规律来进行指纹活体检测。也即是，在本申请实施例中，获取的第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像，第一指纹图像和第二指纹图像是指同一指纹的图像。
由前述对图1的相关介绍可知，采集指纹图像之前，需要先唤醒图像采集器，可以根据光强变化或者电容值来确定当前是否检测到采集对象，例如手指，当检测到采集对象时，可以唤醒图像采集器，开始采集指纹图像。
由于本申请实施例需要根据红色通道图像和蓝色通道图像来进行指纹活体检测，基于此，本申请实施例提供了两种图像采集方法，也即是提供了两种获取第一指纹图像和第二指纹图像的实现方式，接下来对这两种实现方式进行介绍。
第一种实现方式，指纹活体检测设备可以获取第一原始图像，第一原始图 像为绿灯关闭、红灯和蓝灯开启的状态下采集的指纹的图像，之后，指纹活体检测设备可以提取第一原始图像的红色通道的图像数据，生成第一指纹图像，提取第一原始图像的蓝色通道的图像数据，生成第二指纹图像。
在本申请实施例中，图像传感器包括红色通道、蓝色通道和绿色通道，例如图像传感器可以是如图3所示的拜尔模式的彩色图像传感器，其中，R为红色(R)通道，Gr和Gb均为绿色(G)通道，B为蓝色(B)通道。
在唤醒图像采集器之后，可以同时开启红灯和蓝灯，通过镜头和图像传感器采集第一原始图像，也即是，第一原始图像为绿灯关闭、红灯和蓝灯开启的状态下采集的该指纹的图像。之后，图像采集器可以将采集的原始图像发送给处理器。处理器可以提取第一原始图像的红色通道的图像数据，生成第一指纹图像，并提取蓝色通道的图像数据，生成第二指纹图像。
示例性地，以图3所示的图像传感器为例，假设采集对象为活体指纹，第一原始图像的大小为w0*h0，w0和h0分别表示第一原始图像的宽度和高度，也可以分别表示宽度方向和高度方向的像素点的个数。处理器可以将R通道的图像数据提取出来，得到如图4所示的红色通道图像数据，根据该红色通道图像数据生成如图5所示的第一指纹图像。处理器可以将B通道的图像数据提取出来，得到如图6所示的蓝色通道图像数据，根据该蓝色通道图像数据生成如图7所示的第二指纹图像，第一指纹图像和第二指纹图像的图像大小均为w0*h0/4。
第二种实现方式，指纹活体检测设备可以获取第二原始图像和第三原始图像，第二原始图像为绿灯和蓝灯关闭、红灯开启的状态下采集的该指纹的图像，第三原始图像为绿灯和红灯关闭、蓝灯开启的状态下采集的该指纹的图像。之后，指纹活体检测设备可以提取第二原始图像的红色通道的图像数据，生成第一指纹图像，提取第三原始图像的蓝色通道的图像数据，生成第二指纹图像。
可见，第二种实现方式是通过仅开启红灯采集第一幅原始图像，再仅开启蓝灯采集第二幅原始图像，进而提取第一幅原始图像的红色通道的图像数据，来生成第一指纹图像，提取第二幅原始图像的蓝色通道的图像数据，来生成第二指纹图像。其中，提取图像数据的方法可以参照第一种实现方式中的相关介绍，这里不再赘述。
需要说明的是，在第二种实现方式中，通过仅开启红灯和仅开启蓝灯的方式来采集两幅原始图像，这样可以避免同时开启红灯和蓝灯的情况下，红灯对 蓝色通道的影响，以及蓝灯对红色通道的影响。而在第一种实现方式中，通过绿灯关闭、红灯和蓝灯同时开启的方法来采集一幅原始图像，相较于第二种实现方式，可以减少图像采集的时间。
可选地，由于采集的指纹图像的边缘可能存在畸变，因此，指纹活体检测设备可以在提取出红色通道的图像数据和蓝色通道的图像数据之后，还可以对红色通道图像和蓝色通道图像进行畸变校正。另外，指纹活体检测设备还可以根据预设图像大小参数，对红色通道图像和蓝色通道图像分别进行裁剪，以将边缘部分不清晰的图像去掉，并得到图像大小相同的第一指纹图像和第二指纹图像。
步骤202：对第一指纹图像和第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区。
在本申请实施例中，指纹活体检测设备可以根据活体指纹的红色通道图像和蓝色通道图像中明区和暗区的分布规律，对第一指纹图像和第二指纹图像分别进行区域划分，以将第一指纹图像划分为多个第一预估明区和多个第一预估暗区，将第二指纹图像划分为多个第二预估明区和多个第二预估暗区。
由图5所示的第一指纹图像和图7所示的第二指纹图像可以看出，红色通道图像沿图像高度方向从上到下的明区和暗区的分布规律为：明、暗、明、暗、明，蓝色通道图像沿图像高度方向从上到下的明区和暗区的分布规律为：暗、明、暗、明、暗，也即是红色通道图像和蓝色通道图像均呈现明暗交替的分布规律，指纹活体检测设备可以根据该分布规律，对第一指纹图像和第二指纹图像进行区域划分。
在本申请实施例中，指纹活体检测设备可以根据活体指纹的红色通道图像和蓝色通道图像中明区和暗区的分布规律，确定分区参数，该分区参数包括多个分区宽度比例，之后，根据多个分区宽度比例，对第一指纹图像和第二指纹图像分别进行区域划分。
需要说明的是，该分区参数也可以是事先根据各个区域的分布规律设置的参数，例如，可以统计多幅活体指纹的红色通道图像和蓝色通道图像的明暗分布情况，根据明暗分布情况确定活体指纹的红色通道图像和蓝色通道图像每个图像中明区和暗区的总数量，以及从上到下处于同一位置上的明区和暗区占图像高度的比例，以此来得到每个分区宽度比例，并在指纹活体检测设备中设置 该分区参数。
示例性地，图8是本申请实施例中提供的一种确定图像各个区域的示意图，参见图8，图8中的左上图为活体指纹的红色通道图像，即第一指纹图像，右上图为活体指纹的蓝色通道图像，即第二指纹图像，通过红色通道图像和蓝色通道图像均呈现明暗交替的分布规律，可以将红色通道图像和蓝色通道图像从上至下均分为5个区域，红色通道图像包括的5个区域从上到下依次为明区、暗区、明区、暗区和明区，蓝色通道图像包括的5个区域从上到下依次为暗区、明区、暗区、明区和暗区，且红色通道图像的5个区域与蓝色通道的5个区域一一对应，也即两幅图像中对应区域的宽度比例相同，不同的是，红色通道图像中的明区，在蓝色通道图像中相同高度范围上对应的区域为暗区。综合红色通道图像和蓝色通道图像的明暗区分布，可以得出不管蓝色通道图像还是红色通道图像都可以以相同的划分规则划分为如图8所示的五个区域，分别为区域1、区域2、区域3、区域4和区域5，5个区域对应的分区宽度比例如图8所示，例如从上至下可以分别为20/66、10/66、15/66、9/66和12/66，还可以是其他的比例。
指纹活体检测设备可以按照图8所示的分区宽度比例分别对第一指纹图像和第二指纹图像进行分区，划分后的第一指纹图像包括3个第一预估明区和2个第一预估暗区，其中，3个第一预估明区对应图8中的区域1、区域3和区域5，2个第一预估暗区对应图8中的区域2和区域4，同理，划分后的第二指纹图像包括2个第二预估明区和3个第二预估暗区，其中，2个第二预估明区对应图8中的区域2和区域4，3个第二预估暗区对应图8中的区域1、区域3和区域5。
可选地，在第一指纹图像和第二指纹图像的图像大小为预设图像大小的情况下，分区参数可以包括多个宽度值。示例性地，假设预设图像大小的高度为66像素，图8所示的5个区域的宽度值从上之下依次可以为20、10、15、9和12像素。指纹活体检测设备可以根据该多个宽度值，将第一指纹图像和第二指纹图像进行区域划分。
步骤203：根据该多个第一预估明区、多个第一预估暗区、多个第二预估明区和多个第二预估暗区，确定多分区灰度分布特征向量。
在本申请实施例中，第一指纹图像和第二指纹图像的图像大小相同，在对第一指纹图像和第二指纹图像分别进行区域划分之后，指纹活体检测设备可以 根据统计划分出的各个区域的灰度值，来确定多分区灰度分布特征向量。
在本申请实施例中，指纹活体检测设备可以根据该多个第一预估明区和该多个第二预估暗区，确定第一灰度特征向量，根据该多个第一预估暗区和该多个第二预估明区，确定第二灰度特征向量。之后，指纹活体检测设备可以根据第一灰度特征向量和第二灰度特征向量，生成多分区灰度分布特征向量。
指纹活体检测设备可以通过统计第一指纹图像和第二指纹图像的各个区域的灰度值，来确定第一灰度特征向量和第二灰度特征向量。
在本申请实施例中，指纹活体检测设备可以确定该多个第一预估明区和多个第二预估暗区内各个灰度值分别对应的像素点个数，并根据该多个第一预估明区和多个第二预估暗区内各个灰度值分别对应的像素点个数，生成第一灰度特征向量。也即是，指纹活体检测设备可以统计第一指纹图像的多个第一预估明区内和第二指纹图像的多个第二预估暗区内各个灰度值分别对应的像素点个数，来确定第一灰度特征向量。其中，多个第一预估明区与多个第二预估暗区一一对应，且对应的区域是处于两幅图像的同一个高度范围的区域。
示例性地，以图8所介绍的分区示意图为例，指纹活体检测设备可以统计第一指纹图像的区域1、区域3和区域5内，灰度值‘0’至‘255’中每个灰度值对应的像素点个数，将统计得到的各个灰度值对应的像素点的个数，按照灰度值从0-255的顺序依次排列，从而得到向量histr1。统计第二指纹图像的区域1、区域3和区域5内，灰度值‘0’至‘255’中每个灰度值的像素点个数，同样将统计得到的各个灰度值对应的像素点的个数，按照灰度值从0-255的顺序依次排列，得到向量histb1。其中，向量histr1和向量histb1的长度均为256。然后可以将向量histr1和向量histb1相加，得到第一灰度特征向量hisfp1，向量hisfp1的长度也为256。
同样地，指纹活体检测设备可以确定该多个第一预估暗区和该多个第二预估明区内各个灰度值分别对应的像素点个数，并根据该多个第一预估暗区和该多个第二预估明区内各个灰度值分别对应的像素点个数，生成第二灰度特征向量。也即是，指纹活体检测设备可以统计第一指纹图像的多个第一预估暗区内和第二指纹图像的多个第二预估明区内各个灰度值分别对应的像素点个数，来确定第二灰度特征向量。
示例性地，仍以图8所介绍的分区示意图为例，指纹活体检测设备可以统计第一指纹图像的区域2和区域4内，灰度值‘0’至‘255’中每个灰度值的 像素点个数，将统计得到的各个灰度值对应的像素点的个数，按照灰度值从0-255的顺序依次排列，从而得到向量histr2。统计第二指纹图像的区域2和区域4内，灰度值‘0’至‘255’中每个灰度值的像素点个数，将统计得到的各个灰度值对应的像素点的个数，按照灰度值从0-255的顺序依次排列，从而得到向量histb2。其中，向量histr2和向量histb2的长度均为256。然后可以将向量histr2和向量histb2相加，得到第二灰度特征向量hisfp2，向量hisfp2的长度也为256。
可选地，由于指纹活体检测设备采集图像时的曝光度会对采集的图像产生影响，曝光度过高，采集的图像的过亮，曝光度过低，采集的图像过暗，为了减少曝光的影响，保留有效的灰度信息，可以缩小灰度值的统计范围，也即是可以根据采集该指纹的图像的图像采集设备的曝光参数来确定一个灰度值统计范围。在实际应用中，不同的图像采集设备的曝光参数可能不同，可以根据各个设备的曝光参数来设置对应的灰度值统计范围。需要说明的是，在本申请实施例中，指纹活体检测设备即为图像采集设备。另外，通过缩小灰度值统计范围，可以减少计算量，加快检测速度。
基于上述介绍，在本申请实施例中，指纹活体检测设备可以获取灰度值统计范围，并根据该灰度值统计范围，确定该多个第一预估明区和该多个第二预估暗区内处于该灰度值统计范围的各个灰度值对应的像素点个数，根据该多个第一预估明区和多个第二预估暗区内处于该灰度值统计范围的各个灰度值对应的像素点个数，生成第一灰度特征向量。
同样地，指纹活体检测设备可以根据该灰度值统计范围，确定该多个第一预估暗区和该多个第二预估明区内处于该灰度值统计范围的各个灰度值对应的像素点个数，根据该多个第一预估暗区和多个第二预估明区内处于该灰度值统计范围的各个灰度值对应的像素点个数，生成第二灰度特征向量。
也即是，指纹活体检测设备可以仅统计第一指纹图像和第二指纹图像的各个区域内处于该灰度值统计范围内的各个灰度值对应的像素点个数，并按照前述介绍的方法来生成第一灰度特征向量和第二灰度特征向量。
示例性地，灰度值统计范围可以为‘11’至‘230’，这样根据前述方法得到的第一灰度特征向量和第二灰度特征向量的长度均为220，向量的长度减小了，可以减少后续的计算量。
在得到第一灰度特征向量和第二灰度特征向量之后，指纹活体检测设备可 以将第一灰度特征向量和第二灰度特征向量进行拼接，得到多分区灰度分布特征向量。
示例性地，第一灰度特征向量为hisfp1，第二灰度特征向量为hisfp2，将第一灰度特征向量和第二灰度特征向量拼接之后，得到的多分区灰度分布特征向量可以为[hisfp1，hisfp2]。
以上介绍的指纹活体检测设备得到多分区灰度分布特征向量的步骤为，先统计第一指纹图像的多个第一预估明区和第二指纹图像的多个第二预估暗区，分别得到两个向量，这两个向量相加得到第一灰度特征向量，再统计第一指纹图像的多个第一预估暗区和第二指纹图像的多个第二预估明区，分别得到两个向量，这两个向量相加得到第二灰度特征向量，之后将第一灰度特征向量和第二灰度特征向量进行拼接，得到多分区灰度分布特征向量。
可选性，在本申请实施例中，指纹活体检测设备也可以先统计第一指纹图像的多个第一预估明区，得到向量histr1，统计第一指纹图像的多个第一预估暗区，得到向量histr2，将向量histr1和向量histr2进行拼接，得到第一灰度拼接向量histr，再统计第二指纹图像的多个第二预估暗区，得到向量histb1，统计第二指纹图像的多个第二预估明区，得到向量histb2，将向量histb1和向量histb2进行拼接，得到第二灰度拼接向量histb，之后，将向量histr和向量histb相加，得到多分区灰度分布特征向量hisfp。也即是，分别统计两幅图像，得到两幅图像各自对应的明区和暗区的灰度拼接向量，然后将两个灰度拼接向量相加，得到多分区灰度分布特征向量。这样，指纹活体检测设备可以先处理一幅图像，再处理另一幅图像，无需来回地更换要统计的图像的图像数据。
需要说明的是，以上介绍的统计各个区域的灰度值，来确定多分区灰度分布特征向量的步骤中，各个区域的统计顺序可以不分先后，根据实际确定一个统计顺序即可。
可选地，指纹活体检测设备也可以将上述各个统计结果用灰度直方图进行表示，根据灰度直方图来确定各个向量。
步骤204：将多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果。
在本申请实施例中，指纹活体检测设备在得到多分区灰度分布特征向量之后，可以将该向量输入指纹活体检测模型，得到指纹活体检测结果。
可选地，指纹活体检测模型可以为SVM(Support Vector Machine,支持向量 机)分类器、神经网络模型、随机森林模型等，本申请实施例对此不作限定。指纹活体检测结果可以为‘是活体指纹’或者‘不是活体指纹’，或者为‘0’或‘1’，‘0’表示不是活体指纹，‘1’表示是活体指纹。
需要说明的是，在本申请实施例中，指纹活体检测模型可以是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的，指纹活体检测模型为事先训练好的模型，可以在指纹活体检测设备上训练得到该指纹活体检测模型，或者还可以由其他的设备训练得到该指纹活体检测模型，例如后台设备，或其他的计算机设备，之后将该指纹活体检测模型部署在指纹活体检测设备上。本申请实施例以在指纹活体检测设备上训练得到该指纹活体检测模型为例进行介绍。
在本申请实施例中，指纹活体检测设备可以获取活体指纹样本图像和非活体指纹样本图像，活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个活体指纹样本的图像，非活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个非活体指纹样本的图像。之后，指纹活体检测设备可以根据该活体指纹样本图像和非活体指纹样本图像，通过监督学习训练得到指纹活体检测模型。
需要说明的是，获取活体指纹样本图像和非活体指纹样本图像的方法，与前述采集第一原始图像的方法相同，这里不再赘述。
在获取活体指纹样本图像和非活体指纹样本图像之后，指纹活体检测设备可以按照前述得到多分区灰度分布特征向量的方法，得到每个样本图像对应的多分区灰度分布特征向量。之后，可以根据各个样本图像对应的多分区灰度分布特征向量，训练得到指纹活体检测模型。
可选地，每个样本图像对应一个样本标签，活体指纹样本图像的样本标签可以为‘1’，非活体指纹样本图像的样本标签可以为‘0’，指纹活体检测设备可以根据各个样本图像对应的多分区灰度分布特征向量和样本标签，通过监督学习训练得到指纹活体检测模型。
示例性地，假设指纹活体检测模型为SVM分类器，指纹活体检测设备可以根据各个样本图像对应的多分区灰度分布特征向量和样本标签，通过监督学习训练得到该SVM分类器，该SVM分类器为一个二分类器。
图9是本申请实施例提供的另一种指纹活体检测方法的流程图。参见图9，假设指纹活体检测模型为SVM分类器，分别对第一指纹图像(ImgR)和第二 指纹图像(ImgB)，进行特征提取，得到向量histr和向量histb，再进行特征合成，也即将向量histr和向量histb进行拼接，得到多分区灰度分布特征向量hiafp，将向量hisfp输入SVM分类器中，输出指纹活体检测结果。
以上介绍了本申请实施例提供的指纹活体检测的方法，在实际应用中，需要将指纹识别和指纹活体检测结合，来确定最终的安全验证结果。也即是，如果指纹识别结果指示所采集的指纹为有效指纹，且指纹活体检测结果指示所采集的指纹为活体指纹，则确定安全验证通过。基于此，本申请实施例还提供了一种指纹安全验证方法，接下来将对此进行介绍。
在本申请实施例中，指纹活体检测设备还可以获取第四原始图像，第四原始图像为绿灯开启的状态下采集的所述指纹的图像，根据第四原始图像，确定指纹识别结果。之后，指纹活体检测设备可以根据指纹识别结果和指纹活体检测结果，确定安全验证结果。
由于活体指纹的绿色通道图像能够呈现较为清晰的指纹纹路，也即没有明暗交替的特征，相较于红色通道图像和蓝色通道图像，绿色通道图像更适合于指纹识别。另外，在三种颜色的灯均开启的状态下，所采集的活体指纹的图像，也是呈现很清晰的指纹纹路，也适合于指纹识别。基于此，可以在唤醒图像采集器进行指纹采集时，不仅采集红灯和蓝灯开启状态下该指纹的原始图像，还可以采集在绿灯开启的状态下该指纹的图像，得到第四原始图像。其中，绿灯开启的状态可以是指绿灯开启、红灯和蓝灯关闭的状态，或者，绿灯、红灯和蓝灯均开启的状态。也即是，在唤醒图像采集器之后，可以采集第一原始图像和第四原始图像，以完成一次采集，或者采集第二原始图像、第三原始图像和第四原始图像，以完成一次采集。
另外，每次完成采集后，可以关闭所有的灯，且本申请实施例对要采集的原始图像的顺序不作限定。例如，可以先仅开启绿灯采集第四原始图像，再关闭绿灯、开启红灯和蓝灯采集第一原始图像。又如，可以先开启绿灯、红灯和蓝灯采集第四原始图像，再关闭绿灯、保持红灯和蓝灯开启的状态采集第一原始图像。再如，可以先开启红灯采集第二原始图像，再关闭红灯、开启蓝灯采集第三原始图像，最后再关闭蓝灯、开启绿灯采集第四原始图像。
在获取到第四原始图像之后，指纹活体检测设备可以提取第四原始图像的绿色通道的图像数据，生成第三指纹图像，之后，提取第三指纹图像的指纹特征，例如纹数、纹形、核心点、断点、交叉点、方向、曲率等，根据提取的指 纹特征与存储的各个有效指纹的图像对应的特征进行对比，得到指纹识别结果。其中，指纹识别结果可以为‘是有效指纹’或者‘不是有效指纹’。
其中，提取第四原始图像的绿色通道的图像数据，生成第三指纹图像的实现方式可以参照前述有关生成第一指纹图像和第二指纹图像的相关介绍。
示例性地，以图3所示的拜尔模式的彩色图像传感器为例，指纹活体检测设备在采集到第四原始图像之后，可以提取Gr通道的图像数据，或者提取Gb通道的图像数据，得到图像大小为w0*h0/4的第三指纹图像。可选地，指纹活体检测设备也可以对提取之后得到的图像进行畸变校正，得到校正之后的第三指纹图像。可选地，进一步地，指纹活体检测设备还可以根据预设图像大小参数，对校正之后的图像进行裁剪得到中间较为清晰的部分，作为第三指纹图像。
需要说明的是，第三指纹图像的图像大小可以与第一指纹图像的图像大小相同或不同。如果相同，则设置一组预设图像大小参数即可。
可选地，在获取到第四原始图像之后，指纹活体检测设备也可以直接提取第四原始图像的指纹特征，根据提取的指纹特征与存储的各个有效指纹的图像对应的特征进行对比，得到指纹识别结果。
在根据以上方法得到指纹识别结果和指纹活体检测结果之后，指纹活体检测设备可以根据这两个结果，确定最终的安全验证结果。示例性地，如果指纹识别结果指示所采集的指纹为有效指纹，且指纹活体检测结果指示所采集的指纹为活体指纹，则确定安全验证通过。如果指纹识别结果指示所采集的指纹不是有效指纹，或者指纹活体检测结果指示所采集的指纹不是活体指纹，则确定安全验证不通过，指纹活体检测设备可以以文字提示或者声音来提示用户。
综上所述，在本申请实施例中，可以对第一指纹图像和第二指纹图像进行区域划分，进而根据划分的区域来确定多分区灰度分布特征向量，以通过指纹活体检测模型进行指纹活体检测，其中，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像。由于非活体指纹的各个区域灰度值的分布规律与活体指纹的不同，因此，通过多分区灰度分布特征向量可以识别出指纹图像是否存在活体指纹的灰度分布特征，进而确定出该指纹是否为活体指纹，安全性较高。
图10是本申请实施例提供的一种指纹活体检测装置1000的结构示意图，该指纹活体检测装置1000可以由软件、硬件或者两者的结合实现成为计算机设 备的部分或者全部。请参考图10，该装置包括：第一获取模块1001、第一获取模块1002、第一获取模块1003和第一获取模块1004。
第一获取模块1001，用于获取第一指纹图像和第二指纹图像，第一指纹图像和第二指纹图像是指同一指纹的图像，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像；
分区模块1002，用于对第一指纹图像和第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区；
第一确定模块1003，用于根据多个第一预估明区、多个第一预估暗区、多个第二预估明区和多个第二预估暗区，确定多分区灰度分布特征向量；
检测模块1004，用于将多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果，指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
可选地，第一获取模块1001包括：
第一获取单元，用于获取第一原始图像，第一原始图像为绿灯关闭、红灯和蓝灯开启的状态下采集的指纹的图像；
第一提取单元，用于提取第一原始图像的红色通道的图像数据，生成第一指纹图像；
第二提取单元，用于提取第一原始图像的蓝色通道的图像数据，生成第二指纹图像。
可选地，第一获取模块1001包括：
第二获取单元，用于获取第二原始图像和第三原始图像，第二原始图像为绿灯和蓝灯关闭、红灯开启的状态下采集的指纹的图像，第三原始图像为绿灯和红灯关闭、蓝灯开启的状态下采集的指纹的图像；
第三提取单元，用于提取第二原始图像的红色通道的图像数据，生成第一指纹图像；
第四提取单元，用于提取第三原始图像的蓝色通道的图像数据，生成第二指纹图像。
可选地，分区模块1002包括：
第一确定单元，用于根据活体指纹的红色通道图像和蓝色通道图像中明区和暗区的分布规律，确定分区参数，分区参数包括多个分区宽度比例；
分区单元，用于根据多个分区宽度比例，对第一指纹图像和第二指纹图像分别进行区域划分。
可选地，第一指纹图像和第二指纹图像的图像大小相同，第一确定模块1003包括：
第二确定单元，用于根据多个第一预估明区和多个第二预估暗区，确定第一灰度特征向量；
第三确定单元，用于根据多个第一预估暗区和多个第二预估明区，确定第二灰度特征向量；
生成单元，用于根据第一灰度特征向量和第二灰度特征向量，生成多分区灰度分布特征向量。
可选地，第二确定单元包括：
第一确定子单元，用于确定多个第一预估明区和多个第二预估暗区内各个灰度值分别对应的像素点个数；
第一生成子单元，用于根据多个第一预估明区和多个第二预估暗区内各个灰度值分别对应的像素点个数，生成第一灰度特征向量。
可选地，第三确定单元包括：
第二确定子单元，用于确定多个第一预估暗区和多个第二预估明区内各个灰度值分别对应的像素点个数；
第二生成子单元，用于根据多个第一预估暗区和多个第二预估明区内各个灰度值分别对应的像素点个数，生成第二灰度特征向量。
可选地，第一确定模块1003还包括：
第三获取单元，用于获取灰度值统计范围，灰度值统计范围根据采集指纹的图像的图像采集设备的曝光参数确定；
第二确定单元包括：
第三确定子单元，用于根据灰度值统计范围，确定多个第一预估明区和多个第二预估暗区内处于灰度值统计范围的各个灰度值对应的像素点个数；
第三生成子单元，用于根据多个第一预估明区和多个第二预估暗区内处于灰度值统计范围的各个灰度值对应的像素点个数，生成第一灰度特征向量。
可选地，生成单元包括：
拼接子单元，用于将第一灰度特征向量和第二灰度特征向量进行拼接，得到多分区灰度分布特征向量。
可选地，该装置1000还包括：
第二获取模块，用于获取活体指纹样本图像和非活体指纹样本图像，活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个活体指纹样本的图像，非活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个非活体指纹样本的图像；
训练模块，用于根据活体指纹样本图像和非活体指纹样本图像，通过监督学习训练得到指纹活体检测模型。
可选地，参见图11，该装置1000还包括：
第三获取模块1005，用于获取第四原始图像，第四原始图像为绿灯开启的状态下采集的指纹的图像；
识别模块1006，用于根据第四原始图像，确定指纹识别结果；
第二确定模块1007，用于根据指纹识别结果和指纹活体检测结果，确定安全验证结果。
综上所述，在本申请实施例中，可以对第一指纹图像和第二指纹图像进行区域划分，进而根据划分区域来确定多分区灰度分布特征向量，以通过指纹活体检测模型进行指纹活体检测，其中，第一指纹图像为红色通道图像，第二指纹图像为蓝色通道图像。由于非活体指纹的这两个通道图像的明区和暗区的灰度值的分布规律与活体指纹的不同，因此，通过多分区灰度分布特征向量可以识别出指纹图像是否存在活体指纹的灰度分布特征，进而确定出该指纹是否为活体指纹，安全性较高。
需要说明的是：上述实施例提供的指纹活体检测装置在检测指纹时，仅以上述各功能模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能模块完成，即将装置的内部结构划分成不同的功能模块，以完成以上描述的全部或者部分功能。另外，上述实施例提供的指纹活体检测装置与指纹活体检测方法实施例属于同一构思，其具体实现过程详见方法实施例，这里不再赘述。
图12是本申请实施例提供的一种计算机设备1200的结构框图。该计算机设备1200可以是智能手机、平板电脑、笔记本电脑或台式电脑等可以通过采集指纹来进行安全验证的指纹活体检测设备。
通常，计算机设备1200包括有：处理器1201和存储器1202。
处理器1201可以包括一个或多个处理核心，比如4核心处理器、12核心处理器等。处理器1201可以采用DSP(Digital Signal Processing，数字信号处理)、FPGA(Field－Programmable Gate Array，现场可编程门阵列)、PLA(Programmable Logic Array，可编程逻辑阵列)中的至少一种硬件形式来实现。处理器1201也可以包括主处理器和协处理器，主处理器是用于对在唤醒状态下的数据进行处理的处理器，也称CPU(Central Processing Unit，中央处理器)；协处理器是用于对在待机状态下的数据进行处理的低功耗处理器。在一些实施例中，处理器1201可以集成有GPU(Graphics Processing Unit，图像处理器)，GPU用于负责显示屏所需要显示的内容的渲染和绘制。一些实施例中，处理器1201还可以包括AI(Artificial Intelligence，人工智能)处理器，该AI处理器用于处理有关机器学习的计算操作。
存储器1202可以包括一个或多个计算机可读存储介质，该计算机可读存储介质可以是非暂态的。存储器1202还可包括高速随机存取存储器，以及非易失性存储器，比如一个或多个磁盘存储设备、闪存存储设备。在一些实施例中，存储器1202中的非暂态的计算机可读存储介质用于存储至少一个指令，该至少一个指令用于被处理器1201所执行以实现本申请实施例中方法实施例提供的指纹活体检测方法。
在一些实施例中，计算机设备1200还可选包括有：外围设备接口1203和至少一个外围设备。处理器1201、存储器1202和外围设备接口1203之间可以通过总线或信号线相连。各个外围设备可以通过总线、信号线或电路板与外围设备接口1203相连。具体地，外围设备包括：射频电路1204、显示屏1205、摄像头组件1206、音频电路1207、定位组件1208和电源1209中的至少一种。
外围设备接口1203可被用于将I/O(Input/Output，输入/输出)相关的至少一个外围设备连接到处理器1201和存储器1202。在一些实施例中，处理器1201、存储器1202和外围设备接口1203被集成在同一芯片或电路板上；在一些其他实施例中，处理器1201、存储器1202和外围设备接口1203中的任意一个或两个可以在单独的芯片或电路板上实现，本实施例对此不加以限定。
射频电路1204用于接收和发射RF(Radio Frequency，射频)信号，也称电磁信号。射频电路1204通过电磁信号与通信网络以及其他通信设备进行通信。射频电路1204将电信号转换为电磁信号进行发送，或者，将接收到的电磁信号转换为电信号。可选地，射频电路1204包括：天线系统、RF收发器、一个或 多个放大器、调谐器、振荡器、数字信号处理器、编解码芯片组、用户身份模块卡等等。射频电路1204可以通过至少一种无线通信协议来与其它计算机设备进行通信。该无线通信协议包括但不限于：万维网、城域网、内联网、各代移动通信网络(2G、3G、4G及5G)、无线局域网和/或WiFi(Wireless Fidelity，无线保真)网络。在一些实施例中，射频电路1204还可以包括NFC(Near Field Communication，近距离无线通信)有关的电路，本申请实施例对此不加以限定。
显示屏1205用于显示UI(User Interface，用户界面)。该UI可以包括图形、文本、图标、视频及其它们的任意组合。当显示屏1205是触摸显示屏时，显示屏1205还具有采集在显示屏1205的表面或表面上方的触摸信号的能力。该触摸信号可以作为控制信号输入至处理器1201进行处理。此时，显示屏1205还可以用于提供虚拟按钮和/或虚拟键盘，也称软按钮和/或软键盘。在一些实施例中，显示屏1205可以为一个，设置于计算机设备1200的前面板；在另一些实施例中，显示屏1205可以为至少两个，分别设置在计算机设备1200的不同表面或呈折叠设计；在其他一些实施例中，显示屏1205可以是柔性显示屏，设置在计算机设备1200的弯曲表面上或折叠面上。甚至，显示屏1205还可以设置成非矩形的不规则图形，也即异形屏。显示屏1205可以采用LCD(Liquid Crystal Display，液晶显示屏)、OLED(Organic Light-Emitting Diode,有机发光二极管)等材质制备。
摄像头组件1206用于采集图像或视频。可选地，摄像头组件1206包括前置摄像头和后置摄像头。通常，前置摄像头设置在计算机设备的前面板，后置摄像头设置在计算机设备的背面。在一些实施例中，后置摄像头为至少两个，分别为主摄像头、景深摄像头、广角摄像头、长焦摄像头中的任意一种，以实现主摄像头和景深摄像头融合实现背景虚化功能、主摄像头和广角摄像头融合实现全景拍摄以及VR(Virtual Reality，虚拟现实)拍摄功能或者其它融合拍摄功能。在一些实施例中，摄像头组件1206还可以包括闪光灯。闪光灯可以是单色温闪光灯，也可以是双色温闪光灯。双色温闪光灯是指暖光闪光灯和冷光闪光灯的组合，可以用于不同色温下的光线补偿。
音频电路1207可以包括麦克风和扬声器。麦克风用于采集用户及环境的声波，并将声波转换为电信号输入至处理器1201进行处理，或者输入至射频电路1204以实现语音通信。出于立体声采集或降噪的目的，麦克风可以为多个，分别设置在计算机设备1200的不同部位。麦克风还可以是阵列麦克风或全向采集 型麦克风。扬声器则用于将来自处理器1201或射频电路1204的电信号转换为声波。扬声器可以是传统的薄膜扬声器，也可以是压电陶瓷扬声器。当扬声器是压电陶瓷扬声器时，不仅可以将电信号转换为人类可听见的声波，也可以将电信号转换为人类听不见的声波以进行测距等用途。在一些实施例中，音频电路1207还可以包括耳机插孔。
定位组件1208用于定位计算机设备1200的当前地理位置，以实现导航或LBS(Location Based Service，基于位置的服务)。定位组件1208可以是基于美国的GPS(Global Positioning System，全球定位系统)、中国的北斗系统或俄罗斯的伽利略系统的定位组件。
电源1209用于为计算机设备1200中的各个组件进行供电。电源1209可以是交流电、直流电、一次性电池或可充电电池。当电源1209包括可充电电池时，该可充电电池可以是有线充电电池或无线充电电池。有线充电电池是通过有线线路充电的电池，无线充电电池是通过无线线圈充电的电池。该可充电电池还可以用于支持快充技术。
在一些实施例中，计算机设备1200还包括有一个或多个传感器1210。该一个或多个传感器1210包括但不限于：加速度传感器1211、陀螺仪传感器1212、压力传感器1213、指纹传感器1214、光学传感器1215以及接近传感器1216。
加速度传感器1211可以检测以计算机设备1200建立的坐标系的三个坐标轴上的加速度大小。比如，加速度传感器1211可以用于检测重力加速度在三个坐标轴上的分量。处理器1201可以根据加速度传感器1211采集的重力加速度信号，控制显示屏1205以横向视图或纵向视图进行用户界面的显示。加速度传感器1211还可以用于游戏或者用户的运动数据的采集。
陀螺仪传感器1212可以检测计算机设备1200的机体方向及转动角度，陀螺仪传感器1212可以与加速度传感器1211协同采集用户对计算机设备1200的3D动作。处理器1201根据陀螺仪传感器1212采集的数据，可以实现如下功能：动作感应(比如根据用户的倾斜操作来改变UI)、拍摄时的图像稳定、游戏控制以及惯性导航。
压力传感器1213可以设置在计算机设备1200的侧边框和/或显示屏1205的下层。当压力传感器1213设置在计算机设备1200的侧边框时，可以检测用户对计算机设备1200的握持信号，由处理器1201根据压力传感器1213采集的握持信号进行左右手识别或快捷操作。当压力传感器1213设置在显示屏1205的 下层时，由处理器1201根据用户对显示屏1205的压力操作，实现对UI界面上的可操作性控件进行控制。可操作性控件包括按钮控件、滚动条控件、图标控件、菜单控件中的至少一种。
指纹传感器1214用于采集用户的指纹，由处理器1201根据指纹传感器1214采集到的指纹识别用户的身份，或者，由指纹传感器1214根据采集到的指纹识别用户的身份。在识别出用户的身份为可信身份时，由处理器1201授权该用户执行相关的敏感操作，该敏感操作包括解锁屏幕、查看加密信息、下载软件、支付及更改设置等。指纹传感器1214可以被设置在计算机设备1200的正面、背面或侧面。当计算机设备1200上设置有物理按键或厂商Logo时，指纹传感器1214可以与物理按键或厂商Logo集成在一起。
光学传感器1215用于采集环境光强度。在一个实施例中，处理器1201可以根据光学传感器1215采集的环境光强度，控制显示屏1205的显示亮度。具体地，当环境光强度较高时，调高显示屏1205的显示亮度；当环境光强度较低时，调低显示屏1205的显示亮度。在另一个实施例中，处理器1201还可以根据光学传感器1215采集的环境光强度，动态调整摄像头组件1206的拍摄参数。
接近传感器1216，也称距离传感器，通常设置在计算机设备1200的前面板。接近传感器1216用于采集用户与计算机设备1200的正面之间的距离。在一个实施例中，当接近传感器1216检测到用户与计算机设备1200的正面之间的距离逐渐变小时，由处理器1201控制显示屏1205从亮屏状态切换为息屏状态；当接近传感器1216检测到用户与计算机设备1200的正面之间的距离逐渐变大时，由处理器1201控制显示屏1205从息屏状态切换为亮屏状态。
本领域技术人员可以理解，图12中示出的结构并不构成对计算机设备1200的限定，可以包括比图示更多或更少的组件，或者组合某些组件，或者采用不同的组件布置。
在一些实施例中，还提供了一种计算机可读存储介质，该存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现上述实施例中指纹活体检测方法的步骤。例如，所述计算机可读存储介质可以是ROM、RAM、CD-ROM、磁带、软盘和光数据存储设备等。
值得注意的是，本申请实施例提到的计算机可读存储介质可以为非易失性存储介质，换句话说，可以是非瞬时性存储介质。
应当理解的是，实现上述实施例的全部或部分步骤可以通过软件、硬件、固件或者其任意结合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。所述计算机指令可以存储在上述计算机可读存储介质中。
也即是，在一些实施例中，还提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述所述的指纹活体检测方法的步骤。
以上所述为本申请提供的实施例，并不用以限制本申请，凡在本申请的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (14)

  1. 一种指纹活体检测方法，其特征在于，所述方法包括：
    获取第一指纹图像和第二指纹图像，所述第一指纹图像和所述第二指纹图像是指同一指纹的图像，所述第一指纹图像为红色通道图像，所述第二指纹图像为蓝色通道图像；
    对所述第一指纹图像和所述第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区；
    根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量；
    将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果，所述指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
  2. 根据权利要求1所述的方法，其特征在于，所述获取第一指纹图像和第二指纹图像，包括：
    获取第一原始图像，所述第一原始图像为绿灯关闭、红灯和蓝灯开启的状态下采集的所述指纹的图像；
    提取所述第一原始图像的红色通道的图像数据，生成所述第一指纹图像；
    提取所述第一原始图像的蓝色通道的图像数据，生成所述第二指纹图像。
  3. 根据权利要求1所述的方法，其特征在于，所述获取第一指纹图像和第二指纹图像，包括：
    获取第二原始图像和第三原始图像，所述第二原始图像为绿灯和蓝灯关闭、红灯开启的状态下采集的所述指纹的图像，所述第三原始图像为绿灯和红灯关闭、蓝灯开启的状态下采集的所述指纹的图像；
    提取所述第二原始图像的红色通道的图像数据，生成所述第一指纹图像；
    提取所述第三原始图像的蓝色通道的图像数据，生成所述第二指纹图像。
  4. 根据权利要求1所述的方法，其特征在于，所述对所述第一指纹图像和所 述第二指纹图像分别进行区域划分，包括：
    根据活体指纹的红色通道图像和蓝色通道图像中明区和暗区的分布规律，确定分区参数，所述分区参数包括多个分区宽度比例；
    根据所述多个分区宽度比例，对所述第一指纹图像和所述第二指纹图像分别进行区域划分。
  5. 根据权利要求1-4任一所述的方法，其特征在于，所述第一指纹图像和所述第二指纹图像的图像大小相同，所述根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量，包括：
    根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量；
    根据所述多个第一预估暗区和所述多个第二预估明区，确定第二灰度特征向量；
    根据所述第一灰度特征向量和所述第二灰度特征向量，生成所述多分区灰度分布特征向量。
  6. 根据权利要求5所述的方法，其特征在于，所述根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量，包括：
    确定所述多个第一预估明区和所述多个第二预估暗区内各个灰度值分别对应的像素点个数；
    根据所述多个第一预估明区和所述多个第二预估暗区内各个灰度值分别对应的像素点个数，生成所述第一灰度特征向量。
  7. 根据权利要求5所述的方法，其特征在于，所述根据所述多个第一预估暗区和所述多个第二预估明区，确定第二灰度特征向量，包括：
    确定所述多个第一预估暗区和所述多个第二预估明区内各个灰度值分别对应的像素点个数；
    根据所述多个第一预估暗区和所述多个第二预估明区内各个灰度值分别对应的像素点个数，生成所述第二灰度特征向量。
  8. 根据权利要求5所述的方法，其特征在于，所述根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量之前，还包括：
    获取灰度值统计范围，所述灰度值统计范围根据采集所述指纹的图像的图像采集设备的曝光参数确定；
    所述根据所述多个第一预估明区和所述多个第二预估暗区，确定第一灰度特征向量，包括：
    根据所述灰度值统计范围，确定所述多个第一预估明区和所述多个第二预估暗区内处于所述灰度值统计范围的各个灰度值对应的像素点个数；
    根据所述多个第一预估明区和所述多个第二预估暗区内处于所述灰度值统计范围的各个灰度值对应的像素点个数，生成所述第一灰度特征向量。
  9. 根据权利要求5所述的方法，其特征在于，所述根据所述第一灰度特征向量和所述第二灰度特征向量，生成所述多分区灰度分布特征向量，包括：
    将所述第一灰度特征向量和所述第二灰度特征向量进行拼接，得到所述多分区灰度分布特征向量。
  10. 根据权利要求1-9任一所述的方法，其特征在于，所述将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果之前，还包括：
    获取所述活体指纹样本图像和所述非活体指纹样本图像，所述活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个活体指纹样本的图像，所述非活体指纹样本图像包括在绿灯关闭、红灯和蓝灯开启的状态下采集的多个非活体指纹样本的图像；
    根据所述活体指纹样本图像和所述非活体指纹样本图像，通过监督学习训练得到所述指纹活体检测模型。
  11. 根据权利要求1-10任一所述的方法，其特征在于，所述方法还包括：
    获取第四原始图像，所述第四原始图像为绿灯开启的状态下采集的所述指纹的图像；
    根据所述第四原始图像，确定指纹识别结果；
    根据所述指纹识别结果和所述指纹活体检测结果，确定安全验证结果。
  12. 一种指纹活体检测设备，其特征在于，所述指纹活体检测设备包括图像采集器和处理器，所述图像采集器包括补光灯和图像传感器；
    所述图像采集器，用于采集第一原始图像；
    所述补光灯，用于在采集所述第一原始图像的过程中，关闭绿灯、开启红灯和蓝灯为所述图像传感器进行补光；
    所述处理器，用于对所述第一原始图像进行处理，得到第一指纹图像和第二指纹图像，所述第一指纹图像为红色通道图像，所述第二指纹图像为蓝色通道图像；对所述第一指纹图像和所述第二指纹图像分别进行区域划分，其中，划分后的第一指纹图像中包括多个第一预估明区和多个第一预估暗区，划分后的第二指纹图像中包括多个第二预估明区和多个第二预估暗区；根据所述多个第一预估明区、所述多个第一预估暗区、所述多个第二预估明区和所述多个第二预估暗区，确定多分区灰度分布特征向量；将所述多分区灰度分布特征向量输入指纹活体检测模型，得到指纹活体检测结果，所述指纹活体检测模型是根据活体指纹样本图像和非活体指纹样本图像通过监督学习训练得到的。
  13. 根据权利要求12所述的指纹活体检测设备，其特征在于，
    所述图像采集器，还用于采集第四原始图像；
    所述补光灯，还用于在采集所述第四原始图像的过程中，开启绿灯为所述图像传感器进行补光；
    所述处理器，还用于根据所述第四原始图像，确定指纹识别结果，并根据所述指纹识别结果和所述指纹活体检测结果，确定安全验证结果。
  14. 一种计算机可读存储介质，其特征在于，所述存储介质内存储有计算机程序，所述计算机程序被处理器执行时实现权利要求1-11任一所述方法的步骤。
PCT/CN2021/089361 2020-04-30 2021-04-23 指纹活体检测方法、设备及存储介质 WO2021218823A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
KR1020227030315A KR20220130804A (ko) 2020-04-30 2021-04-23 지문 생체 검출 방법, 기기 및 저장 매체
EP21797544.0A EP4145343A4 (en) 2020-04-30 2021-04-23 METHOD AND DEVICE FOR DETECTING FINGERPRINT VIABILITY AND STORAGE MEDIUM

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010364403.XA CN113591514B (zh) 2020-04-30 2020-04-30 指纹活体检测方法、设备及存储介质
CN202010364403.X 2020-04-30

Publications (1)

Publication Number Publication Date
WO2021218823A1 true WO2021218823A1 (zh) 2021-11-04

Family

ID=78237292

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/089361 WO2021218823A1 (zh) 2020-04-30 2021-04-23 指纹活体检测方法、设备及存储介质

Country Status (4)

Country Link
EP (1) EP4145343A4 (zh)
KR (1) KR20220130804A (zh)
CN (1) CN113591514B (zh)
WO (1) WO2021218823A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114998841A (zh) * 2022-08-01 2022-09-02 易凯医疗建筑设计(深圳)有限公司 核酸采样的样本识别方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104392227A (zh) * 2014-12-15 2015-03-04 金虎林 活体指纹判断方法及系统
CN106295555A (zh) * 2016-08-08 2017-01-04 深圳芯启航科技有限公司 一种活体指纹图像的检测方法
US20180137329A1 (en) * 2016-11-11 2018-05-17 Samsung Electronics Co., Ltd. User authentication method using fingerprint image and method of generating coded model for user authentication
CN108549884A (zh) * 2018-06-15 2018-09-18 天地融科技股份有限公司 一种活体检测方法及装置
CN110765857A (zh) * 2019-09-12 2020-02-07 敦泰电子（深圳）有限公司 指纹识别方法、芯片及电子装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7668350B2 (en) * 2003-04-04 2010-02-23 Lumidigm, Inc. Comparative texture analysis of tissue for biometric spoof detection
BR112016007929B1 (pt) * 2013-10-11 2021-03-02 Hid Global Corporation sistema de acesso biométrico

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104392227A (zh) * 2014-12-15 2015-03-04 金虎林 活体指纹判断方法及系统
CN106295555A (zh) * 2016-08-08 2017-01-04 深圳芯启航科技有限公司 一种活体指纹图像的检测方法
US20180137329A1 (en) * 2016-11-11 2018-05-17 Samsung Electronics Co., Ltd. User authentication method using fingerprint image and method of generating coded model for user authentication
CN108549884A (zh) * 2018-06-15 2018-09-18 天地融科技股份有限公司 一种活体检测方法及装置
CN110765857A (zh) * 2019-09-12 2020-02-07 敦泰电子（深圳）有限公司 指纹识别方法、芯片及电子装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4145343A4

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114998841A (zh) * 2022-08-01 2022-09-02 易凯医疗建筑设计(深圳)有限公司 核酸采样的样本识别方法、装置、设备及存储介质

Also Published As

Publication number Publication date
EP4145343A4 (en) 2023-09-27
CN113591514B (zh) 2024-03-29
KR20220130804A (ko) 2022-09-27
CN113591514A (zh) 2021-11-02
EP4145343A1 (en) 2023-03-08

Similar Documents

Publication Publication Date Title
CN109034102B (zh) 人脸活体检测方法、装置、设备及存储介质
US11244170B2 (en) Scene segmentation method and device, and storage medium
US20220301341A1 (en) Method for processing images and electronic device
CN108594997B (zh) 手势骨架构建方法、装置、设备及存储介质
WO2019101021A1 (zh) 图像识别方法、装置及电子设备
CN110992493B (zh) 图像处理方法、装置、电子设备及存储介质
US12148250B2 (en) AI-based face recognition method and apparatus, device, and medium
CN111079576A (zh) 活体检测方法、装置、设备及存储介质
CN111915305B (zh) 支付方法、装置、设备以及存储介质
CN111127509B (zh) 目标跟踪方法、装置和计算机可读存储介质
US11386586B2 (en) Method and electronic device for adding virtual item
CN109285178A (zh) 图像分割方法、装置及存储介质
CN111144365A (zh) 活体检测方法、装置、计算机设备及存储介质
CN109522863B (zh) 耳部关键点检测方法、装置及存储介质
CN109360222A (zh) 图像分割方法、装置及存储介质
CN112184581B (zh) 图像处理方法、装置、计算机设备及介质
CN110163862B (zh) 图像语义分割方法、装置及计算机设备
WO2021218823A1 (zh) 指纹活体检测方法、设备及存储介质
CN110889391B (zh) 人脸图像处理的方法、装置、计算设备和存储介质
CN112990424A (zh) 神经网络模型训练的方法和装置
CN113938606B (zh) 确定球机架设参数的方法、装置及计算机存储介质
CN109117895A (zh) 数据聚类方法、装置及存储介质
CN111723615B (zh) 对检测物图像进行检测物匹配判定的方法和装置
CN113920222A (zh) 获取地图建图数据的方法、装置、设备及可读存储介质
CN110728275B (zh) 车牌识别方法、装置及存储介质

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: 21797544

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20227030315

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 202217067652

Country of ref document: IN

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021797544

Country of ref document: EP

Effective date: 20221130