Biometric Identification Method and Device, Electronic Device, Storage Medium

Abstract

The present application provides a biometric recognition method, device, electronic device, and storage medium. The method includes: obtaining biometric feature samples of multiple modalities of a person to be recognized; for each modality, comparing the biometric feature sample corresponding to the modality with the biometric registration template of the target object in the modality to obtain a comparison score corresponding to the modality; if the comparison scores of at least one modality are greater than the similarity threshold corresponding to the modality, for each modality, converting the comparison score corresponding to the modality into a false acceptance rate; according to the relative sample quality and false acceptance rate corresponding to each modality, fusing to obtain a comprehensive result of the false recognition rate; and determining whether the person to be recognized is consistent with the target object according to the comprehensive result of the false recognition rate and the confidence threshold. This solution integrates the reliability weight of the sample quality and the feature comparison credibility of each modality, improving the passing rate and reliability of biometric recognition.

Description

Technical Field

[0001] This application relates to the field of information technology, and particularly to a biometric recognition method and device, an electronic device, and a computer-readable storage medium. Background Art

[0002] Biometric recognition technology closely combines computers with high-tech means such as optics, acoustics, biosensors, and biostatistics principles, and uses inherent physiological characteristics of the human body such as fingerprints, faces, irises, finger veins, etc. for personal identity recognition.

[0003] With the increasing requirements for social security, the accuracy and reliability of identity authentication, the limitations of single biometric recognition in terms of accuracy and reliability have become increasingly prominent and cannot meet the needs of product and technology development. At the same time, with the continuous maturity of biometric recognition technology, the research and application of multimodal biometric recognition technology will make up for the limitations and deficiencies of single biometric recognition technology, further reduce the false acceptance rate of biometric recognition systems, and improve the discrimination accuracy. Therefore, multimodal biometric recognition technology has attracted more and more attention and has become one of the most potential and advantageous research directions in the field of biometric recognition. Summary of the Invention

[0004] Embodiments of this application provide a biometric recognition method to solve the problems of low security, low reliability, and low passing rate in the process of biometric recognition.

[0005] Embodiments of this application provide a biometric recognition method, including:

[0006] Obtain biometric feature samples of multiple modalities of the person to be recognized;

[0007] For each modality, compare the biometric feature sample corresponding to the modality with the biometric registration template of the target object in the modality to obtain the comparison score corresponding to the modality;

[0008] If the comparison score of at least one modality is greater than the similarity threshold corresponding to the modality, for each modality, convert the comparison score corresponding to the modality into a false acceptance rate;

[0009] Fuse the relative sample quality and false acceptance rate corresponding to each modality to obtain a comprehensive result of the false recognition rate;

[0010] Determine whether the person to be recognized is the same as the target object according to the comprehensive result of the false recognition rate and the confidence threshold.

[0011] In one embodiment, after obtaining the biometric feature samples of multiple modalities of the person to be recognized, the method further includes:

[0012] For each modality, determine whether the sample quality score of the biometric sample of the modality is greater than the sample quality threshold. If it is less, re-obtain the biometric sample of the modality.

[0013] In one embodiment, before converting the comparison score corresponding to each modality into a false acceptance rate, the method further includes:

[0014] For each test sample set corresponding to each modality, calculate the false acceptance rate corresponding to different similarity thresholds under the modality, and construct a mapping relationship between the similarity threshold and the false acceptance rate under the modality.

[0015] In one embodiment, converting the comparison score corresponding to each modality into a false acceptance rate includes:

[0016] For each modality, according to the comparison score corresponding to the modality and the mapping relationship between the similarity threshold and the false acceptance rate under the modality, obtain the false acceptance rate corresponding to the comparison score under the modality.

[0017] In one embodiment, before fusing according to the relative sample quality and false acceptance rate corresponding to each modality to obtain a comprehensive result of the false recognition rate, the method further includes:

[0018] For each modality, according to the sample quality score of the biometric sample of the modality and the maximum quality score corresponding to the modality, obtain the relative sample quality of the biometric sample of the modality.

[0019] In one embodiment, fusing according to the relative sample quality and false acceptance rate corresponding to each modality to obtain a comprehensive result of the false recognition rate includes:

[0020] For each modality, convert the false acceptance rate corresponding to the modality into a corresponding credibility;

[0021] Perform weighted summation on the credibility corresponding to each modality and the relative sample quality to obtain a fused credibility;

[0022] Convert the fused credibility into the comprehensive result of the false recognition rate.

[0023] In one embodiment, before determining whether the person to be recognized is consistent with the target object according to the comprehensive result of the false recognition rate and the credibility threshold, the method further includes:

[0024] Fuse according to the similarity threshold corresponding to each modality to obtain the credibility threshold.

[0025] An embodiment of the present application further provides a biometric recognition device, including:

[0026] A feature acquisition module, configured to acquire biometric feature samples of multiple modalities of the person to be identified;

[0027] A feature comparison module, configured to, for each modality, compare the similarity between the biometric feature sample corresponding to the modality and the biometric feature registration template of the target object in the modality, and obtain a comparison score corresponding to the modality;

[0028] A score conversion module, configured to, if the comparison scores of at least one modality are greater than the similarity threshold corresponding to the modality, convert the comparison scores corresponding to each modality into false acceptance rates for each modality;

[0029] A data fusion module, configured to fuse to obtain a comprehensive result of the false recognition rate according to the relative sample quality and the false acceptance rate corresponding to each modality;

[0030] A result comparison module, configured to determine whether the person to be identified is consistent with the target object according to the comprehensive result of the false recognition rate and the confidence threshold.

[0031] An embodiment of the present application further provides an electronic device, where the electronic device includes:

[0032] A processor;

[0033] A memory for storing instructions executable by the processor;

[0034] Wherein, the processor is configured to execute the biometric identification method described in any one of the above.

[0035] An embodiment of the present application further provides a computer-readable storage medium, where the storage medium stores a computer program, and the computer program can be executed by the processor to complete the biometric identification method described in any one of the above.

[0036] The technical solution provided by the above embodiment of the present application, by acquiring biometric feature samples of multiple modalities of the person to be identified, and respectively comparing the similarity between the biometric feature sample corresponding to each modality and the biometric feature registration template of the target object in this modality; when the comparison score in one modality is greater than the similarity threshold corresponding to this modality, convert the comparison score corresponding to each modality into a false acceptance rate; fuse the false acceptance rate and the relative sample quality corresponding to each modality to obtain a comprehensive result of the false recognition rate, and compare the comprehensive result of the false recognition rate with the confidence threshold to determine whether the person to be identified is consistent with the target object, thereby effectively, safely and reliably identifying biometric features. Description of the Drawings

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required to be used in the embodiments of the present application will be briefly introduced below.

[0038] Figure 1 Schematic structural diagram of the electronic device provided by an embodiment of the present application;

[0039] Figure 2 Schematic flow chart of a biometric recognition method provided by an embodiment of the present application;

[0040] Figure 3 Similarity threshold of a modality provided by an embodiment of the present application And false acceptance rate FAR X Mapping relationship therebetween;

[0041] Figure 4 Schematic flow chart of a method for obtaining a comprehensive result of false recognition rate provided by an embodiment of the present application;

[0042] Figure 5 Block diagram of a biometric recognition device provided by an embodiment of the present application. Detailed implementation manners

[0043] Next, the technical solutions in the embodiments of the present application will be described with reference to the accompanying drawings in the embodiments of the present application.

[0044] Similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first", "second", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0045] The biometric recognition method provided by the embodiment of the present application can be applied to the following scenarios: obtaining three feature samples of a person's fingerprint, face, and iris and corresponding feature templates through a camera and a sensor of an image acquisition device; obtaining the relative sample quality of the fingerprint, face, and iris according to the feature samples, and obtaining a comparison score through similarity comparison between the feature samples and the corresponding feature templates; when the comparison score of the fingerprint, face, or iris is greater than the similarity threshold corresponding to its modality, converting the comparison score into a false acceptance rate; obtaining a corresponding comprehensive result of false recognition rate based on the false acceptance rates corresponding to the fingerprint, face, and iris and the relative sample quality; and judging whether the feature recognition of this person is successful according to the comparison between the comprehensive result of false recognition rate and the confidence threshold.

[0046] Figure 1 An electronic device 1 shown in an embodiment of the present application, the electronic device 1 includes: at least one processor 11 and a memory 12, Figure 1Take a processor as an example. A processor 11 and a memory 12 are connected through a bus 10. The memory 12 stores instructions executable by the processor 11, and the instructions are executed by the processor 11. Among them, the processor 11 is configured to execute the biometric identification method provided in the embodiments of the present application.

[0047] The processor 11 may be a device including a central processing unit (CPU), a graphics processing unit (GPU), or other forms of processing units having data processing capabilities and / or instruction execution capabilities. It can process data of other components in the electronic device 1 and can also control other components in the electronic device 1 to execute desired functions.

[0048] The memory 12 may include one or more computer program products. The computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may run the program instructions to implement the biometric identification method described below. Various application programs and various data may also be stored in the computer-readable storage medium, such as various data used and / or generated by the application programs, etc.

[0049] Figure 1 The components and structure of the illustrated electronic device 1 are merely exemplary and not restrictive. According to needs, in one embodiment, the electronic device 1 may also have other components and structures.

[0050] In one embodiment, the exemplary electronic device 1 for implementing the biometric identification method of the embodiments of the present application may be implemented as a smart device such as a smart phone, a tablet computer, a desktop computer, a laptop computer, a vehicle-mounted terminal, etc.

[0051] Figure 2 It is a schematic flowchart of a biometric identification method provided by the embodiments of the present application. As Figure 2 shown, this method can be executed by the Figure 1 illustrated electronic device 1 to implement biometric identification. This method includes the following steps S210 - S250.

[0052] Step S210: Obtain biometric feature samples of multiple modalities of the person to be identified.

[0053] In this step, biometric samples of the person to be identified are obtained through the camera, sensors, etc. of the image acquisition device. Among them, the obtained biometric samples are multi-modal biometric samples. For example, biometric samples of two or more modalities such as a person's face, fingerprint, iris, and finger vein are obtained.

[0054] For each modality of biometric samples, judge its sample quality score QS X whether it is greater than the sample quality threshold If it is less than, re-obtain the biometric samples of this modality. Among them, the sample quality score QS X is not less than zero and not greater than the maximum quality score MS X , the maximum quality score MS X is usually 100; the sample quality threshold can be set according to specific circumstances. For example, the sample quality threshold can be set to 60.

[0055] The sample quality score of biometric samples can be obtained by scoring the samples. For example, score the obtained face samples, and comprehensively evaluate the quality of the face samples according to the brightness, clarity, face size, angle, position, and contrast of the obtained face images. Evaluate the brightness, clarity, face size, angle, position, and face image respectively to obtain their respective evaluation coefficients Q i (0-100). Then perform weighted calculation through the weights a i occupied by each coefficient. Finally, obtain the sample quality score QS X of each face sample. The calculation formula is as follows:

[0056]

[0057] Among them, i represents brightness, clarity, face size, angle, position, and face image, 0≤Q i ≤100,

[0058] Step S220: For each modality, compare the biometric samples corresponding to the modality with the biometric registration template of the target object in the modality to obtain the comparison score corresponding to the modality.

[0059] In this step, the biometric samples P X of each modality and the biometric registration template R X in this modality are respectively compared for similarity to obtain the comparison score CS X . Among them, the biometric registration template can be the biometrics such as the face, fingerprint, iris, and finger vein of a person with a known identity.

[0060] Step S230: If the matching scores of at least one modality are greater than the similarity threshold corresponding to the modality, for each modality, convert the matching score corresponding to the modality into a false acceptance rate.

[0061] In this step, when there is a matching score CS of one modality X greater than the similarity threshold corresponding to its modality then for each modality, convert the matching score CS corresponding to each modality X into a false acceptance rate FAR X . Among them, the similarity threshold can be set manually. The similarity threshold can be set to 60 or 70; the false acceptance rate FAR X represents the proportion of impostors being wrongly judged as accepted in the verification and identification of biometric samples.

[0062] Before converting the matching score CS corresponding to each modality X into a false acceptance rate FAR X , for the test sample set corresponding to each modality, calculate the false acceptance rate FAR corresponding to different similarity thresholds in this modality X , and construct the mapping relationship between the similarity threshold and the false acceptance rate FAR X in this modality.

[0063] As Figure 3 shown, it is the mapping relationship between the similarity threshold in one modality and the false acceptance rate FAR and false rejection rate FRR in this modality. Among them, the false rejection rate FRR represents the proportion of genuine users being wrongly judged as rejected. The false acceptance rate FAR and false rejection rate FRR are obtained from the test of the sample set. The false acceptance rate FAR is jointly evaluated by the false match rate FMR of misidentifying samples of different people as the same person and the failed sample quality evaluation rate FTEQR; the false rejection rate FRR is jointly evaluated by the false non-match rate FNMR of misidentifying samples of the same person as different people and the failed sample quality evaluation rate FTEQR.

[0064] The calculation formula of the false acceptance rate FAR is as follows:

[0065] FAR = FMR × (1 - FTEQR)

[0066] Among them, the calculation formulas of the false match rate FMR of misidentifying samples of different people as the same person and the failed sample quality evaluation rate FTEQR are as follows:

[0067]

[0068] In the above formula, C1 is the total number of comparisons of samples from different people, and C2 is the number of times samples from different people are recognized as the same person.

[0069]

[0070] In the above formula, FTEQR is the ratio of biological samples that do not meet the quality requirements, M1 is the total number of biometric samples participating in the evaluation, and M2 is the number of samples that do not meet the quality requirements.

[0071] The calculation formula for the false rejection rate FRR is as follows:

[0072] FRR = FTEQR + FNMR × (1 - FTEQR)

[0073] Among them, the calculation formula for the ratio FNMR of misidentifying samples of the same person as different people is as follows:

[0074]

[0075] In the above formula, C3 is the total number of comparisons of samples of the same person, and C4 is the number of times samples of the same person are recognized as different people.

[0076] According to the comparison score CS corresponding to each modality X and the similarity threshold under each modality and the mapping relationship with the false acceptance rate FAR X between them, the false acceptance rate FAR X corresponding to the comparison score CS X .

[0077] Step S240: According to the relative sample quality and false acceptance rate corresponding to each modality, fuse to obtain the comprehensive result of the misidentification rate.

[0078] In this step, the relative sample quality RQ X corresponding to each modality and the false acceptance rate FAR X are fused respectively to obtain the comprehensive result FAR z of the misidentification rate. Among them, the relative sample quality RQ X is obtained by the ratio of the sample quality score QS X and the maximum quality score MS X ; the calculation formula for the comprehensive result FAR z of the misidentification rate is as follows:

[0079]

[0080] Among them, RQ n is the relative sample quality corresponding to the nth modality, and FAR n is the false acceptance rate corresponding to the nth modality.

[0081] Step S250: Determine whether the person to be identified is the same as the target object according to the comprehensive result of the misrecognition rate and the confidence threshold.

[0082] In this step, the comprehensive result of the misrecognition rate FAR z is compared with the confidence threshold T FAR to determine whether the comprehensive result of the misrecognition rate FAR z is less than or equal to the confidence threshold T FAR . If the comprehensive result of the misrecognition rate FAR z is less than or equal to the confidence threshold T FAR , then the person to be identified is the same as the target object, that is, the recognition result is passed; if the comprehensive result of the misrecognition rate FAR z is greater than the confidence threshold T FAR , then the person to be identified is not the same as the target object, that is, the recognition result is not passed. Among them, the confidence threshold T FAR can be set according to actual needs, and the confidence threshold T FAR can be set to 10 -4 or 10 -5 .

[0083] In another embodiment, the confidence threshold T can be obtained by fusing according to the similarity threshold corresponding to each modality FAR . The calculation formula for obtaining the confidence threshold T by fusing the similarity threshold FAR is as follows:

[0084]

[0085]

[0086] where is the similarity threshold corresponding to the nth modality.

[0087] In one embodiment, as Figure 4 shown, the above step S240 specifically includes steps S410 - S430.

[0088] Step S410: For each modality, obtain the relative sample quality of the biometric sample of the modality according to the sample quality score of the biometric sample of the modality and the maximum quality score corresponding to the modality.

[0089] In this step, for each modality, divide the sample quality score QS X of its biometric sample by the maximum quality score MS X to obtain the relative sample quality RQ X of its biometric sample. That is, RQX The calculation formula is as follows:

[0090]

[0091] Where 0 ≤ RQ X ≤ 1.

[0092] Step S420: For each modality, convert the false acceptance rate corresponding to the modality into the corresponding credibility.

[0093] In this step, convert the false acceptance rate FAR X corresponding to each modality into the corresponding credibility That is, the smaller the false acceptance rate FAR X corresponding to the modality, the greater the credibility corresponding to the modality.

[0094] Step S430: Perform a weighted sum of the credibility corresponding to each modality and the relative sample quality to obtain the fused credibility.

[0095] In this step, respectively perform a weighted sum of the credibility corresponding to each modality and the relative sample quality RQ X to obtain the fused credibility That is, the calculation formula for the fused credibility is as follows:

[0096]

[0097] Where RQ n is the relative sample quality corresponding to the nth modality, and FAR n is the false acceptance rate corresponding to the nth modality.

[0098] Step S440: Convert the fused credibility into the comprehensive result of the false recognition rate.

[0099] In this step, obtain the comprehensive result FAR of the false recognition rate through the fused credibility z The calculation formula is as follows:

[0100]

[0101] Where RQ n is the relative sample quality corresponding to the nth modality, and FAR n is the false acceptance rate corresponding to the nth modality.

[0102] The above biometric recognition method obtains biometric feature samples of multiple modalities of the person to be recognized, and respectively compares the biometric feature samples corresponding to each modality with the biometric registration template of the target object in that modality; when the comparison score in one modality is greater than the similarity threshold corresponding to this modality, the comparison score corresponding to each modality is converted into a false acceptance rate; the false acceptance rate corresponding to each modality and the relative sample quality are fused to obtain a comprehensive misrecognition rate result, and the comprehensive misrecognition rate result is compared with the confidence threshold to determine whether the person to be recognized is consistent with the target object, thereby effectively, safely and reliably recognizing biometric features.

[0103] The following is an embodiment of the device of the present application, which can be used to execute the above-mentioned embodiment of the biometric recognition method of the present application. For the details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the biometric recognition method of the present application.

[0104] Figure 5 It is a block diagram of a biometric recognition device provided by an embodiment of the present application. As Figure 5 shown, the device includes: a feature acquisition module 510, a feature comparison module 520, a score conversion module 530, a data fusion module 540, and a result comparison module 550.

[0105] The feature acquisition module 510 is used to acquire biometric feature samples of multiple modalities of the person to be recognized;

[0106] The feature comparison module 520 is used to, for each modality, compare the biometric feature sample corresponding to the modality with the biometric registration template of the target object in the modality to obtain the comparison score corresponding to the modality;

[0107] The score conversion module 530 is used to, if the comparison score of at least one modality is greater than the similarity threshold corresponding to the modality, for each modality, convert the comparison score corresponding to the modality into a false acceptance rate;

[0108] The data fusion module 540 is used to fuse according to the relative sample quality and the false acceptance rate corresponding to each modality to obtain a comprehensive misrecognition rate result;

[0109] The result comparison module 550 is used to determine whether the person to be recognized is consistent with the target object according to the comprehensive misrecognition rate result and the confidence threshold.

[0110] The implementation processes of the functions and roles of each module in the above device are specifically detailed in the implementation processes of the corresponding steps in the above biometric recognition method, and will not be elaborated here.

[0111] In several embodiments provided in this application, the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and a module, a program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0112] In addition, in each embodiment of this application, the various functional modules can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0113] If the function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

Claims

1. A method for biometric recognition, characterized in that, Including: Obtain biometric feature samples of multiple modalities of the person to be identified; For each modality, compare the biometric feature sample corresponding to the modality with the biometric registration template of the target object in the modality to obtain the comparison score corresponding to the modality; If the comparison scores of at least one modality are greater than the similarity thresholds corresponding to the modalities, for each modality, convert the comparison score corresponding to the modality into a false acceptance rate; Based on the relative sample quality and false acceptance rate corresponding to each modality, fuse to obtain a comprehensive misidentification rate result; Based on the comprehensive misidentification rate result and the confidence threshold, determine whether the person to be identified is consistent with the target object; Among them, the fusing to obtain a comprehensive misidentification rate result based on the relative sample quality and false acceptance rate corresponding to each modality includes: For each modality, convert the false acceptance rate corresponding to the modality into a corresponding confidence level; Perform weighted summation on the confidence levels and relative sample qualities corresponding to each modality to obtain a fused confidence level; Convert the fused confidence level into the comprehensive misidentification rate result; The formula for converting the false acceptance rate corresponding to the modality into a corresponding confidence level is as follows: Among them, FAR X is the false acceptance rate.

2. The method according to claim 1, wherein After obtaining the biometric feature samples of multiple modalities of the person to be identified, the method further includes: For each modality, determine whether the sample quality score of the biometric feature sample of the modality is greater than the sample quality threshold. If it is less, re-obtain the biometric feature sample of the modality.

3. The method according to claim 1, characterized in that Before converting the comparison score corresponding to each modality into a false acceptance rate, the method further includes: For the test sample set corresponding to each modality, calculate the false acceptance rates corresponding to different similarity thresholds in the modality, and construct a mapping relationship between the similarity thresholds and the false acceptance rates in the modality.

4. The method according to claim 3, wherein The converting the comparison score corresponding to each modality into a false acceptance rate includes: For each modality, based on the comparison score corresponding to the modality and the mapping relationship between the similarity threshold and the false acceptance rate in the modality, obtain the false acceptance rate corresponding to the comparison score in the modality.

5. The method according to claim 1, wherein Before fusing to obtain a comprehensive misidentification rate result based on the relative sample quality and false acceptance rate corresponding to each modality, the method further includes: For each modality, based on the sample quality score of the biometric feature sample of the modality and the maximum quality score corresponding to the modality, obtain the relative sample quality of the biometric feature sample of the modality.

6. The method according to claim 1, characterized in that, Before determining whether the person to be identified is consistent with the target object based on the comprehensive misidentification rate result and the confidence threshold, the method further includes: Based on the similarity thresholds corresponding to each modality, fuse to obtain the confidence threshold.

7. A biometric recognition device, characterized in that Including: A feature acquisition module for obtaining biometric feature samples of multiple modalities of the person to be identified; A feature comparison module for, for each modality, comparing the biometric feature sample corresponding to the modality with the biometric registration template of the target object in the modality to obtain the comparison score corresponding to the modality; A score conversion module, configured to, if the comparison scores of at least one modality are greater than the similarity threshold corresponding to the modality, convert, for each modality, the comparison score corresponding to the modality into a false acceptance rate; A data fusion module, configured to fuse the relative sample quality and false acceptance rate corresponding to each modality to obtain a comprehensive result of the false recognition rate; A result comparison module, configured to determine whether the person to be identified is the same as the target object according to the comprehensive result of the false recognition rate and the confidence threshold; Wherein, the fusing the relative sample quality and false acceptance rate corresponding to each modality to obtain a comprehensive result of the false recognition rate includes: For each modality, converting the false acceptance rate corresponding to the modality into a corresponding confidence level; Performing a weighted sum of the confidence levels corresponding to each modality and the relative sample quality to obtain a fused confidence level; Converting the fused confidence level into the comprehensive result of the false recognition rate; The formula for converting the false acceptance rate corresponding to the modality into a corresponding confidence level is as follows: Among them, FAR X is the false acceptance rate.

8. An electronic device, characterized in that, The electronic device includes: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the biometric identification method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program can be executed by the processor to complete the biometric identification method according to any one of claims 1-6.

Images