WO2022241792A1 - 指纹识别方法、装置、电子设备及存储介质 - Google Patents

指纹识别方法、装置、电子设备及存储介质 Download PDF

Info

Publication number
WO2022241792A1
WO2022241792A1 PCT/CN2021/095328 CN2021095328W WO2022241792A1 WO 2022241792 A1 WO2022241792 A1 WO 2022241792A1 CN 2021095328 W CN2021095328 W CN 2021095328W WO 2022241792 A1 WO2022241792 A1 WO 2022241792A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
fingerprint
sub
area
calibration
Prior art date
Application number
PCT/CN2021/095328
Other languages
English (en)
French (fr)
Inventor
朱强
Original Assignee
深圳市汇顶科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市汇顶科技股份有限公司 filed Critical 深圳市汇顶科技股份有限公司
Priority to PCT/CN2021/095328 priority Critical patent/WO2022241792A1/zh
Publication of WO2022241792A1 publication Critical patent/WO2022241792A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition

Definitions

  • the present application relates to the technical field of fingerprint identification, in particular to a fingerprint identification method, device, electronic equipment and storage medium.
  • the present application provides a fingerprint identification method, device, electronic equipment and storage medium to solve the problems existing in the prior art.
  • the present application provides a fingerprint identification method, including:
  • the first fingerprint data and the second fingerprint data corresponding to the fingerprint to be detected are obtained respectively, and the first fingerprint data is collected when the light emitting unit part corresponding to the pressing area emits light data, the second fingerprint data is the data collected when all the light emitting units corresponding to the pressing area emit light;
  • the bright area data is the data in the area where the light emitting unit in the light emitting state is located
  • the The data in the dark area is the data in the area where the light-emitting unit in the extinguished state is located
  • the authenticity of the fingerprint to be detected is determined according to the fingerprint characteristic information.
  • the determining fingerprint feature information according to the bright area data and dark area data corresponding to the first fingerprint data and the second fingerprint data includes:
  • fingerprint feature information is determined.
  • the calibration data includes first calibration data corresponding to the first fingerprint data and second calibration data corresponding to the second fingerprint data;
  • the determining the calibration data according to the bright area data and dark area data corresponding to the first fingerprint data and the second fingerprint data includes:
  • the first type of preset data set is A data set obtained when the light-emitting unit partially emits light and used to calibrate the fingerprint data collected when the light-emitting unit partially emits light during the pressing process;
  • the fingerprint feature information includes the repetition degree of fingerprint data and the size relationship between fingerprint data
  • the determining the fingerprint feature information according to the calibration data includes:
  • the first sub-data is row data including bright area data and dark area data, and/or, the first sub-data is row data including bright area data and dark area data column data of zone data;
  • the data amount is used to characterize the difference between the first sub-data and the second sub-data size relationship.
  • the determining the authenticity of the fingerprint to be detected according to the fingerprint feature information includes:
  • both the first sub-data and the second sub-data are valid, if the data amount of the dark area data in the first sub-data is smaller than the fingerprint data at the corresponding position in the second sub-data reaches the first preset If a threshold is set, it is determined that the fingerprint to be detected is a true fingerprint.
  • the first sub-data and the second sub-data are determined according to the degree of repetition between the bright area data in the first sub-data and the fingerprint data in the corresponding position in the second sub-data Are all valid, including:
  • the bright area data in the first sub-data is equal to the fingerprint data at the corresponding position in the second sub-data and the data amount reaches a second preset threshold, then it is determined that the first sub-data and the second sub-data are both efficient.
  • the fingerprint feature information includes the size relationship of the fingerprint data, the average value of the fingerprint data, the difference between the average value of the fingerprint data, the difference between the dark area data in the first sub-data and the fingerprint data in the corresponding position in the second sub-data. degree of difference;
  • the determining the fingerprint feature information according to the calibration data includes:
  • the first sub-data is line data including the bright area data and the dark area data, and/or, the first sub-data is the light field data and column data of said dark field data;
  • the calibration data includes third calibration data corresponding to the first fingerprint data
  • the determining the calibration data according to the bright area data and dark area data corresponding to the first fingerprint data and the second fingerprint data includes:
  • the fingerprint feature information includes quantitative feature information
  • the determining the fingerprint feature information according to the calibration data includes:
  • the quantitative characteristic information includes at least one of data mean value, data variance, data standard deviation, and gradient value.
  • the calibration data further includes third calibration data corresponding to the first fingerprint data
  • the determining the calibration data according to the bright area data and dark area data corresponding to the first fingerprint data and the second fingerprint data further includes:
  • the fingerprint feature information also includes quantitative feature information
  • the determining the fingerprint feature information according to the calibration data also includes:
  • the quantitative characteristic information includes at least one of data mean value, data variance, data standard deviation, and gradient value.
  • the determining the authenticity of the fingerprint to be detected according to the fingerprint feature information includes:
  • the authenticity of the fingerprint to be detected is determined according to the comparison result of the score and the preset fingerprint threshold.
  • the present application provides a fingerprint identification device, including:
  • the fingerprint data acquisition module is used to obtain the first fingerprint data and the second fingerprint data corresponding to the fingerprint to be detected respectively when the pressing operation of the finger in the pressing area is detected, and the first fingerprint data is corresponding to the pressing area.
  • a feature information determination module configured to determine fingerprint feature information according to the bright area data and dark area data corresponding to the first fingerprint data, and the second fingerprint data, the bright area data is where the light emitting unit in the light emitting state is located.
  • the data in the area, the data in the dark area is the data in the area where the light-emitting unit in the extinguished state is located;
  • the fingerprint authenticity determination module is configured to determine the authenticity of the fingerprint to be detected according to the fingerprint feature information.
  • the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and operable on the processor.
  • the processor executes the program, the above-mentioned Fingerprinting method.
  • the present application provides a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and the computer-executable instructions are used to implement the above-mentioned fingerprint identification method when executed by a processor.
  • the fingerprint identification method, device, electronic equipment, and storage medium provided by the present application respectively obtain the first fingerprint data corresponding to the special-shaped light spot and the second fingerprint data corresponding to the normal light spot in a single pressing process, and then obtain the fingerprint data corresponding to the first fingerprint data.
  • the bright area data and dark area data and the fingerprint feature information related to the second fingerprint data because the bright area data and dark area data corresponding to the first fingerprint data can be used to represent the difference between the real finger and the 2.5D fake fingerprint, that is, the real finger
  • the difference in the amount of transmitted light produced by the 2.5D fake fingerprint so that based on the fingerprint feature information, the authenticity of the fingerprint to be detected can be effectively determined, and the accuracy of the fingerprint recognition result for the 2.5D fake fingerprint can be improved.
  • FIG. 1 is a schematic cross-sectional view of an electronic device to which an embodiment of the present application can be applied;
  • Fig. 2 is a schematic diagram of all or part of the light-emitting units corresponding to the fingerprint collection area in the embodiment of the present application;
  • FIG. 3 is a schematic diagram of a real finger pressing the screen in the embodiment of the present application.
  • Fig. 4 is a schematic diagram of a 2.5D fake fingerprint pressing the screen in the embodiment of the present application.
  • FIG. 5 is a schematic diagram of a fingerprint recognition method provided in an embodiment of the present application.
  • FIG. 6 is a schematic diagram of obtaining a first preset data set provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram of obtaining a second preset data set provided by an embodiment of the present application.
  • Fig. 8 is a schematic diagram of selecting the first sub-data from the first calibration data according to the embodiment of the present application.
  • FIG. 9 is a schematic diagram of selecting second sub-data from second calibration data according to an embodiment of the present application.
  • Fig. 10 is an example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • FIG. 11 is another example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • Fig. 12 is another example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • Fig. 13 is another example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • Fig. 14 is a schematic diagram of the first fingerprint data in the embodiment of the present application.
  • Fig. 15 is a schematic diagram of the second fingerprint data in the embodiment of the present application.
  • Fig. 16 is another schematic diagram of selecting the sixth sub-data and the seventh sub-data in the embodiment of the present application.
  • Fig. 17 is another schematic diagram of selecting the sixth sub-data and the seventh sub-data in the embodiment of the present application.
  • FIG. 18 is a schematic diagram of a fingerprint identification device provided by an embodiment of the present application.
  • the 2.5D fake fingerprint produced by the prior art is a type of fake fingerprint that is as easy to be produced as the 2D fake fingerprint.
  • Printing, etc. to make a fake fingerprint mold (Mold), and then use fake fingerprint materials (such as white glue, wood glue, black glue, silica gel, beautifying agent, paint, gelatin, etc.)
  • fake fingerprint materials such as white glue, wood glue, black glue, silica gel, beautifying agent, paint, gelatin, etc.
  • the existing fingerprint recognition schemes can better identify real fingerprints and fake 2D fake fingerprints, but the interception effect on 2.5D fake fingerprints is poor, which seriously affects the information security of end users.
  • the fingerprint identification method, device, electronic equipment and storage medium provided by this application aim to solve the above technical problems in the prior art.
  • the fingerprint recognition device may be specifically an optical fingerprint device, which may be arranged in a local area or the entire area under the display screen, thereby forming an under-display or under-screen optical fingerprint system .
  • the fingerprint identification device receives light returned from the top surface of the display screen of the electronic device, and this returned light carries information of an object (such as a finger) that is in contact with the top surface of the display screen, thereby , by collecting and detecting the returned light to obtain the fingerprint information of the finger.
  • FIG. 1 is a schematic cross-sectional view of an electronic device to which an embodiment of the present application can be applied.
  • the electronic device includes a display screen 12 and a fingerprint identification device 13 , wherein the display screen 12 includes an upper cover 121 , a light emitting layer 122 and a lower cover 123 .
  • the display screen 12 may be a display screen with self-luminous display units, or a non-self-luminous display screen.
  • the display screen 12 is a display screen with a self-luminous display unit.
  • a self-luminous display unit for example, as shown in FIG. , for example, micro-light-emitting diodes (Micro-LEDs) can also be used.
  • the fingerprint recognition device 13 can use the OLED light source corresponding to the location of the fingerprint collection area of the display screen 12 as an excitation light source for fingerprint detection.
  • the light source including the light emitting unit
  • the light beam is reflected on the surface of the object 11 in contact with the screen.
  • Reflected light is formed, and the light beam passes through part of the object 11 to form transmitted light. After the reflected light and transmitted light of different intensities pass through the optical components, they are received by the fingerprint identification device 13 and converted into corresponding electrical signals, that is, fingerprint detection signals. Based on the fingerprint detection signal, fingerprint data can be obtained for realizing the fingerprint recognition function in the electronic device.
  • the display screen 12 is a non-self-illuminating display screen, such as a liquid crystal display screen.
  • the fingerprint identification device 13 needs to use a built-in light source or an external light source as an excitation light source to provide an optical signal for fingerprint detection.
  • the principle of fingerprint detection when a built-in light source or an external light source is used as an excitation light source is the same as the above-mentioned fingerprint detection principle when an OLED display is used, and will not be repeated here.
  • a fingerprint identification device 13 may be disposed under the display screen 12 and may include a multi-optical path structure 131 and an optical detection component 132 .
  • the multi-optical path structure 131 can be arranged above the optical detection component 132 , and is mainly used to guide the light signal reflected or scattered from the finger to the optical detection component for optical detection by the optical detection component 132 .
  • the optical detection unit 132 includes a photosensitive array, a reading circuit and other auxiliary circuits electrically connected to the photosensitive array.
  • the photosensitive array may include a plurality of photosensitive units distributed in an array, which may also be called pixel units or photosensitive pixels. The photosensitive array is mainly used to detect the received light signal, so as to generate fingerprint data through the reading circuit electrically connected with it.
  • the light-emitting unit corresponding to the fingerprint collection area pressed by the object 11 may emit light in whole or in part.
  • full light emission means that all light emitting units in the fingerprint collection area emit light beams, thereby forming a complete light spot
  • partial light emission means that some light emitting units in the fingerprint collection area emit light beams, while the other part of light emitting units do not emit light beams, so that there is no A complete spot is formed, which is called a special-shaped spot.
  • FIG. 2 is a schematic diagram of all or part of the light-emitting units corresponding to the fingerprint collection area in the embodiment of the present application.
  • the fingerprint collection area can be at position 12a of the display screen 12.
  • the corresponding light-emitting units all emit light (white means light-emitting, and the light-emitting area is called a bright area), thereby forming a complete light spot; the fingerprint collection area can also be at the 12b position of the display screen 12.
  • the corresponding light-emitting unit of the fingerprint collection area Partially emit light (black means no light, and the non-light-emitting area is called a dark area), thereby forming a special-shaped light spot; Changes occur, and this application does not specifically limit this.
  • Figure 3 is a schematic diagram of a real finger pressing the screen in the embodiment of the present application
  • Figure 4 is a schematic diagram of a 2.5D fake fingerprint pressing the screen in the embodiment of the present application, as shown in Figures 3 and 4, the corresponding luminescent light in the fingerprint collection area
  • the bright area in the figure indicates the area where the light-emitting unit emits light
  • the dark area indicates the area where the light-emitting unit does not emit light.
  • the photosensitive unit located at the edge of the dark area can receive the surrounding small-angle reflected light (light indicated by the solid line), and the photosensitive unit located at the center of the dark area can receive real finger light. Or the transmitted light of the fake fingerprint material (rays indicated by dotted lines).
  • the thickness of the object (real finger + fake fingerprint material) located above the display screen in Figure 4 is greater than the thickness of the object (real finger) located above the display screen in Figure 3, and the greater the thickness, the darker areas
  • the transmitted light generated when the 2.5D fake fingerprint presses the screen is more than the transmitted light produced when the real finger presses the screen. Therefore, based on this difference, this application can realize Identify the authenticity of fingerprints.
  • the processing steps of the fingerprint identification method in this application can be implemented by a processor inside the electronic device, and the processor can identify the authenticity of the fingerprint based on the fingerprint identification method in this application, so as to perform fingerprint identification based on the authenticity of the fingerprint.
  • the processor determines that the fingerprint is a real fingerprint, it may perform an unlocking operation of the electronic device and the like.
  • Fig. 5 is a schematic diagram of the fingerprint identification method provided by the embodiment of the present application. The method is explained by taking the execution subject of the method as an example of the processor inside the electronic device. As shown in Fig. 5, the method mainly includes the following steps:
  • the first fingerprint data and the second fingerprint data corresponding to the fingerprint to be detected are obtained respectively, the first fingerprint data is the data collected when the light emitting unit part corresponding to the pressing area emits light, The second fingerprint data is the data collected when all the light emitting units corresponding to the pressing area emit light.
  • the first fingerprint data and the second fingerprint data acquired by the processor can be obtained by the optical detection component in the fingerprint identification device according to the multi-channel optical signal guided by the multi-channel optical structure, and the first fingerprint data and the second fingerprint data are objects Two copies of fingerprint data collected during a single press.
  • the light emitting unit corresponding to the fingerprint collection area can sequentially perform partial and full light emission under the action of the control signal, thereby sequentially obtaining the first fingerprint data and the second fingerprint data. fingerprint data.
  • the light-emitting unit may first emit light entirely, and then partly emit light, so as to sequentially obtain the second fingerprint data and the first fingerprint data.
  • the first fingerprint data collected when the light-emitting unit corresponding to the fingerprint collection area is partially illuminated since the light spot formed by the partial light emission is a special-shaped light spot, the first fingerprint data can also be called special-shaped light spot data.
  • the second fingerprint data collected when all the light-emitting units corresponding to the fingerprint collection area emit light since the light spot formed by all the light-emitting units is a complete light spot, the second fingerprint data can also be called normal light spot data.
  • the bright area data is the data in the area where the light emitting unit is in the light emitting state
  • the dark area data is the data in the extinguished state The data in the area where the light-emitting unit of the state is located;
  • the processor determines the fingerprint feature information according to the bright area data and dark area data corresponding to the first fingerprint data and the second fingerprint data, so it can be considered that the fingerprint feature information and the bright area data and dark area data corresponding to the first fingerprint data
  • the data is related to the second fingerprint data, because the difference between the real finger and the 2.5D fake fingerprint is mainly reflected in the amount of transmitted light generated, and this difference can be reflected by the bright area data and dark area data corresponding to the first fingerprint data, thus,
  • the fingerprint feature information obtained by the processor can be used to determine the authenticity of the fingerprint to be detected.
  • the difference between the real finger and the 2.5D fake fingerprint is mainly reflected in the amount of transmitted light generated, and this difference can be reflected by the data of the bright area and the data of the dark area corresponding to the first fingerprint data. After the bright area data and dark area data corresponding to the data and the fingerprint feature information related to the second fingerprint data, the authenticity of the fingerprint to be detected can be determined according to the fingerprint feature information.
  • This embodiment provides a fingerprint identification method.
  • the first fingerprint data corresponding to the special-shaped light spot and the second fingerprint data corresponding to the normal light spot are respectively obtained, and then the bright area data and the dark area data corresponding to the first fingerprint data are obtained.
  • area data and fingerprint feature information related to the second fingerprint data because the bright area data and dark area data corresponding to the first fingerprint data can be used to represent the difference between the real finger and the 2.5D fake fingerprint, that is, the difference between the real finger and the 2.5D fake fingerprint Due to the difference in the amount of transmitted light generated, the authenticity of the fingerprint to be detected can be effectively determined based on the fingerprint feature information, and the accuracy of fingerprint recognition results for 2.5D fake fingerprints can be improved.
  • the fingerprint feature information is determined according to the bright area data and dark area data corresponding to the first fingerprint data, and the second fingerprint data, including:
  • the light-emitting unit of the display screen when performing fingerprint recognition under the screen, will emit light signals to objects placed on the display screen.
  • some of the light signals emitted by the light-emitting unit of the display screen (also known as screen light leakage) are directly downward and are received by the optical detection component through multiple refraction, reflection and diffraction.
  • the optical detection component generates fingerprint original data according to the received light signal.
  • the original fingerprint data includes not only fingerprint information, but also interference information (also referred to as noise floor) such as background noise of the display screen (such as screen light leakage), and the noise floor of each optical detection component is different. Therefore, by performing calibration processing on the original fingerprint data, for example, regularization processing, the influence of the noise floor difference of the optical detection components on the original fingerprint data can be eliminated, so as to ensure the accuracy of the fingerprint data.
  • the calibration data includes first calibration data corresponding to the first fingerprint data and second calibration data corresponding to the second fingerprint data;
  • the processor can perform regularization processing on the first fingerprint data and the second fingerprint data respectively, so that the influence of the difference in the noise floor of the optical detection component on the first fingerprint data and the second fingerprint data can be eliminated, so that The accuracy of the first fingerprint data and the second fingerprint data is guaranteed.
  • the calibration data is determined, including:
  • the first type of preset data set includes a first preset data set and a second preset data set, and the first preset data set and the second preset data set are acquired when the light-emitting unit partially emits light for use in Calibrate the data set of the fingerprint data collected when the light-emitting unit partially emits light during the pressing process;
  • Fig. 6 is a schematic diagram of obtaining the first preset data set provided by the embodiment of the present application.
  • the first preset data set may be simulated by a flesh-colored flat-headed fingerprint model 51 during the calibration phase of the optical detection component.
  • the flesh-colored flat-head fingerprint model 51 is used to simulate a user's finger without fingerprints, that is, the flesh-colored flat-head fingerprint model 51 is equivalent to a finger full of fingerprint valleys.
  • the flesh-colored flat-headed fingerprint model 51 can be pressed on the fingerprint collection area (that is, the local area of the display screen 12 corresponding to the optical detection component), and the light-emitting unit part corresponding to the fingerprint collection area emits light.
  • the optical detection component determines the first preset data set according to the received optical signal. Since the flesh-colored flat-head fingerprint model 51 is equivalent to a finger full of fingerprint valleys, the first preset data set not only contains information related to the light reflected by the central concave surface of the flesh-colored flat-head fingerprint model 51, but also includes the background noise of the display screen 12. (such as the light leakage information of the display screen 12) and other disturbing information.
  • Fig. 7 is a schematic diagram of obtaining the second preset data set provided by the embodiment of the present application.
  • the second preset data set may be a black flat-headed fingerprint model 52 used to simulate the user's finger during the calibration stage of the optical detection component And the data set obtained.
  • the black flat head fingerprint model 52 is used to simulate the pressing state without finger touch.
  • the black flat-headed fingerprint model 52 can be pressed on the fingerprint collection area, and when the light-emitting unit corresponding to the fingerprint collection area is in the same part of the light-emitting situation, the optical detection component determines the first according to the received light signal. Two preset data sets.
  • the second fingerprint data only includes the background noise of the display screen (such as the light leakage information of the display screen 12) and other interference information, that is, only includes the bottom of the optical detection part. noise.
  • Subtracting the second preset data set from the first preset data set can obtain fingerprint data that does not contain background noise.
  • the first preset data set and the second preset data set are used to perform regularization processing on the first fingerprint data corresponding to the fingerprint to be detected, thereby eliminating the influence of the noise floor of different optical detection components on the first fingerprint data, Improve the accuracy of fingerprint recognition.
  • the first preset data set and the second preset data set are represented as H_Fleshspot and H_blackspot respectively, and the first fingerprint data of the fingerprint to be detected is represented as Spotdata, then the first fingerprint data can be regularized by the following formula Processing:
  • the first calibration data corresponding to the first fingerprint data can be calculated. It should be noted that the calculation method here is only to illustrate the specific principle of regularization processing, and matrix operation can also be used in actual calculation to improve the processing speed.
  • the regularization processing method performs regularization processing on the first fingerprint data.
  • the second type of preset data set includes a third preset data set and a fourth preset data set
  • the third preset data set and the fourth preset data set are acquired when all light-emitting units emit light for use in Calibrate the data set of the fingerprint data collected when all the light-emitting units emit light during the pressing process.
  • the basic principle of the process of obtaining the third preset data set can refer to the process of obtaining the first preset data set above, the difference is that when obtaining the third preset data set, the flesh-colored flat head
  • the fingerprint model 51 is pressed on the fingerprint collection area (that is, the local area of the display screen 12 corresponding to the optical detection part), and when all the light-emitting units corresponding to the fingerprint collection area emit light, the optical detection part determines the third according to the received light signal.
  • the third preset data set not only includes information related to the light reflected by the central concave surface of the flesh-colored flat head fingerprint model 51 , but also includes background noise of the display screen 12 (such as light leakage information of the display screen 12 ) and other interference information.
  • the basic principle of the process of obtaining the fourth preset data set can refer to the process of obtaining the second preset data set above, the difference is that when obtaining the fourth fingerprint data set, the black flat head fingerprint model 52 can be pressed In the fingerprint collection area, when the light emitting units corresponding to the fingerprint collection area are all emitting light, the optical detection component determines the fourth preset data set according to the received light signal.
  • the fourth fingerprint data only includes interference information such as background noise of the display screen (such as light leakage information of the display screen 12 ), that is, only includes the background noise of the optical detection component.
  • Subtracting the fourth preset data set from the third preset data set can obtain fingerprint data that does not contain background noise.
  • the third preset data set and the fourth preset data set are used to perform regularization processing on the second fingerprint data corresponding to the fingerprint to be detected, thereby eliminating the influence of the noise floor of different optical detection components on the second fingerprint data, Improve the accuracy of fingerprint recognition.
  • the third preset data set and the fourth preset data set are represented as H_Flesh and H_black respectively, and the second fingerprint data of the fingerprint to be detected is represented as Nordata, then the second fingerprint data can be regularized by the following formula Processing:
  • the second calibration data corresponding to the second fingerprint data can be calculated. It should be noted that the calculation method here is only to illustrate the specific principle of regularization processing, and matrix operation can also be used in actual calculation to improve the processing speed.
  • the regularization processing method performs regularization processing on the second fingerprint data.
  • the first calibration data and the second calibration data are obtained.
  • the dark area data in the first calibration data is usually larger than the fingerprint data at the corresponding position in the second calibration data; if it is a real finger, the dark area data in the first calibration data is usually smaller than the fingerprint at the corresponding position in the second calibration data Therefore, based on the above principles, the authenticity identification of the fingerprint to be detected can be realized.
  • the fake fingerprint material when there is a large difference in the amount of transmitted light generated by the fake fingerprint material and the real finger, for example, the fake fingerprint material is relatively thick, and the authenticity of the fingerprint can be identified directly based on the data characteristics of the fingerprint data itself.
  • the fingerprint feature information includes the degree of repetition of the fingerprint data and the size relationship between the fingerprint data, wherein the degree of repetition is used to characterize the similarity between the first calibration data and the second calibration data, and the size relationship is the first calibration data.
  • the data is related to the data size of the second calibration data.
  • determine the fingerprint feature information including:
  • the first sub-data is the row data including the bright area data and the dark area data, and/or the first sub-data is the column including the bright area data and the dark area data data;
  • FIG. 8 is a schematic diagram of selecting the first sub-data from the first calibration data according to the embodiment of the present application.
  • the first calibration data includes bright area data (the white circle part in the figure) and dark area data (the black circle part in the figure), when selecting the first sub-data, since it is necessary to rely on the dark area data for authenticity identification, the selected first sub-data is row data and / or column data.
  • the mth row of data and/or the nth column of data in FIG. 8 may be selected as the first sub-data.
  • Fig. 9 is a schematic diagram of selecting the second sub-data from the second calibration data according to the embodiment of the present application.
  • the second calibration data only includes bright area data (the white circle part in the figure) when selecting the second sub-data
  • the position of the second sub-data in the second fingerprint data is correspondingly the same as the position of the first sub-data in the first fingerprint data. For example, if the m-th row of data in the first calibration data in FIG. 8 is selected as the first sub-data, then the m-th row of data in the second calibration data in FIG. 9 is also selected as the second sub-data.
  • the processor determines the fingerprint feature information based on the first sub-data and the second sub-data.
  • the fingerprint feature information specifically includes bright area data in the first sub-data, The degree of repetition with the fingerprint data at the corresponding position in the second sub-data, and the data volume of the dark area data in the first sub-data is smaller than the fingerprint data at the corresponding position in the second sub-data, thus, based on the above-mentioned fingerprint feature information, it can be determined Detect the authenticity of fingerprints.
  • determining the authenticity of the fingerprint to be detected includes:
  • the first fingerprint data and the second fingerprint data are two pieces of fingerprint data collected during a single press, the pressed area of the object does not change significantly during the single press, therefore, the first sub-data
  • the similarity between the bright area data in the first sub-data and the fingerprint data in the corresponding position in the second sub-data is relatively high, thus, the repetition degree between the bright area data in the first sub-data and the fingerprint data in the corresponding position in the second sub-data can be used for Determine the validity of the first fingerprint data and the second fingerprint data, that is, the validity of the first sub-data and the second sub-data.
  • both the first sub-data and the second sub-data are valid data
  • the data amount of the dark area data in the first sub-data is smaller than the fingerprint data of the corresponding position in the second sub-data reaches the first preset threshold, then it is determined to The detected fingerprint is a true fingerprint, otherwise, it is determined that the fingerprint to be detected is a false fingerprint.
  • the first sub-data and the second sub-data are both valid, including: if the second sub-data When the bright area data in one sub-data is equal to the fingerprint data at the corresponding position in the second sub-data and reaches a second preset threshold, it is determined that both the first sub-data and the second sub-data are valid.
  • the bright area data in the first sub-data is equal to the fingerprint data at the corresponding position in the second sub-data and reaches the second preset threshold, it means that the bright area data in the first sub-data corresponds to the fingerprint data in the second sub-data.
  • the similarity of the fingerprint data of the position is relatively high, therefore, it can be determined that both the first fingerprint data and the second fingerprint data are valid data, and thus the first sub-data and the second sub-data are also valid data, thus, based on The accuracy and credibility of the fingerprint authenticity identification results obtained from the first sub-data and the second sub-data are relatively high.
  • the bright area data in the first sub-data is equal to the fingerprint data at the corresponding position in the second sub-data, and the two data may be completely equal, or the difference between the two data is less than a certain threshold, Therefore, in the above two cases, it can be considered that the bright area data in the first sub-data is highly similar to the fingerprint data in the corresponding position in the second sub-data.
  • the bright area data in the first sub-data is equal to the fingerprint data at the corresponding position in the second sub-data, and the data volume does not reach the second preset threshold, it means that the bright area data in the first sub-data is different from the fingerprint data in the second sub-data.
  • the similarity of the fingerprint data at the corresponding position is low. It is determined that the first sub-data and the second sub-data are invalid data.
  • the prompt information of the first fingerprint data and the second fingerprint data is acquired.
  • Figure 10 is an example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • the horizontal axis of the coordinate axis indicates the position of the data
  • the vertical axis indicates the size of the data
  • the first curve indicates that the first calibration data contains The first sub-data of a certain line of bright area data and dark area data
  • the second curve represents the second sub-data corresponding to the position in the second calibration data, refer to the curve shown in Figure 10, in the two kinds of curves, located at the two sides of the abscissa
  • the repeatability of the bright area data at the end is high, so both curves are valid data.
  • the first curve has more parts below the second curve, it means that the dark area data in the first sub-data is smaller than the fingerprint data at the corresponding position in the second sub-data.
  • the first preset threshold so it can be determined that the fingerprint to be detected is a true fingerprint.
  • Figure 11 is another example diagram of fingerprint authentication in the embodiment of the present application.
  • the horizontal axis of the coordinate axis represents the position of the data
  • the vertical axis represents the size of the data
  • the first curve represents the first calibration data.
  • the second curve represents the second sub-data of the corresponding position in the second calibration data, refer to the curve shown in Figure 11, of the two curves
  • the horizontal The repeatability of the bright area data at both ends of the coordinates is high, so the two curves are valid data.
  • the first curve has more parts below the second curve, it means that the dark area data in the first sub-data is smaller than the fingerprint data at the corresponding position in the second sub-data.
  • the first preset threshold so it can be determined that the fingerprint to be detected is a true fingerprint.
  • Fig. 12 is another example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • the abscissa of the coordinate axis represents the position of the data
  • the ordinate represents the data size
  • the first curve represents the data in the first calibration data.
  • the second curve represents the second sub-data of the corresponding position in the second calibration data, refer to the curve shown in Figure 12, of the two curves, the horizontal The repeatability of the bright area data at both ends of the coordinates is high, so the two curves are valid data.
  • the dark area data located in the middle of the abscissa because the first curve has more parts above the second curve, it means that the dark area data in the first sub-data is smaller than the fingerprint data at the corresponding position in the second sub-data.
  • the first preset threshold it can be determined that the fingerprint to be detected is a fake fingerprint.
  • Fig. 13 is another example diagram of fingerprint authenticity identification in the embodiment of the present application.
  • the horizontal axis of the coordinate axis represents the position of the data
  • the vertical axis represents the data size
  • the first curve represents the data in the first calibration data.
  • the second curve represents the second sub-data at the corresponding position in the second calibration data, refer to the curve shown in Figure 13, of the two curves, the horizontal The repeatability of the bright area data at both ends of the coordinates is high, so the two curves are valid data.
  • the dark area data located in the middle of the abscissa because the first curve has more parts above the second curve, it means that the dark area data in the first sub-data is smaller than the fingerprint data at the corresponding position in the second sub-data.
  • the first preset threshold it can be determined that the fingerprint to be detected is a fake fingerprint.
  • the authenticity of the fingerprint can be effectively identified, thereby improving the security performance of the electronic device.
  • the fake fingerprint material when the difference in the amount of transmitted light produced by the fake fingerprint material and the real finger is small, for example, the fake fingerprint material is relatively thin and transparent (such as gelatin), and it cannot be accurately performed directly based on the data characteristics of the fingerprint data itself. Authenticity identification of fingerprints. At this time, the authenticity of fingerprints can be identified in combination with deeper data features obtained from fingerprint data.
  • the fingerprint feature information includes the size relationship of the fingerprint data, the average value of the fingerprint data, the difference between the average value of the fingerprint data, the difference between the dark area data in the first sub-data and the fingerprint data in the corresponding position in the second sub-data Spend.
  • determine the fingerprint feature information including:
  • the first sub-data is row data including bright area data and dark area data, and/or the first sub-data is column data of bright area data and dark area data ;
  • the dark area data in the RowSpotData are calculated respectively (for example, from the mth data
  • the second data average value AvgRowNorData of the corresponding position data in RowNorData that is, also from the mth data to the nth data
  • the difference AvgRowSub between AvgRowNorData and AvgRowSpotData.
  • both the first sub-data and the second sub-data are column data, the same processing is performed.
  • the sum sumRow can be used to represent the first The degree of difference between the dark area data in the first sub-data and the fingerprint data in the corresponding position in the second sub-data.
  • both the first sub-data and the second sub-data are column data, the same processing is performed.
  • determining the authenticity of the fingerprint to be detected includes:
  • S314. Determine the authenticity of the fingerprint to be detected according to the comparison result of the score and the preset fingerprint threshold.
  • the decision tree model is trained according to the fingerprint feature information of each fingerprint sample in the fingerprint sample set and the authenticity result of each fingerprint sample.
  • the fingerprint feature information of each fingerprint sample has the same feature type as the fingerprint feature information of the fingerprint to be detected.
  • the feature type of the fingerprint feature information includes the first data mean value of the first sub-data, the second data mean value of the second sub-data, the difference between the first data mean value and the second data mean value, each difference (for RowSpotData The sum of the absolute value of each dark area data of each dark area data and the fingerprint data of the corresponding position in RowNorData, the difference obtained pixel by pixel), or any combination thereof.
  • RowSpotData The sum of the absolute value of each dark area data of each dark area data and the fingerprint data of the corresponding position in RowNorData, the difference obtained pixel by pixel
  • the fingerprint samples in the fingerprint sample set can include, for example, real fingers and fake fingers in various scenarios, for example, real fingers and fake fingers, dry fingers and/or wet fingers in low temperature scenarios, high temperature scenarios, normal temperature scenarios, oily state finger.
  • the fingerprint feature information of the fingerprint sample is the fingerprint feature information determined according to the fingerprint data corresponding to the fingerprint samples in various scenarios.
  • each feature type of fingerprint feature information can be regarded as a decision node, and each decision node is used to classify the fingerprint samples, so as to train the generated decision tree model, which can include the same The judgment threshold and weight corresponding to each feature type.
  • the fingerprint feature information generated according to the fingerprint data corresponding to the fingerprint to be detected is input into the pre-trained decision tree model, and the pre-trained decision tree model generates the score of the fingerprint to be detected according to the predetermined judgment threshold and weight .
  • the feature type of the fingerprint feature information includes the first data mean value of the first sub-data, the second data mean value of the second sub-data, the difference between the first data mean value and the second data mean value, and the absolute value of each difference
  • the first data mean of the first sub-data corresponding to each fingerprint sample in the fingerprint sample set, the second data mean of the second sub-data, the first data mean and the second The sum of the difference between the data mean and the absolute value of each difference is used as the fingerprint feature information corresponding to the fingerprint sample, and the extracted fingerprint feature information of each fingerprint sample and the authenticity structure corresponding to each fingerprint sample are input into the decision tree model
  • the trained prediction model includes a first judgment threshold corresponding to the first data mean value of the first sub-data and a first weight, and a second judgment threshold corresponding to the second data mean value of the second sub-data
  • the trained decision tree model is based on the comparison result of the first data mean of the first sub-data of the fingerprint to be detected and the first judgment threshold, the first data mean of the first sub-data of the fingerprint to be detected and the first judgment threshold.
  • the comparison result of the two judgment thresholds, the comparison result of the difference between the first data mean value and the second data mean value of the fingerprint to be detected and the third judgment threshold, the sum of the absolute values of each difference of the fingerprint to be detected and the fourth judgment threshold The comparison results, and the weights corresponding to each judgment condition determine the score used to indicate that the fingerprint to be detected is a true fingerprint.
  • the preset fingerprint threshold can be flexibly set according to the user's security level requirements. For example, in application scenarios with lower security level requirements, such as in the application scenario of unlocking electronic devices through fingerprint verification, the preset fingerprint threshold can be set to Set it relatively low, such as 0.5. However, in an application scenario requiring a higher security level, for example, in an application scenario in which fee payment is performed through fingerprint verification, the preset fingerprint threshold can be set relatively high, such as 0.7.
  • a pre-trained decision tree model is used for scoring, and the authenticity of the fingerprint to be detected is determined according to the comparison result of the score and the preset fingerprint threshold, so that the true and false fingerprints can be accurately identified.
  • the calibration data includes third calibration data corresponding to the first fingerprint data
  • the calibration data is determined, including:
  • Figure 14 is a schematic diagram of the first fingerprint data in the embodiment of the present application, wherein the data in the solid line circle is the dark area data corresponding to the first fingerprint data, and the data outside the solid line circle and in the box are the data corresponding to the first fingerprint data
  • the third sub-data one or more sub-data in the bright-area data can be selected, for example, P11, P21, P31, and P41 in FIG. 14 can be used as the third sub-data.
  • Figure 15 is a schematic diagram of the second fingerprint data in the embodiment of the present application, wherein the data in the dotted circle is the fingerprint data of the corresponding position of the dark area data in the first fingerprint data in the second fingerprint data, outside the dotted circle and in the box
  • the data is the fingerprint data of the corresponding position of the bright area data in the first fingerprint data in the second fingerprint data.
  • P11, P21, P31, and P41 in FIG. 14 are used as the third sub-data
  • P12, P22, P32, and P42 in FIG. 15 can be used as the fourth sub-data.
  • the data mean values corresponding to P11, P21, P31, and P41 are represented by AvgSpotPart1, AvgSpotPart2, AvgSpotPart3, AvgSpotPart4 respectively, and P12, P22, P32,
  • the data mean values corresponding to P42 are represented by AvgNorPart1, AvgNorPart2, AvgNorPart3, AvgNorPart4 respectively, then the ratio of the third data mean value to the fourth data mean value includes K1, K2, K3, K4, where:
  • the first fingerprint data is regularized by the following formula to obtain the third calibration data SpotNomalized:
  • the accuracy of the fingerprint data can be improved, thereby improving the accuracy of the fingerprint recognition result.
  • determining the authenticity of the fingerprint to be detected includes:
  • S315. Determine quantitative feature information of the bright area data and/or dark area data corresponding to the third calibration data, where the quantitative feature information includes at least one of data mean, data variance, data standard deviation, and gradient value;
  • S317 Determine the authenticity of the fingerprint to be detected according to the comparison result of the score and the preset fingerprint threshold.
  • the fifth sub-data and the sixth sub-data can be selected from the dark area data of the third calibration data, and the first The fifth data mean, data variance, and data standard deviation of the fifth sub-data, and the sixth data mean, data variance, and data standard deviation of the sixth sub-data, and the positions of the fifth sub-data and the sixth sub-data are different;
  • the data at the same position as C1 in FIG. 14 in the third calibration data can be selected as the fifth sub-data, that is, the center position data of the dark area data in the third calibration data.
  • the data in the third calibration data at the same position as L1, R1, U1, and D1 in FIG. 14 can be selected as the sixth sub-data, that is, the edge position data of the dark area data in the third calibration data.
  • the corresponding data mean, data variance, and data standard deviation are respectively calculated.
  • the fifth data mean avgcenter, variance varCenter, etc. corresponding to the fifth sub-data
  • the sixth data mean avgL/avgR/avgU/avgD, variance varL/varR/varU/varD etc. corresponding to the sixth sub-data.
  • the seventh sub-data may also be selected from the bright area data of the third calibration data, and the first sub-data of the seventh sub-data may be determined.
  • the data in the third calibration data having the same positions as P11 , P21 , P31 , and P41 in FIG. 14 can be selected as the seventh sub-data.
  • the corresponding data mean, data variance, and data standard deviation are respectively calculated.
  • the sixth data mean and the seventh data mean it is also possible to determine the gradient between different regions in the third calibration data based on the fifth data mean, the sixth data mean and the seventh data mean value, that is, two gradients in four directions from the peripheral area to the central area.
  • Fig. 16 and Fig. 17 are another two schematic diagrams for selecting the sixth sub-data and the seventh sub-data in the embodiment of the present application, the specific positions of the sixth sub-data and the seventh sub-data can be adjusted, and similarly, the specific shapes and The specific amount can also be adjusted.
  • a pre-trained decision tree model is used for scoring, and the authenticity of the fingerprint to be detected is determined according to the comparison result of the score and the preset fingerprint threshold, so that the true and false fingerprints can be accurately identified.
  • the processor may also perform regularization processing on the first fingerprint data and the second fingerprint data respectively, and obtain the size relationship of the fingerprint data of the first calibration data and the second calibration data, the average value of the fingerprint data, The difference between the average value of the fingerprint data, the degree of difference between the dark area data in the first sub-data and the fingerprint data at the corresponding position in the second sub-data, and, by performing regularization processing on the first fingerprint data, and obtaining the third calibration Quantified feature information of the bright area data and/or dark area data corresponding to the data, and then input the above fingerprint feature information into the pre-trained decision tree model to obtain a score for indicating that the fingerprint to be detected is a true fingerprint; according to the The comparison result of the score and the preset fingerprint threshold value is used to determine the authenticity of the fingerprint to be detected.
  • scoring judgment is performed by combining various types of fingerprint feature information, which can further improve the accuracy of fingerprint authenticity identification results.
  • a fingerprint identification device is provided.
  • Fig. 18 is a schematic diagram of a fingerprint identification device provided by the embodiment of the present application. As shown in Fig. 18, the device includes:
  • the fingerprint data acquiring module 100 is configured to acquire the first fingerprint data and the second fingerprint data corresponding to the fingerprint to be detected when the pressing operation of the finger in the pressing area is detected, the first fingerprint data is the part of the light emitting unit corresponding to the pressing area The data collected when emitting light, the second fingerprint data is the data collected when all the light emitting units corresponding to the pressing area emit light;
  • the feature information determination module 200 is configured to determine the fingerprint feature information according to the bright area data and dark area data corresponding to the first fingerprint data and the second fingerprint data, the bright area data is the data in the area where the light emitting unit in the light emitting state is located, The data in the dark area is the data in the area where the light-emitting unit in the extinguished state is located;
  • the fingerprint authenticity determination module 300 is configured to determine the authenticity of the fingerprint to be detected according to the fingerprint feature information.
  • Each module in the above-mentioned fingerprint identification device can be fully or partially realized by software, hardware and a combination thereof.
  • the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, and can also be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of the above modules.
  • the present application provides a fingerprint identification device, which obtains the first fingerprint data corresponding to the special-shaped light spot and the second fingerprint data corresponding to the normal light spot in a single pressing process, and then obtains the bright area data and dark area corresponding to the first fingerprint data data and the fingerprint feature information related to the second fingerprint data, because the bright area data and dark area data corresponding to the first fingerprint data can be used to represent the difference between the real finger and the 2.5D fake fingerprint, that is, the difference between the real finger and the 2.5D fake fingerprint Therefore, based on the fingerprint feature information, the authenticity of the fingerprint to be detected can be effectively determined, and the accuracy of the fingerprint recognition result for the 2.5D fake fingerprint can be improved.
  • an electronic device including: a memory, a processor, and a computer program stored on the memory and operable on the processor, and the above-mentioned fingerprint identification method is implemented when the processor executes the program.
  • the memory and the processor are electrically connected directly or indirectly to realize data transmission or interaction.
  • these components may be electrically connected to each other through one or more communication buses or signal lines, for example, they may be connected through a bus.
  • Computer-executed instructions for implementing the data access control method are stored in the memory, including at least one software function module that can be stored in the memory in the form of software or firmware.
  • the processor runs the software programs and modules stored in the memory to execute various Functional application and data processing.
  • the memory can be, but not limited to, random access memory (Random Access Memory, RAM), read-only memory (Read Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), erasable only memory Read memory (Erasable Programmable Read-Only Memory, EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), etc.
  • RAM Random Access Memory
  • ROM read-only memory
  • PROM programmable read-only memory
  • PROM Programmable Read-Only Memory
  • EPROM Erasable Programmable Read-Only Memory
  • EEPROM Electric Erasable Programmable Read-Only Memory
  • the memory is used to store programs, and the processor executes the programs after receiving execution instructions.
  • the software programs and modules in the memory may also include an operating system, which may include various software components and/or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may Communicate with various hardware or software components to provide an operating environment for other software components.
  • the processor can be an integrated circuit chip with signal processing capability.
  • the above-mentioned processor may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP) and the like.
  • CPU Central Processing Unit
  • NP Network Processor
  • Various methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed.
  • a general-purpose processor may be a microprocessor, or the processor may be any conventional processor, or the like.
  • a computer-readable storage medium is provided, and computer-executable instructions are stored in the computer-readable storage medium, and the computer-executable instructions are used to implement the steps of the various method embodiments of the present application when executed by a processor.
  • a computer program product including a computer program, and when the computer program is executed by a processor, the steps of the various method embodiments of the present application are implemented.
  • Nonvolatile memory can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory.
  • Volatile memory can include random access memory (RAM) or external cache memory.
  • RAM is available in many forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Chain Synchlink DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
  • SRAM Static RAM
  • DRAM Dynamic RAM
  • SDRAM Synchronous DRAM
  • DDRSDRAM Double Data Rate SDRAM
  • ESDRAM Enhanced SDRAM
  • SLDRAM Synchronous Chain Synchlink DRAM
  • Rambus direct RAM
  • DRAM direct memory bus dynamic RAM
  • RDRAM memory bus dynamic RAM

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Collating Specific Patterns (AREA)
  • Image Input (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

本申请提供一种指纹识别方法、装置、电子设备及存储介质,在单次按压过程中分别得到异形光斑对应的第一指纹数据以及正常光斑对应的第二指纹数据,进而得到与第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据相关的指纹特征信息,由于第一指纹数据对应的亮区数据和暗区数据可以用于表征真手指与2.5D假指纹的差异,即真手指与2.5D假指纹所产生的透射光的量的差异,从而,基于指纹特征信息可以有效确定待检测指纹的真伪,提高对于2.5D假指纹的指纹识别结果的准确性。

Description

指纹识别方法、装置、电子设备及存储介质 技术领域
本申请涉及指纹识别技术领域,尤其涉及一种指纹识别方法、装置、电子设备及存储介质。
背景技术
随着光学指纹识别技术在终端设备中的广泛应用,用户对指纹识别的安全性要求越来越高。
现有的指纹识别方案,可以较好地识别真指纹和伪造的平面假指纹(不具备立体指纹特征的指纹,也称为2D假指纹)。然而,由于根据提取的用户指纹通过腐蚀电路板等简单工艺制作的假指纹(也称为2.5D假指纹)具有3D深度特征,使得现有的指纹识别方案对此类假指纹的拦截效果较差,严重影响终端用户的信息安全。
因此,如何识别真指纹与2.5D假指纹,以提升指纹识别的安全性是一项亟需解决的问题。
发明内容
本申请提供一种指纹识别方法、装置、电子设备及存储介质,用以解决现有技术存在的问题。
第一方面,本申请提供一种指纹识别方法,包括:
在检测到手指在按压区域内的按压操作时,分别获取待检测指纹对应的第一指纹数据以及第二指纹数据,所述第一指纹数据为所述按压区域对应的发光单元部分发光时采集的数据,所述第二指纹数据为所述按压区域对应的发光单元全部发光时采集的数据;
根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定指纹特征信息,所述亮区数据为处于发光状态的发光单元所在区域内的数据,所述暗区数据为处于熄灭状态的发光单元所在区域内的数据;
根据所述指纹特征信息,确定所述待检测指纹的真伪。
在一些实施例中,所述根据所述第一指纹数据对应的亮区数据和暗区数 据,以及所述第二指纹数据,确定指纹特征信息,包括:
根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据;
根据所述校准数据,确定指纹特征信息。
在一些实施例中,所述校准数据包括第一指纹数据对应的第一校准数据和所述第二指纹数据对应的第二校准数据;
所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据,包括:
通过第一类预设数据集合对所述第一指纹数据对应的亮区数据和暗区数据进行正则化处理,得到所述第一校准数据,其中,所述第一类预设数据集合为在所述发光单元部分发光时获取的、并用于标定在按压过程中发光单元部分发光时采集得到的指纹数据的数据集合;
通过第二类预设数据集合对所述第二指纹数据进行正则化处理,得到所述第二校准数据,其中,所述第一类预设数据集合为在所述发光单元全部发光时获取的、并用于标定在按压过程中发光单元全部发光时采集得到的指纹数据的数据集合。
在一些实施例中,所述指纹特征信息包括指纹数据的重复度以及指纹数据之间的大小关系;
所述根据所述校准数据,确定指纹特征信息,包括:
从所述第一校准数据中选择第一子数据,所述第一子数据为包含亮区数据以及暗区数据的行数据,和/或,所述第一子数据为包含亮区数据以及暗区数据的列数据;
从所述第二校准数据中选择第二子数据,所述第二子数据在所述第二指纹数据中的位置与所述第一子数据在所述第一指纹数据中的位置相同;
确定所述第一子数据中的亮区数据与所述第二子数据中对应位置的指纹数据的重复度;
确定所述第一子数据中暗区数据小于所述第二子数据中对应位置的指纹数据的数据量,所述数据量用于表征所述第一子数据与所述第二子数据之间的大小关系。
在一些实施例中,所述根据所述指纹特征信息,确定所述待检测指纹的 真伪,包括:
根据所述第一子数据中的亮区数据,与所述第二子数据中对应位置的指纹数据的重复度,确定所述第一子数据以及所述第二子数据是否均有效;
在确定所述第一子数据以及所述第二子数据均有效时,若所述第一子数据中暗区数据小于所述第二子数据中对应位置的指纹数据的数据量达到第一预设阈值,则确定所述待检测指纹为真指纹。
在一些实施例中,根据所述第一子数据中的亮区数据,与所述第二子数据中对应位置的指纹数据的重复度,确定所述第一子数据以及所述第二子数据是否均有效,包括:
若所述第一子数据中亮区数据等于所述第二子数据中对应位置的指纹数据的数据量达到第二预设阈值,则确定所述第一子数据与所述第二子数据均有效。
在一些实施例中,所述指纹特征信息包括指纹数据的大小关系、指纹数据均值、指纹数据均值的差值、第一子数据中暗区数据与第二子数据中对应位置的指纹数据之间的差异度;
所述根据所述校准数据,确定指纹特征信息,包括:
从所述第一校准数据中选择第一子数据,所述第一子数据为包括所述亮区数据和所述暗区数据的行数据,和/或,所述第一子数据为所述亮区数据和所述暗区数据的列数据;
从所述第二校准数据中选择第二子数据,所述第二子数据在所述第二指纹数据中的位置与所述第一子数据在所述第一指纹数据中的位置相同;
确定所述第一子数据中暗区数据小于所述第二子数据中对应位置的指纹数据的数据量;
确定所述第一子数据中暗区数据的第一数据均值、所述第二子数据中对应位置的数据的第二数据均值、以及所述第一数据均值与所述第二数据均值的差值;
确定所述第一子数据中的每个暗区数据与所述第二子数据中对应位置的指纹数据的差值,并确定每个差值的绝对值的总和,所述总和用于表征所述第一子数据与所述第二子数据的差异度。
在一些实施例中,所述校准数据包括第一指纹数据对应的第三校准数据;
所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据,包括:
从所述第一指纹数据对应的亮区数据中选择第三子数据,确定所述第三子数据的第三数据均值;
从所述第二指纹数据中选择第四子数据,确定所述第四子数据的第四数据均值,所述第四子数据在所述第二指纹数据中的位置与所述第三子数据在所述第一指纹数据中的位置对应相同;
确定所述第三数据均值与所述第四数据均值的比值;
基于所述第一指纹数据的亮区数据和暗区数据、所述第二指纹数据以及所述比值,对所述第一指纹数据进行正则化处理,得到所述第三校准数据。
在一些实施例中,所述指纹特征信息包括量化特征信息;
所述根据所述校准数据,确定指纹特征信息,包括:
确定所述第三校准数据对应的亮区数据和/或暗区数据的量化特征信息,所述量化特征信息包括数据均值、数据方差、数据标准差以及梯度值中的至少一项。
在一些实施例中,所述校准数据还包括第一指纹数据对应的第三校准数据;
所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据,还包括:
从所述第一指纹数据对应的亮区数据中选择第三子数据,确定所述第三子数据的第三数据均值;
从所述第二指纹数据中选择第四子数据,确定所述第四子数据的第四数据均值,所述第四子数据在所述第二指纹数据中的位置与所述第三子数据在所述第一指纹数据中的位置对应相同;
确定所述第三数据均值与所述第四数据均值的比值;
基于所述第一指纹数据的亮区数据和暗区数据、所述第二指纹数据以及所述比值,对所述第一指纹数据进行正则化处理,得到所述第三校准数据。
在一些实施例中,所述指纹特征信息还包括量化特征信息;
所述根据所述校准数据,确定指纹特征信息,还包括:
确定所述第三校准数据对应的亮区数据和/或暗区数据的量化特征信息, 所述量化特征信息包括数据均值、数据方差、数据标准差以及梯度值中的至少一项。
在一些实施例中,所述根据所述指纹特征信息,确定所述待检测指纹的真伪,包括:
将所述指纹特征信息输入至预先训练的决策树模型,得到用于指示所述待检测指纹为真指纹的得分;
根据所述得分和预设指纹阈值的比较结果,确定所述待检测指纹的真伪。
第二方面,本申请提供一种指纹识别装置,包括:
指纹数据获取模块,用于在检测到手指在按压区域内的按压操作时,分别获取待检测指纹对应的第一指纹数据以及第二指纹数据,所述第一指纹数据为所述按压区域对应的发光单元部分发光时采集的数据,所述第二指纹数据为所述按压区域对应的发光单元全部发光时采集的数据;
特征信息确定模块,用于根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定指纹特征信息,所述亮区数据为处于发光状态的发光单元所在区域内的数据,所述暗区数据为处于熄灭状态的发光单元所在区域内的数据;
指纹真伪确定模块,用于根据所述指纹特征信息,确定所述待检测指纹的真伪。
第三方面,本申请提供一种电子设备,包括:存储器,处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述处理器执行所述程序时实现上述的指纹识别方法。
第四方面,本申请提供一种计算机可读存储介质,所述计算机可读存储介质中存储有计算机执行指令,所述计算机执行指令被处理器执行时用于实现上述的指纹识别方法。
本申请提供的指纹识别方法、装置、电子设备及存储介质,在单次按压过程中分别得到异形光斑对应的第一指纹数据以及正常光斑对应的第二指纹数据,进而得到与第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据相关的指纹特征信息,由于第一指纹数据对应的亮区数据和暗区数据可以用于表征真手指与2.5D假指纹的差异,即真手指与2.5D假指纹所产生的透射光的量的差异,从而,基于指纹特征信息可以有效确定待检测指纹的真伪, 提高对于2.5D假指纹的指纹识别结果的准确性。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本公开的实施例,并与说明书一起用于解释本公开的原理。
图1为本申请实施例可以适用的电子设备的剖面示意图;
图2为本申请实施例中指纹采集区域对应的发光单元全部发光或者部分发光的示意图;
图3为本申请实施例中真手指按压屏幕时的示意图;
图4为本申请实施例中2.5D假指纹按压屏幕时的示意图;
图5为本申请实施例提供的指纹识别方法的示意图;
图6为本申请实施例提供的获取第一预设数据集合的示意图;
图7为本申请实施例提供的获取第二预设数据集合的示意图;
图8为本申请实施例从第一校准数据中选择第一子数据的示意图;
图9为本申请实施例从第二校准数据中选择第二子数据的示意图;
图10为本申请实施例中进行指纹真伪识别的示例图;
图11为本申请实施例中进行指纹真伪识别的另一示例图;
图12为本申请实施例中进行指纹真伪识别的另一示例图;
图13为本申请实施例中进行指纹真伪识别的另一示例图;
图14为本申请实施例中第一指纹数据的示意图;
图15为本申请实施例中第二指纹数据的示意图;
图16为本申请实施例中选择第六子数据、第七子数据的另一示意图;
图17为本申请实施例中选择第六子数据、第七子数据的另一示意图;
图18为本申请实施例提供的指纹识别装置的示意图。
通过上述附图,已示出本公开明确的实施例,后文中将有更详细的描述。这些附图和文字描述并不是为了通过任何方式限制本公开构思的范围,而是通过参考特定实施例为本领域技术人员说明本公开的概念。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚,下面将结合本发 明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
在本申请实施例中使用的术语是仅仅出于描述特定实施例的目的,而非旨在限制本发明。在本申请实施例中所使用的单数形式的“一种”和“该”也旨在包括多数形式,除非上下文清楚地表示其他含义。
应当理解,本文中使用的术语“和/或”仅仅是一种描述关联对象的关联关系,表示可以存在三种关系,例如,A和/或B,可以表示:单独存在A,同时存在A和B,单独存在B这三种情况。另外,本文中字符“三种,一般表示前后关联对象是一种“或”的关系。
还需要说明的是,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的商品或者系统不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种商品或者系统所固有的要素。在没有更多限制的情况下,由语句“包括一个……”限定的要素,并不排除在包括所述要素的商品或者系统中还存在另外的相同要素。
现有技术所制作的2.5D假指纹是一种和2D假指纹一样容易被制作的假指纹类型,首先将提取到指纹采用简单的工艺(普通打印机打印在胶片纸上、腐蚀线路板工艺、UV打印等)制作成假指纹模具(Mold),然后用假指纹制作材料(例如白胶,木胶,黑胶,硅胶,美缝剂,油漆,明胶等)进行拓印,待定型后取下即得到具有3D立体特征的2.5D假指纹。
在镜头式指纹识别系统中,现有的指纹识别方案可以较好地识别真指纹和伪造的2D假指纹,但是对2.5D假指纹的拦截效果较差,严重影响终端用户的信息安全。
本申请提供的指纹识别方法、装置、电子设备及存储介质,旨在解决现有技术的如上技术问题。
本申请实施例提供的技术方案可以应用于各种电子设备。例如,智能手机、平板电脑以及其他具有显示屏和指纹识别装置的移动终端或者其他电子设备。更具体地,在上述电子设备中,指纹识别装置可以具体为光学指纹装 置,其可以设置在显示屏下方的局部区域或者全部区域,从而形成屏下(Under-display或Under-screen)光学指纹系统。具体地,在该电子设备中,指纹识别装置接收从电子设备的显示屏的顶面返回的光,这种返回的光携带有与显示屏的顶面接触的物体(例如手指)的信息,从而,通过采集和检测这种返回来的光来获取手指的指纹信息。
图1为本申请实施例可以适用的电子设备的剖面示意图。如图1所示,电子设备包括显示屏12和指纹识别装置13,其中,显示屏12包括上盖板121、发光层122和下盖板123。具体的,根据发光层不同,显示屏12可以是具有自发光显示单元的显示屏,也可以是非自发光的显示屏。
在显示屏12是具有自发光显示单元的显示屏,例如,如图1所示,显示屏12可以为采用有机发光二极管(Organic Light-Emitting Diode,OLED)显示屏,然而,本申请不限于此,例如也可以采用微型发光二极管(Micro-LED)。在采用OLED显示屏时,指纹识别装置13可以利用显示屏12的与指纹采集区域位置对应的OLED光源作为指纹检测的激励光源。当物体11按压在显示屏12的指纹采集区域时,显示屏12中对应位置的光源(包括发光单元)向指纹采集区域上方的物体11发射光束,该光束在物体11与屏幕接触的表面发生反射形成反射光,以及,光束透过部分物体11形成透射光。不同强度的反射光以及透射光经过光学部件后,由指纹识别装置13接收并转换为相应的电信号,即指纹检测信号。基于该指纹检测信号,可以获得指纹数据,用于在电子设备实现指纹识别功能。
另外,在显示屏12是非自发光的显示屏,例如液晶显示屏。指纹识别装置13需要采用内置光源或外置光源作为激励光源,以提供用于进行指纹检测的光信号。在采用内置光源或外置光源作为激励光源时的指纹检测原理与上面提及的采用OLED显示屏时的指纹检测原理相同,此处不再赘述。
参考图1,指纹识别装置13,具体地为光学指纹识别装置,可以设置在显示屏12下方,并且可以包括多光路结构131和光学检测部件132。其中,多光路结构131可以设置在光学检测部件132的上方,主要用于将从手指处反射或散射的光信号引导至光学检测部件以由光学检测部件132进行光学检测。光学检测部件132包括感光阵列和与感光阵列电连接的读取电路及其他辅助电路。感光阵列可以包括多个呈阵列分布的多个感光单元,其也可以称 为像素单元或感光像素。感光阵列主要用于对接收到的光信号进行检测,以便通过与其电连接的读取电路等生成指纹数据。
在本申请的实施例中,在采集物体11的指纹数据时,物体11所按压的指纹采集区域对应的发光单元可以全部发光或者部分发光。其中,全部发光是指指纹采集区域内的所有发光单元均发射光束,从而形成完整的光斑,部分发光是指指纹采集区域内的一部分发光单元发射光束,而另一部分发光单元不发射光束,从而无法形成完整的光斑,即称为异形光斑。
例如,图2为本申请实施例中指纹采集区域对应的发光单元全部发光或者部分发光的示意图,如图2所示,指纹采集区域可以是在显示屏12的12a位置,此时,指纹采集区域对应的发光单元全部发光(白色表示发光,发光的区域称为亮区),从而形成完整的光斑;指纹采集区域也可以是在显示屏12的12b位置,此时,指纹采集区域对应的发光单元部分发光(黑色表示不发光,不发光的区域称为暗区),从而形成异形光斑;另外,如12c、12d、12e、12f所示,不发光的发光单元所在的位置以及所形成的形状可以发生变化,本申请对此不做具体限定。
图3为本申请实施例中真手指按压屏幕时的示意图,图4为本申请实施例中2.5D假指纹按压屏幕时的示意图,如图3以及图4所示,在指纹采集区域对应的发光单元部分发光形成异形光斑的情况下,图中亮区表示发光单元发光的区域,暗区表示发光单元不发光的区域,另外,在图4中,由于常见的假指纹材料本身通常无法触发电子设备的指纹识别功能,因此,需要将具有触发采图功能的其他材料或物体,如真手指放置在假指纹材料上方以触发指纹识别功能,从而导致图4中的真手指与假指纹材料共同形成较厚的厚度。
参考图3以及图4,在异形光斑中,位于暗区边缘位置的感光单位可以接收到周围的小角度反射光(实线表示的光线),位于暗区中心位置的感光单元可以接收到真手指或者假指纹材料的透射光(虚线表示的光线)。
而由于假指纹材料的存在,使得图4中位于显示屏上方的物体(真手指+假指纹材料)厚度大于图3中位于显示屏上方的物体(真手指)厚度,而厚度越大,暗区中心位置的感光单元接收到的透射光越多,因此,2.5D假指纹按压屏幕时所产生的透射光比真手指按压屏幕时产生的透射光更多,从而, 本申请基于该差异,可以实现对指纹进行真伪识别。
下面以具体地实施例对本申请的技术方案以及本申请的技术方案如何解决上述技术问题进行详细说明。下面这几个具体的实施例可以相互结合,对于相同或相似的概念或过程可能在某些实施例中不再赘述。下面将结合附图,对本申请的实施例进行描述。
可以理解,本申请中指纹识别方法的处理步骤可以由电子设备内部的处理器实现,该处理器可以基于本申请中的指纹识别方法对指纹的真伪进行识别,从而根据指纹真伪识别结果进行进一步处理,例如,处理器在确定指纹为真指纹时,可以执行电子设备的解锁操作等。
图5为本申请实施例提供的指纹识别方法的示意图,以方法的执行主体为电子设备内部的处理器为例进行解释说明,如图5所示,该方法主要包括以下步骤:
S1、在检测到手指在按压区域内的按压操作时,分别获取待检测指纹对应的第一指纹数据以及第二指纹数据,第一指纹数据为按压区域对应的发光单元部分发光时采集的数据,第二指纹数据为按压区域对应的发光单元全部发光时采集的数据。
处理器获取的第一指纹数据以及第二指纹数据可以是由指纹识别装置中的光学检测部件根据多路光学结构引导的多路光信号得到,并且,第一指纹数据以及第二指纹数据为物体单次按压过程中采集得到的两份指纹数据。
具体的,在检测到物体按压显示屏时,指纹采集区域(即按压区域)对应的发光单元可以在控制信号的作用下,依次进行部分发光以及全部发光,从而依次得到第一指纹数据以及第二指纹数据。可选的,发光单元也可以是先全部发光,再部分发光,从而依次得到第二指纹数据以及第一指纹数据。
其中,在指纹采集区域对应的发光单元部分发光的情况下采集得到的第一指纹数据,由于部分发光所形成的光斑为异形光斑,因此第一指纹数据也可以被称为异形光斑数据。
另外,在指纹采集区域对应的发光单元全部发光的情况下采集得到的第二指纹数据,由于全部发光所形成的光斑为完整光斑,因此第二指纹数据也可以被称为正常光斑数据。
S2、根据第一指纹数据对应的亮区数据和暗区数据,以及第二指纹数据,确定指纹特征信息,亮区数据为处于发光状态的发光单元所在区域内的数据,暗区数据为处于熄灭状态的发光单元所在区域内的数据;
本步骤中,处理器根据第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据确定指纹特征信息,因此可以认为,指纹特征信息与第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据相关,由于真手指与2.5D假指纹的差异主要体现在产生的透射光的量,而该差异可以通过第一指纹数据对应的亮区数据以及暗区数据体现,从而,处理器得到的指纹特征信息可以用于确定待检测指纹的真伪。
S3、根据指纹特征信息,确定待检测指纹的真伪。
由于真手指与2.5D假指纹的差异主要体现在产生的透射光的量,而该差异可以通过第一指纹数据对应的亮区数据以及暗区数据体现,因此,处理器在得到与第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据相关的指纹特征信息后,根据该指纹特征信息即可确定待检测指纹的真伪。
本实施例提供一种指纹识别方法,在单次按压过程中分别得到异形光斑对应的第一指纹数据以及正常光斑对应的第二指纹数据,进而得到与第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据相关的指纹特征信息,由于第一指纹数据对应的亮区数据和暗区数据可以用于表征真手指与2.5D假指纹的差异,即真手指与2.5D假指纹所产生的透射光的量的差异,从而,基于指纹特征信息可以有效确定待检测指纹的真伪,提高对于2.5D假指纹的指纹识别结果的准确性。
在一些实施例中,根据第一指纹数据对应的亮区数据和暗区数据,以及第二指纹数据,确定指纹特征信息,包括:
S21、根据第一指纹数据对应的亮区数据和暗区数据,以及第二指纹数据,确定校准数据;
S22、根据校准数据,确定指纹特征信息。
具体的,在进行屏下指纹识别时,显示屏的发光单元会向置于显示屏上的物体发射光信号。此外,显示屏的发光单元发射的光信号中有一部分光信号(也称为屏幕漏光)直接向下经过多次折射、反射和衍射等被光学检测部 件接收。光学检测部件根据接收到的光信号生成指纹原始数据。然而,该指纹原始数据不仅包括指纹信息,还包括显示屏背景噪声(例如屏幕漏光)等干扰信息(也称为底噪),各个光学检测部件的底噪不同。因此,通过对指纹原始数据进行校准处理,具体例如进行正则化处理,可以消除由于光学检测部件的底噪差异对指纹原始数据的影响,以保证指纹数据的准确性。
在一些实施例中,校准数据包括第一指纹数据对应的第一校准数据和第二指纹数据对应的第二校准数据;
本实施例中,处理器可以分别对第一指纹数据以及第二指纹数据进行正则化处理,从而,可以消除由于光学检测部件的底噪差异对第一指纹数据以及第二指纹数据的影响,以保证第一指纹数据以及第二指纹数据的准确性。
对应的,根据第一指纹数据对应的亮区数据和暗区数据,以及第二指纹数据,确定校准数据,包括:
S211、通过第一类预设数据集合对第一指纹数据对应的亮区数据和暗区数据进行正则化处理,得到第一校准数据,其中,第一类预设数据集合为在发光单元部分发光时获取的、并用于标定在按压过程中发光单元部分发光时采集得到的指纹数据的数据集合;
具体的,第一类预设数据集合包括第一预设数据集合和第二预设数据集合,第一预设数据集合和第二预设数据集合为在发光单元部分发光时获取、以用于标定在按压过程中发光单元部分发光时采集得到的指纹数据的数据集合;
图6为本申请实施例提供的获取第一预设数据集合的示意图,如图6所示,第一预设数据集合可以是在光学检测部件的校准阶段利用肉色平头指纹模型51来模拟用户的手指而获取的数据集合。肉色平头指纹模型51用于模拟无指纹的用户手指,即肉色平头指纹模型51相当于全是指纹谷的手指。在获取第一预设数据集合时,可以将肉色平头指纹模型51按压在指纹采集区域上(即光学检测部件对应的显示屏12的局部区域),在指纹采集区域对应的发光单元部分发光的情况下,光学检测部件根据接收到的光信号确定第一预设数据集合。由于肉色平头指纹模型51相当于全是指纹谷的手指,因此,第一预设数据集合不仅包含有与肉色平头指纹模型51的中央凹面反射的光有关 的信息,还包括显示屏12的背景噪声(例如显示屏12的漏光信息)等干扰信息。
图7为本申请实施例提供的获取第二预设数据集合的示意图,如图7所示,第二预设数据集合可以是在光学检测部件的校准阶段利用黑色平头指纹模型52模拟用户的手指而获取的数据集合。黑色平头指纹模型52用于模拟没有手指触摸的按压状态。在获取第二指纹数据集合时,可以将黑色平头指纹模型52按压在指纹采集区域,在指纹采集区域对应的发光单元处于同样的部分发光的情况下,光学检测部件根据接收到的光信号确定第二预设数据集合。由于黑色平头指纹模型52会吸收向显示屏上方透射的光,因此,第二指纹数据仅包括显示屏的背景噪声(例如显示屏12的漏光信息)等干扰信息,即仅包括光学检测部件的底噪。
将第一预设数据集合减去第二预设数据集合,可以得到不包含底噪的指纹数据。本实施例中利用第一预设数据集合和第二预设数据集合对待检测指纹对应的第一指纹数据进行正则化处理,由此消除不同光学检测部件的底噪对第一指纹数据的影响,提高指纹识别的准确度。
为了便于描述,将第一预设数据集合和第二预设数据集合分别表示为H_Fleshspot和H_blackspot,将待检测指纹的第一指纹数据表示为Spotdata,则可以通过以下公式对第一指纹数据进行正则化处理:
Spotdata=(Spotdata-H_blackspot)/(H_Fleshspot-H_blackspot)
通过上述公式,可以计算得到第一指纹数据对应的第一校准数据。需要说明的是,此处的计算方式仅是为了说明正则化处理的具体原理,在实际计算时也可以利用矩阵运算,提高处理速度。
可以理解,上述正则化处理的实现方式仅是一种示例,本实施例对正则化处理的具体实现方式不做限定,也可以利用第一预设数据集合和第二预设数据集合通过其他合适的正则化处理方式对第一指纹数据进行正则化处理。
S212、通过第二类预设数据集合对第二指纹数据进行正则化处理,得到第二校准数据,其中,第一类预设数据集合为在发光单元全部发光时获取的、并用于标定在按压过程中发光单元全部发光时采集得到的指纹数据的数据集合。
具体的,第二类预设数据集合包括第三预设数据集合和第四预设数据集合,第三预设数据集合和第四预设数据集合为在发光单元全部发光时获取、 以用于标定在按压过程中发光单元全部发光时采集得到的指纹数据的数据集合。
本实施例中,获取第三预设数据集合的过程,基本原理可以参考前文中获取第一预设数据集合的过程,不同之处在于,在获取第三预设数据集合时,可以将肉色平头指纹模型51按压在指纹采集区域上(即光学检测部件对应的显示屏12的局部区域),在指纹采集区域对应的发光单元全部发光的情况下,光学检测部件根据接收到的光信号确定第三预设数据集合。第三预设数据集合不仅包含有与肉色平头指纹模型51的中央凹面反射的光有关的信息,还包括显示屏12的背景噪声(例如显示屏12的漏光信息)等干扰信息。
另外,获取第四预设数据集合的过程,基本原理可以参考前文中获取第二预设数据集合的过程,不同之处在于,在获取第四指纹数据集合时,可以将黑色平头指纹模型52按压在指纹采集区域,在指纹采集区域对应的发光单元处于全部发光的情况下,光学检测部件根据接收到的光信号确定第四预设数据集合。第四指纹数据仅包括显示屏的背景噪声(例如显示屏12的漏光信息)等干扰信息,即仅包括光学检测部件的底噪。
将第三预设数据集合减去第四预设数据集合,可以得到不包含底噪的指纹数据。本实施例中利用第三预设数据集合和第四预设数据集合对待检测指纹对应的第二指纹数据进行正则化处理,由此消除不同光学检测部件的底噪对第二指纹数据的影响,提高指纹识别的准确度。
为了便于描述,将第三预设数据集合和第四预设数据集合分别表示为H_Flesh和H_black,将待检测指纹的第二指纹数据表示为Nordata,则可以通过以下公式对第二指纹数据进行正则化处理:
Nordata=(Nordata-H_black)/(H_Flesh-H_black)
通过上述公式,可以计算得到第二指纹数据对应的第二校准数据。需要说明的是,此处的计算方式仅是为了说明正则化处理的具体原理,在实际计算时也可以利用矩阵运算,提高处理速度。
可以理解,上述正则化处理的实现方式仅是一种示例,本实施例对正则化处理的具体实现方式不做限定,也可以利用第三预设数据集合和第四预设数据集合通过其他合适的正则化处理方式对第二指纹数据进行正则化处理。
从而,通过分别对第一指纹数据以及第二指纹数据进行正则化处理,得 到第一校准数据以及第二校准数据,基于真手指与2.5D假指纹的性能差异,若为2.5D假指纹,则第一校准数据中的暗区数据通常会大于第二校准数据中对应位置的指纹数据;若为真手指,则第一校准数据中的暗区数据通常会小于第二校准数据中对应位置的指纹数据,从而,基于上述原理,可以实现对待检测指纹的真伪识别。
在一些实施例中,在假指纹材料与真手指所产生的透射光的量的差异较大时,例如假指纹材料比较厚,可以直接根据指纹数据本身的数据特征来进行指纹的真伪识别。
在这种情况下,指纹特征信息包括指纹数据的重复度以及指纹数据之间的大小关系,其中,重复度用于表征第一校准数据与第二校准数据的相似度,大小关系为第一校准数据与第二校准数据的数据大小相关关系。
对应的,根据校准数据,确定指纹特征信息,包括:
S221a、从第一校准数据中选择第一子数据,第一子数据为包含亮区数据以及暗区数据的行数据,和/或,第一子数据为包含亮区数据以及暗区数据的列数据;
S221b、从第二校准数据中选择第二子数据,第二子数据在第二指纹数据中的位置与第一子数据在第一指纹数据中的位置相同;
S221c、确定第一子数据中的亮区数据与第二子数据中对应位置的指纹数据的重复度;
S221d、确定第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量,所述数据量用于表征所述第一子数据与所述第二子数据之间的大小关系。
具体的,图8为本申请实施例从第一校准数据中选择第一子数据的示意图,如图8所示,第一校准数据包含亮区数据(图中白色圆环部分)以及暗区数据(图中黑色圆圈部分),在选择第一子数据时,由于需要依赖暗区数据来进行真伪识别,因此,选择的第一子数据为同时包含亮区数据以及暗区数据的行数据和/或列数据。例如,可以选择图8中的第m行数据和/或第n列数据作为第一子数据。
图9为本申请实施例从第二校准数据中选择第二子数据的示意图,如图 9所示,第二校准数据仅包含亮区数据(图中白色圆圈部分)在选择第二子数据时,第二子数据在第二指纹数据中的位置与第一子数据在第一指纹数据中的位置对应相同。例如,若选择图8中第一校准数据的第m行数据作为第一子数据,则同样选择图9中第二校准数据的第m行数据作为第二子数据。
在选择第一子数据以及第二子数据之后,处理器基于第一子数据以及第二子数据确定指纹特征信息,本实施例中,指纹特征信息具体包括第一子数据中的亮区数据,与第二子数据中对应位置的指纹数据的重复度,以及,第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量,从而,基于上述指纹特征信息可以确定待检测指纹的真伪。
在一些实施例中,根据指纹特征信息,确定待检测指纹的真伪,包括:
S311、根据第一子数据中的亮区数据,与第二子数据中对应位置的指纹数据的重复度,确定第一子数据以及第二子数据是否均有效;
S312、在确定第一子数据以及第二子数据均有效时,若第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量达到第一预设阈值,则确定待检测指纹为真指纹。
其中,由于第一指纹数据以及第二指纹数据是在单次按压过程中采集得到的两份指纹数据,该单次按压过程中,物体的按压区域未发生较大变化,因此,第一子数据中的亮区数据与第二子数据中对应位置的指纹数据的相似程度较高,从而,第一子数据中的亮区数据与第二子数据中对应位置的指纹数据的重复度可以用于确定第一指纹数据以及第二指纹数据的有效性,即第一子数据以及第二子数据的有效性。
在确定第一子数据以及第二子数据均为有效数据时,若第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量达到第一预设阈值,则确定待检测指纹为真指纹,否则,确定待检测指纹为假指纹。
在一些实施例中,根据第一子数据中的亮区数据,与第二子数据中对应位置的指纹数据的重复度,确定第一子数据以及第二子数据是否均有效,包括:若第一子数据中亮区数据等于第二子数据中对应位置的指纹数据的数据量达到第二预设阈值,则确定第一子数据与第二子数据均有效。
具体的,若第一子数据中亮区数据等于第二子数据中对应位置的指纹数据的数据量达到第二预设阈值,说明第一子数据中的亮区数据与第二子数据中对应位置的指纹数据的相似程度较高,因此,可以确定第一指纹数据以及第二指纹数据均为有效数据,由此可得第一子数据与第二子数据也均为有效数据,从而,基于第一子数据与第二子数据得到的指纹真伪识别结果的准确性以及可信度较高。
可以理解,本实施例中,第一子数据中亮区数据等于第二子数据中对应位置的指纹数据,可以是两种数据完全相等,或者,两种数据的差值小于某一阈值内,从而,上述两种情况均可以认为第一子数据中亮区数据与第二子数据中对应位置的指纹数据的相似程度较高。
另外,若第一子数据中亮区数据等于第二子数据中对应位置的指纹数据的数据量未达到第二预设阈值,则说明第一子数据中的亮区数据与第二子数据中对应位置的指纹数据的相似程度较低,确定第一子数据以及第二子数据均为无效数据,此时,需要重新采集第一指纹数据以及第二指纹数据,电子设备可以输出重新按压以重新获取第一指纹数据以及第二指纹数据的提示信息。
图10为本申请实施例中进行指纹真伪识别的示例图,如图10所示,坐标轴的横轴表示数据的位置,纵坐标表示数据大小,第一曲线表示第一校准数据中同时包含亮区数据以及暗区数据的某一行的第一子数据,第二曲线表示第二校准数据中对应位置的第二子数据,参考图10所示的曲线,两种曲线中,位于横坐标两端的亮区数据的重复度较高,因此,两种曲线均为有效数据。另外,对于位于横坐标中间的暗区数据,由于第一曲线在第二曲线下方的部分较多,说明第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量达到第一预设阈值,因此可以确定待检测指纹为真指纹。
图11为本申请实施例中进行指纹真伪识别的另一示例图,如图11所示,坐标轴的横轴表示数据的位置,纵坐标表示数据大小,第一曲线表示第一校准数据中同时包含亮区数据以及暗区数据的某一列的第一子数据,第二曲线表示第二校准数据中对应位置的第二子数据,参考图11所示的曲线,两种曲线中,位于横坐标两端的亮区数据的重复度较高,因此,两种曲线均为有效数据。另外,对于位于横坐标中间的暗区数据,由于第一曲线在第二曲线下 方的部分较多,说明第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量达到第一预设阈值,因此可以确定待检测指纹为真指纹。
图12为本申请实施例中进行指纹真伪识别的另一示例图,如图12所示,坐标轴的横轴表示数据的位置,纵坐标表示数据大小,第一曲线表示第一校准数据中同时包含亮区数据以及暗区数据的某一行的第一子数据,第二曲线表示第二校准数据中对应位置的第二子数据,参考图12所示的曲线,两种曲线中,位于横坐标两端的亮区数据的重复度较高,因此,两种曲线均为有效数据。另外,对于位于横坐标中间的暗区数据,由于第一曲线在第二曲线上方的部分较多,说明第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量未达到第一预设阈值,因此可以确定待检测指纹为假指纹。
图13为本申请实施例中进行指纹真伪识别的另一示例图,如图13所示,坐标轴的横轴表示数据的位置,纵坐标表示数据大小,第一曲线表示第一校准数据中同时包含亮区数据以及暗区数据的某一列的第一子数据,第二曲线表示第二校准数据中对应位置的第二子数据,参考图13所示的曲线,两种曲线中,位于横坐标两端的亮区数据的重复度较高,因此,两种曲线均为有效数据。另外,对于位于横坐标中间的暗区数据,由于第一曲线在第二曲线上方的部分较多,说明第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量未达到第一预设阈值,因此可以确定待检测指纹为假指纹。
从而,基于图10至图13所示的示例图,通过本申请的技术方案,可以有效对指纹的真伪进行识别,从而提高电子设备的安全性能。
在一些实施例中,在假指纹材料与真手指所产生的透射光的量的差异较小时,例如假指纹材料比较薄、比较透明(比如明胶),直接根据指纹数据本身的数据特征无法准确进行指纹的真伪识别,此时,可以结合根据指纹数据得到的更深入的数据特征来识别指纹的真伪。
在这种情况下,指纹特征信息包括指纹数据的大小关系、指纹数据均值、指纹数据均值的差值、第一子数据中暗区数据与第二子数据中对应位置的指纹数据之间的差异度。
对应的,根据校准数据,确定指纹特征信息,包括:
S222a、从第一校准数据中选择第一子数据,第一子数据为包括亮区数据 和暗区数据的行数据,和/或,第一子数据为亮区数据和暗区数据的列数据;
S222b、从第二校准数据中选择第二子数据,第二子数据在第二指纹数据中的位置与第一子数据在第一指纹数据中的位置相同;
S222c、确定第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量;
S222d、确定第一子数据中暗区数据的第一数据均值、第二子数据中对应位置的数据的第二数据均值、以及第一数据均值与第二数据均值的差值;
S222e、确定第一子数据中的每个暗区数据与第二子数据中对应位置的指纹数据的差值,并确定每个差值的绝对值的总和,所述总和用于表征所述第一子数据与所述第二子数据的差异度。
具体的,假设第一子数据和第二子数据均为行数据,将第一子数据和第二子数据分别表示为RowSpotData和RowNorData,则分别计算RowSpotData中暗区数据(例如从第m个数据至第n个数据)的第一数据均值AvgRowSpotData,RowNorData中对应位置的数据(即同样是从第m个数据至第n个数据)的第二数据均值AvgRowNorData,以及AvgRowNorData与AvgRowSpotData的差值AvgRowSub。同理,若第一子数据和第二子数据均为列数据,则执行相同的处理。
另外,对于RowSpotData中的每个暗区数据,以及RowNorData中对应位置的指纹数据,逐像素求取其差值,并确定每个差值的绝对值的总和sumRow,该总和sumRow可以用于表示第一子数据中暗区数据与第二子数据中对应位置的指纹数据之间的差异度。同理,若第一子数据和第二子数据均为列数据,则执行相同的处理。
在一些实施例中,根据指纹特征信息,确定待检测指纹的真伪,包括:
S313、将第一子数据中暗区数据小于第二子数据中对应位置的指纹数据的数据量、第一数据均值、第二数据均值、差值、以及总和中的至少一项输入至预先训练的决策树模型,得到用于指示待检测指纹为真指纹的得分;
S314、根据得分和预设指纹阈值的比较结果,确定待检测指纹的真伪。
其中,决策树模型是根据指纹样本集合中每个指纹样本的指纹特征信息以及每个指纹样本的真伪结果训练得到的。每个指纹样本的指纹特征信息与 待检测指纹的指纹特征信息具有相同的特征类型。例如,指纹特征信息的特征类型包括第一子数据的第一数据均值、第二子数据的第二数据均值、第一数据均值与第二数据均值的差值、每个差值(对于RowSpotData中的每个暗区数据以及RowNorData中对应位置的指纹数据,逐像素求取的差值)的绝对值的总和,或其任意组合。应理解,此处的指纹特征信息的特征类型进行举例说明,本实施例不限于此。
指纹样本集合中的指纹样本例如可以包括在各种场景下的真手指和假手指,例如、在低温场景、高温场景、常温场景、油污状态下的真手指和假手指、干手指和/或湿手指。相应地,指纹样本的指纹特征信息为根据各种场景下的指纹样本对应的指纹数据确定的指纹特征信息。
在训练决策树模型时,可以将指纹特征信息的每个特征类型看作一个决策节点,使用每个决策节点对指纹样本进行分类,由此训练生成的决策树模型,该决策树模型可以包括与每个特征类型对应的判断阈值和权重。
在进行指纹识别时,将根据待检测指纹对应的指纹数据生成的指纹特征信息输入到预先训练的决策树模型,预先训练的决策树模型根据预先确定的判断阈值和权重,生成待检测指纹的得分。
例如,若指纹特征信息的特征类型包括第一子数据的第一数据均值、第二子数据的第二数据均值、第一数据均值与第二数据均值的差值和每个差值的绝对值的总和,在训练决策树模型时,可以提取指纹样本集合中的每个指纹样本对应的第一子数据的第一数据均值、第二子数据的第二数据均值、第一数据均值与第二数据均值的差值和每个差值的绝对值的总和作为指纹样本对应的指纹特征信息,将所提取的每个指纹样本的指纹特征信息和每个指纹样本对应的真伪结构输入决策树模型进行训练,得到训练的预测模型,该训练的预测模型包括第一子数据的第一数据均值对应的第一判断阈值和第一权重、第二子数据的第二数据均值对应的第二判断阈值和第二权重、第一数据均值与第二数据均值的差值对应的第三判断阈值和第三权重、以及每个差值的绝对值的总和对应的第四判断阈值和第四权重。在对待检测指纹进行识别时,可以提取待检测指纹的第一子数据的第一数据均值、第二子数据的第二数据均值、第一数据均值与第二数据均值的差值和每个差值的绝对值的总和,训练的决策树模型根据待检测指纹的第一子数据的第一数据均值与第一判断 阈值的比较结果、待检测指纹的第一子数据的第一数据均值与第二判断阈值的比较结果、待检测指纹的第一数据均值与第二数据均值的差值与第三判断阈值的比较结果、待检测指纹的每个差值的绝对值的总和与第四判断阈值的比较结果,以及各个判断条件对应的权重确定用于指示待检测指纹为真指纹的得分。
另外,预设指纹阈值可以根据用户的安全级别要求进行灵活设置,例如在安全级别要求较低的应用场景中,例如在通过指纹验证对电子设备进行解锁的应用场景中,可以将预设指纹阈值设置得相对较低,例如0.5。而在安全级别要求较高的应用场景中,例如在通过指纹验证进行费用支付的应用场景中,可以将预设指纹阈值设置得相对较高,例如0.7。
本实施例中,基于上述指纹特征信息,采用预先训练好的决策树模型进行打分,并根据得分和预设指纹阈值的比较结果,确定待检测指纹的真伪,从而可以准确识别真假指纹。
在一些实施例中,校准数据包括第一指纹数据对应的第三校准数据;
根据第一指纹数据对应的亮区数据和暗区数据,以及第二指纹数据,确定校准数据,包括:
S213、从第一指纹数据对应的亮区数据中选择第三子数据,确定第三子数据的第三数据均值;
S214、从第二指纹数据中选择第四子数据,确定第四子数据的第四数据均值,第四子数据在第二指纹数据中的位置与第三子数据在第一指纹数据中的位置对应相同;
S215、确定第三数据均值与第四数据均值的比值;
S216、基于第一指纹数据的亮区数据和暗区数据、第二指纹数据以及比值,对第一指纹数据进行正则化处理,得到第三校准数据。
图14为本申请实施例中第一指纹数据的示意图,其中,实线圆圈内的数据为第一指纹数据对应的暗区数据,实线圆圈外、方框内的数据为第一指纹数据对应的亮区数据;在选择第三子数据时,可以是选择亮区数据中的一份或者多份子数据,例如,可以将图14中的P11、P21、P31、P41作为第三子数据。
图15为本申请实施例中第二指纹数据的示意图,其中,虚线圆圈内的数据为第一指纹数据中暗区数据在第二指纹数据中对应位置的指纹数据,虚线圆圈外、方框内的数据为第一指纹数据中亮区数据在第二指纹数据中对应位置的指纹数据。例如,在将图14中的P11、P21、P31、P41作为第三子数据时,可以将图15中的P12、P22、P32、P42作为第四子数据。
在选择第三子数据以及第四子数据后,分别计算对应的数据均值,例如,P11、P21、P31、P41对应的数据均值分别以AvgSpotPart1、AvgSpotPart2、AvgSpotPart3、AvgSpotPart4表示,P12、P22、P32、P42对应的数据均值分别以AvgNorPart1、AvgNorPart2、AvgNorPart3、AvgNorPart4表示,则第三数据均值与第四数据均值的比值包括K1、K2、K3、K4,其中:
K1=AvgNorPart1/AvgSpotPart1
K2=AvgNorPart2/AvgSpotPart2
K3=AvgNorPart3/AvgSpotPart3
K4=AvgNorPart4/AvgSpotPart4
然后,计算比值的均值K:
K=(K1+K2+K3+K4)/4
最后,基于第一指纹数据SpotData,第二指纹数据NorData,通过以下公式对第一指纹数据进行正则化处理,得到第三校准数据SpotNomalized:
SpotNomalized=SpotData/(NorData*K)
从而,通过对第一指纹数据进行正则化处理,可以提高指纹数据的准确性,从而提高指纹识别结果的准确性。
在一些实施例中,根据指纹特征信息,确定待检测指纹的真伪,包括:
S315、确定第三校准数据对应的亮区数据和/或暗区数据的量化特征信息,量化特征信息包括数据均值、数据方差、数据标准差以及梯度值中的至少一项;
S316、将量化特征信息输入至预先训练的决策树模型,得到用于指示待检测指纹为真指纹的得分;
S317、根据得分和预设指纹阈值的比较结果,确定待检测指纹的真伪。
具体的,在确定第三校准数据对应的亮区数据和/或暗区数据的量化特征信息时,可以从第三校准数据的暗区数据中选择第五子数据以及第六子数据, 确定第五子数据的第五数据均值、数据方差、数据标准差,以及第六子数据的第六数据均值、数据方差、数据标准差,第五子数据与第六子数据位置不同;
例如,参考图14,在得到第三校准数据后,可以选择第三校准数据中与图14中C1位置相同的数据作为第五子数据,即第三校准数据中暗区数据的中心位置数据。另外,可以选择第三校准数据中与图14中L1、R1、U1、D1位置相同的数据作为第六子数据,即第三校准数据中暗区数据的边缘位置数据。
在选择第五子数据以及第六子数据后,分别计算对应的数据均值、数据方差、数据标准差。例如,第五子数据对应的第五数据均值avgcenter,方差varCenter等,以及第六子数据对应的第六数据均值avgL/avgR/avgU/avgD、方差varL/varR/varU/varD等。
此外,在确定第三校准数据对应的亮区数据和/或暗区数据的量化特征信息时,也可以从第三校准数据的亮区数据中选择第七子数据,确定第七子数据的第七数据均值、数据方差、数据标准差;
例如,参考图14,在得到第三校准数据后,可以选择第三校准数据中与图14中P11、P21、P31、P41位置相同的数据作为第七子数据。
在选择第七子数据后,分别计算对应的数据均值、数据方差、数据标准差。例如,第七子数据对应的第七数据均值avgPart1/avgPart2/avgPart3/avgPart4等。
另外,在得到第五数据均值、第六数据均值以及第七数据均值后,还可以基于第五数据均值、第六数据均值以及第七数据均值,确定第三校准数据中不同区域之间的梯度值,即四个方向从外围区域到中心区域的两段梯度。
例如,以P11的数据均值减去L1的数据均值,即可得到左侧的第一段梯度:Gra1=avgPart1-avgL;以L1的数据均值减去C1的数据均值,即可得到左侧的第二段梯度:GraL2=avgL-avgCenter。同理,可以得到其他方向的两段梯度。
可以理解,本实施例中,并不限定第六子数据、第七子数据的具体形状、具体位置以及具体数量。例如,图16和图17为本申请实施例中选择第六子数据、第七子数据的另两种示意图,第六子数据、第七子数据的具 体位置可以进行调整,同样,具体形状以及具体数量也可以进行调整。
本实施例中,基于上述指纹特征信息,采用预先训练好的决策树模型进行打分,并根据得分和预设指纹阈值的比较结果,确定待检测指纹的真伪,从而可以准确识别真假指纹。
在一些实施例中,处理器也可以是通过分别对第一指纹数据以及第二指纹数据进行正则化处理,并获取第一校准数据以及第二校准数据的指纹数据的大小关系、指纹数据均值、指纹数据均值的差值、第一子数据中暗区数据与第二子数据中对应位置的指纹数据之间的差异度,以及,通过对第一指纹数据进行正则化处理,并获取第三校准数据对应的亮区数据和/或暗区数据的量化特征信息,然后,将上述指纹特征信息输入至预先训练的决策树模型,得到用于指示所述待检测指纹为真指纹的得分;根据所述得分和预设指纹阈值的比较结果,确定所述待检测指纹的真伪。
其中,上述处理步骤的具体过程可以参考前文的描述内容,在此不再赘述。本实施例通过结合多种指纹特征信息来进行打分判断,可以进一步提高指纹真伪识别结果的准确性。
应该理解的是,虽然上述实施例中的流程图中的各个步骤按照箭头的指示依次显示,但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明,这些步骤的执行并没有严格的顺序限制,其可以以其他的顺序执行。而且,图中的至少一部分步骤可以包括多个子步骤或者多个阶段,这些子步骤或者阶段并不必然是在同一时刻执行完成,而是可以在不同的时刻执行,其执行顺序也不必然是依次进行,而是可以与其他步骤或者其他步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。
在一些实施例中,提供一种指纹识别装置。图18为本申请实施例提供的指纹识别装置的示意图,如图18所示,该装置包括:
指纹数据获取模块100,用于在检测到手指在按压区域内的按压操作时,分别获取待检测指纹对应的第一指纹数据以及第二指纹数据,第一指纹数据为按压区域对应的发光单元部分发光时采集的数据,第二指纹数据为按压区 域对应的发光单元全部发光时采集的数据;
特征信息确定模块200,用于根据第一指纹数据对应的亮区数据和暗区数据,以及第二指纹数据,确定指纹特征信息,亮区数据为处于发光状态的发光单元所在区域内的数据,暗区数据为处于熄灭状态的发光单元所在区域内的数据;
指纹真伪确定模块300,用于根据指纹特征信息,确定待检测指纹的真伪。
关于指纹识别装置的具体限定可以参见上文中对于指纹识别方法的限定,在此不再赘述。上述指纹识别装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中,也可以以软件形式存储于计算机设备中的存储器中,以便于处理器调用执行以上各个模块对应的操作。
本申请提供一种指纹识别装置,在单次按压过程中分别得到异形光斑对应的第一指纹数据以及正常光斑对应的第二指纹数据,进而得到与第一指纹数据对应的亮区数据和暗区数据以及第二指纹数据相关的指纹特征信息,由于第一指纹数据对应的亮区数据和暗区数据可以用于表征真手指与2.5D假指纹的差异,即真手指与2.5D假指纹所产生的透射光的量的差异,从而,基于指纹特征信息可以有效确定待检测指纹的真伪,提高对于2.5D假指纹的指纹识别结果的准确性。
在一些实施例中,提供一种电子设备,包括:存储器,处理器及存储在存储器上并可在处理器上运行的计算机程序,处理器执行程序时实现上述的指纹识别方法。
在上述终端设备中,存储器和处理器之间直接或间接地电性连接,以实现数据的传输或交互。例如,这些元件相互之间可以通过一条或者多条通信总线或信号线实现电性连接,如可以通过总线连接。存储器中存储有实现数据访问控制方法的计算机执行指令,包括至少一个可以软件或固件的形式存储于存储器中的软件功能模块,处理器通过运行存储在存储器内的软件程序以及模块,从而执行各种功能应用以及数据处理。
存储器可以是,但不限于,随机存取存储器(Random Access Memory, RAM),只读存储器(Read Only Memory,ROM),可编程只读存储器(Programmable Read-Only Memory,PROM),可擦除只读存储器(Erasable Programmable Read-Only Memory,EPROM),电可擦除只读存储器(Electric Erasable Programmable Read-Only Memory,EEPROM)等。其中,存储器用于存储程序,处理器在接收到执行指令后,执行程序。进一步地,上述存储器内的软件程序以及模块还可包括操作系统,其可包括各种用于管理系统任务(例如内存管理、存储设备控制、电源管理等)的软件组件和/或驱动,并可与各种硬件或软件组件相互通信,从而提供其他软件组件的运行环境。
处理器可以是一种集成电路芯片,具有信号的处理能力。上述的处理器可以是通用处理器,包括中央处理器(Central Processing Unit,CPU)、网络处理器(Network Processor,NP)等。可以实现或者执行本申请实施例中的公开的各方法、步骤及逻辑框图。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
在一些实施例中,提供一种计算机可读存储介质,计算机可读存储介质中存储有计算机执行指令,计算机执行指令被处理器执行时用于实现本申请各方法实施例的步骤。
在一些实施例中,提供一种计算机程序产品,包括计算机程序,该计算机程序被处理器执行时实现本申请各方法实施例的步骤。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程,是可以通过计算机程序来指令相关的硬件来完成,计算机程序可存储于一非易失性计算机可读取存储介质中,该计算机程序在执行时,可包括如上述各方法的实施例的流程。其中,本申请所提供的各实施例中所使用的对存储器、存储、数据库或其它介质的任何引用,均可包括非易失性和/或易失性存储器。非易失性存储器可包括只读存储器(ROM)、可编程ROM(PROM)、电可编程ROM(EPROM)、电可擦除可编程ROM(EEPROM)或闪存。易失性存储器可包括随机存取存储器(RAM)或者外部高速缓冲存储器。作为说明而非局限,RAM以多种形式可得,诸如静态RAM(SRAM)、动态RAM(DRAM)、同步DRAM(SDRAM)、双数据率SDRAM(DDRSDRAM)、增强型SDRAM(ESDRAM)、同步链路(Synchlink)DRAM(SLDRAM)、 存储器总线(Rambus)直接RAM(RDRAM)、直接存储器总线动态RAM(DRDRAM)、以及存储器总线动态RAM(RDRAM)等。
本领域技术人员在考虑说明书及实践这里公开的申请后,将容易想到本公开的其它实施方案。本申请旨在涵盖本公开的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本公开的一般性原理并包括本公开未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本公开的真正范围和精神由下面的权利要求书指出。
应当理解的是,本公开并不局限于上面已经描述并在附图中示出的精确结构,并且可以在不脱离其范围进行各种修改和改变。本公开的范围仅由所附的权利要求书来限制。

Claims (15)

  1. 一种指纹识别方法,其特征在于,包括:
    在检测到手指在按压区域内的按压操作时,分别获取待检测指纹对应的第一指纹数据以及第二指纹数据,所述第一指纹数据为所述按压区域对应的发光单元部分发光时采集的数据,所述第二指纹数据为所述按压区域对应的发光单元全部发光时采集的数据;
    根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定指纹特征信息,所述亮区数据为处于发光状态的发光单元所在区域内的数据,所述暗区数据为处于熄灭状态的发光单元所在区域内的数据;
    根据所述指纹特征信息,确定所述待检测指纹的真伪。
  2. 根据权利要求1所述的方法,其特征在于,所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定指纹特征信息,包括:
    根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据;
    根据所述校准数据,确定指纹特征信息。
  3. 根据权利要求2所述的方法,其特征在于,所述校准数据包括第一指纹数据对应的第一校准数据和所述第二指纹数据对应的第二校准数据;
    所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据,包括:
    通过第一类预设数据集合对所述第一指纹数据对应的亮区数据和暗区数据进行正则化处理,得到所述第一校准数据,其中,所述第一类预设数据集合为在所述发光单元部分发光时获取的、并用于标定在按压过程中发光单元部分发光时采集得到的指纹数据的数据集合;
    通过第二类预设数据集合对所述第二指纹数据进行正则化处理,得到所述第二校准数据,其中,所述第一类预设数据集合为在所述发光单元全部发光时获取的、并用于标定在按压过程中发光单元全部发光时采集得到的指纹数据的数据集合。
  4. 根据权利要求3所述的方法,其特征在于,所述指纹特征信息包括指纹数据的重复度以及指纹数据之间的大小关系;
    所述根据所述校准数据,确定指纹特征信息,包括:
    从所述第一校准数据中选择第一子数据,所述第一子数据为包含亮区数据以及暗区数据的行数据,和/或,所述第一子数据为包含亮区数据以及暗区数据的列数据;
    从所述第二校准数据中选择第二子数据,所述第二子数据在所述第二指纹数据中的位置与所述第一子数据在所述第一指纹数据中的位置相同;
    确定所述第一子数据中的亮区数据与所述第二子数据中对应位置的指纹数据的重复度;
    确定所述第一子数据中暗区数据小于所述第二子数据中对应位置的指纹数据的数据量,所述数据量用于表征所述第一子数据与所述第二子数据之间的大小关系。
  5. 根据权利要求4所述的方法,其特征在于,所述根据所述指纹特征信息,确定所述待检测指纹的真伪,包括:
    根据所述第一子数据中的亮区数据,与所述第二子数据中对应位置的指纹数据的重复度,确定所述第一子数据以及所述第二子数据是否均有效;
    在确定所述第一子数据以及所述第二子数据均有效时,若所述第一子数据中暗区数据小于所述第二子数据中对应位置的指纹数据的数据量达到第一预设阈值,则确定所述待检测指纹为真指纹。
  6. 根据权利要求5所述的方法,其特征在于,根据所述第一子数据中的亮区数据,与所述第二子数据中对应位置的指纹数据的重复度,确定所述第一子数据以及所述第二子数据是否均有效,包括:
    若所述第一子数据中亮区数据等于所述第二子数据中对应位置的指纹数据的数据量达到第二预设阈值,则确定所述第一子数据与所述第二子数据均有效。
  7. 根据权利要求3所述的方法,其特征在于,所述指纹特征信息包括指纹数据的大小关系、指纹数据均值、指纹数据均值的差值、第一子数据中暗区数据与第二子数据中对应位置的指纹数据之间的差异度;
    所述根据所述校准数据,确定指纹特征信息,包括:
    从所述第一校准数据中选择第一子数据,所述第一子数据为包括所述亮区数据和所述暗区数据的行数据,和/或,所述第一子数据为所述亮区数据和所述暗区数据的列数据;
    从所述第二校准数据中选择第二子数据,所述第二子数据在所述第二指纹数据中的位置与所述第一子数据在所述第一指纹数据中的位置相同;
    确定所述第一子数据中暗区数据小于所述第二子数据中对应位置的指纹数据的数据量;
    确定所述第一子数据中暗区数据的第一数据均值、所述第二子数据中对应位置的数据的第二数据均值、以及所述第一数据均值与所述第二数据均值的差值;
    确定所述第一子数据中的每个暗区数据与所述第二子数据中对应位置的指纹数据的差值,并确定每个差值的绝对值的总和,所述总和用于表征所述第一子数据与所述第二子数据的差异度。
  8. 根据权利要求2所述的方法,其特征在于,所述校准数据包括第一指纹数据对应的第三校准数据;
    所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据,包括:
    从所述第一指纹数据对应的亮区数据中选择第三子数据,确定所述第三子数据的第三数据均值;
    从所述第二指纹数据中选择第四子数据,确定所述第四子数据的第四数据均值,所述第四子数据在所述第二指纹数据中的位置与所述第三子数据在所述第一指纹数据中的位置对应相同;
    确定所述第三数据均值与所述第四数据均值的比值;
    基于所述第一指纹数据的亮区数据和暗区数据、所述第二指纹数据以及所述比值,对所述第一指纹数据进行正则化处理,得到所述第三校准数据。
  9. 根据权利要求8所述的方法,其特征在于,所述指纹特征信息包括量化特征信息;
    所述根据所述校准数据,确定指纹特征信息,包括:
    确定所述第三校准数据对应的亮区数据和/或暗区数据的量化特征信息,所述量化特征信息包括数据均值、数据方差、数据标准差以及梯度值中的至少一项。
  10. 根据权利要求7所述的方法,其特征在于,所述校准数据还包括第一指纹数据对应的第三校准数据;
    所述根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定校准数据,还包括:
    从所述第一指纹数据对应的亮区数据中选择第三子数据,确定所述第三子数据的第三数据均值;
    从所述第二指纹数据中选择第四子数据,确定所述第四子数据的第四数据均值,所述第四子数据在所述第二指纹数据中的位置与所述第三子数据在所述第一指纹数据中的位置对应相同;
    确定所述第三数据均值与所述第四数据均值的比值;
    基于所述第一指纹数据的亮区数据和暗区数据、所述第二指纹数据以及所述比值,对所述第一指纹数据进行正则化处理,得到所述第三校准数据。
  11. 根据权利要求10所述的方法,其特征在于,所述指纹特征信息还包括量化特征信息;
    所述根据所述校准数据,确定指纹特征信息,还包括:
    确定所述第三校准数据对应的亮区数据和/或暗区数据的量化特征信息,所述量化特征信息包括数据均值、数据方差、数据标准差以及梯度值中的至少一项。
  12. 根据权利要求7或9或11任一项所述的方法,其特征在于,所述根据所述指纹特征信息,确定所述待检测指纹的真伪,包括:
    将所述指纹特征信息输入至预先训练的决策树模型,得到用于指示所述待检测指纹为真指纹的得分;
    根据所述得分和预设指纹阈值的比较结果,确定所述待检测指纹的真伪。
  13. 一种指纹识别装置,其特征在于,包括:
    指纹数据获取模块,用于在检测到手指在按压区域内的按压操作时,分别获取待检测指纹对应的第一指纹数据以及第二指纹数据,所述第一指纹数据为所述按压区域对应的发光单元部分发光时采集的数据,所述第二指纹数据为所述按压区域对应的发光单元全部发光时采集的数据;
    特征信息确定模块,用于根据所述第一指纹数据对应的亮区数据和暗区数据,以及所述第二指纹数据,确定指纹特征信息,所述亮区数据为处于发光状态的发光单元所在区域内的数据,所述暗区数据为处于熄灭状态的发光单元所在区域内的数据;
    指纹真伪确定模块,用于根据所述指纹特征信息,确定所述待检测指纹的真伪。
  14. 一种电子设备,其特征在于,包括:存储器,处理器及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述处理器执行所述程序时实现上述如权利要求1-12任一项所述的指纹识别方法。
  15. 一种计算机可读存储介质,其特征在于,所述计算机可读存储介质中存储有计算机执行指令,所述计算机执行指令被处理器执行时用于实现上述如权利要求1-12任一项所述的指纹识别方法。
PCT/CN2021/095328 2021-05-21 2021-05-21 指纹识别方法、装置、电子设备及存储介质 WO2022241792A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/095328 WO2022241792A1 (zh) 2021-05-21 2021-05-21 指纹识别方法、装置、电子设备及存储介质

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2021/095328 WO2022241792A1 (zh) 2021-05-21 2021-05-21 指纹识别方法、装置、电子设备及存储介质

Publications (1)

Publication Number Publication Date
WO2022241792A1 true WO2022241792A1 (zh) 2022-11-24

Family

ID=84140133

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/095328 WO2022241792A1 (zh) 2021-05-21 2021-05-21 指纹识别方法、装置、电子设备及存储介质

Country Status (1)

Country Link
WO (1) WO2022241792A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017183867A1 (ko) * 2016-04-18 2017-10-26 주식회사 유니온커뮤니티 생체의 심장 박동에 따른 지문 이미지의 미세한 밝기 변화에 기초하여 위조지문을 판별할 수 있는 위조지문 판별 장치 및 그 방법
CN110582780A (zh) * 2019-08-01 2019-12-17 深圳市汇顶科技股份有限公司 指纹识别和防伪的方法、装置和电子设备
CN111274906A (zh) * 2020-01-16 2020-06-12 上海思立微电子科技有限公司 指纹防伪方法及装置
CN211529170U (zh) * 2020-01-06 2020-09-18 深圳市汇顶科技股份有限公司 指纹识别装置和电子设备

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2017183867A1 (ko) * 2016-04-18 2017-10-26 주식회사 유니온커뮤니티 생체의 심장 박동에 따른 지문 이미지의 미세한 밝기 변화에 기초하여 위조지문을 판별할 수 있는 위조지문 판별 장치 및 그 방법
CN110582780A (zh) * 2019-08-01 2019-12-17 深圳市汇顶科技股份有限公司 指纹识别和防伪的方法、装置和电子设备
CN211529170U (zh) * 2020-01-06 2020-09-18 深圳市汇顶科技股份有限公司 指纹识别装置和电子设备
CN111274906A (zh) * 2020-01-16 2020-06-12 上海思立微电子科技有限公司 指纹防伪方法及装置

Similar Documents

Publication Publication Date Title
EP3690701A1 (en) Fingerprint recognition method, fingerprint recognition device and terminal equipment
US10599933B2 (en) Biometric image capturing apparatus and biometric image capturing method
US4896363A (en) Apparatus and method for matching image characteristics such as fingerprint minutiae
US11495050B2 (en) Method for distinguishing a real three-dimensional object from a two-dimensional spoof of the real object
US20150309663A1 (en) Flexible air and surface multi-touch detection in mobile platform
Greco et al. Benchmarking deep network architectures for ethnicity recognition using a new large face dataset
CN113177516A (zh) 指纹识别方法、装置、电子设备及存储介质
EP3659064B1 (en) Optical fingerprint sensor with scattered light image detection
WO2011126411A1 (ru) Способ обработки банкнот (варианты)
US10380408B2 (en) Method of detecting fraud
CN112115917B (zh) 指纹识别方法、指纹识别装置、电子设备及存储介质
US20210012026A1 (en) Tokenization system for customer data in audio or video
WO2022241792A1 (zh) 指纹识别方法、装置、电子设备及存储介质
EP3387573A1 (en) Methods and apparatuses for birefringence based biometric authentication
Jaiswal et al. Fake region identification in an image using deep learning segmentation model
WO2022067543A1 (zh) 指纹识别方法、指纹识别装置、电子设备及存储介质
CN114067401A (zh) 目标检测模型的训练及身份验证方法和装置
Yu et al. Comparison of paired ROC curves through a two-stage test
CN109564624A (zh) 辨识指纹Logo的方法、装置和电子设备
TWI766569B (zh) 一種影像感測方法
CN117150369B (zh) 超重预测模型的训练方法及电子设备
TW201137762A (en) System and method for determining whether an individual to be identified as a registered individual
EP4250173A1 (en) Authentication system, device and method
US10984220B2 (en) Method and a device for verifying liveness of a finger
Bhargava et al. Face Detection and Recognition System

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: 21940245

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21940245

Country of ref document: EP

Kind code of ref document: A1