Sampling method, device and equipment for biological characteristic verification data

Abstract

One or more embodiments of the present specification relate to a sampling method and apparatus for face scanning data, a computing device and a computer program product. The method comprises the steps that a first sample set is obtained from a sample population through a first sampling method, the sample population comprises at least two pieces of biological feature verification data and an overall feature probability, the biological feature verification data are labeled with sample feature values, and the sample feature values comprise verification success and verification failure; the overall feature probability is used for representing the proportion of verification success data and verification failure data in the sample overall. The method further comprises the step of determining a first feature probability corresponding to the first sample set, wherein the first feature probability is used for representing the proportion of verification success data and verification failure data in the first sample set. Moreover, the method further includes obtaining a second sample set from the sample population by a second sampling method in response to the first feature probability and the population feature probability satisfying a probability deviation condition, the second sampling method being different from the first sampling method.

Description

Technical Field

[0001] The embodiments of the present specification generally relate to the field of computer technology, and more specifically to a method, apparatus, computing device, and computer program product for sampling biometric verification data. Background Art

[0002] Sampling is a statistical method that refers to selecting some individuals from a large number of items, materials or populations as representatives, and making inferences about the population through observation and analysis of these individuals. The basic principle of sampling is to ensure that the sample can fully represent the population, so as to infer the characteristics of the population through the research results of the sample. There are many sampling methods, including but not limited to simple random sampling, stratified sampling, cluster sampling, systematic sampling, etc.

[0003] Biometric verification data refers to information that can represent individual biological characteristics obtained through scanning, collection and processing during the biometric verification process. These biometric characteristics usually include fingerprints, faces, irises, voiceprints, etc., which are unique and stable.

[0004] Computer simulation testing is a method of testing computer models by simulating real environments or scenarios. It can evaluate the behavior and performance of the system by simulating different scenarios and conditions, helping analysts to better understand the operation of the system. Simulation testing has a wide range of applications in data analysis, software development, engineering design and other fields. Summary of the invention

[0005] In view of this, one or more embodiments of the present specification provide a sampling method, apparatus, computing device and computer program product for biometric verification data, which can obtain a first sample set from a total sample set containing biometric verification data through a first sampling method, and compare the deviation between the feature probabilities of the two, and then determine whether to change the sampling method and re-sample based on the deviation, thereby improving the distribution consistency of the sampled samples and the sample population, and further improving the accuracy of the simulation results based on the sampled samples.

[0006] In a first aspect of the present specification, a sampling method for biometric verification data is provided. The method includes obtaining a first sample set from a sample population by a first sampling method, the sample population includes at least two biometric verification data and a total feature probability, the biometric verification data is annotated with a sample feature value, the sample feature value includes verification success and verification failure, and the total feature probability is used to characterize the proportion of verification success data to verification failure data in the sample population. The method also includes determining a first feature probability corresponding to the first sample set, the first feature probability is used to characterize the proportion of verification success data to verification failure data in the first sample set. In addition, the method also includes obtaining a second sample set from the sample population by a second sampling method in response to the first feature probability and the total feature probability satisfying a probability deviation condition, and the second sampling method is different from the first sampling method.

[0007] In a second aspect of the present specification, a sampling device for biometric verification data is provided. The device includes a first sample set acquisition unit, configured to acquire a first sample set from a sample population by a first sampling method, the sample population includes at least two biometric verification data and a total feature probability, the biometric verification data is annotated with a sample feature value, the sample feature value includes verification success and verification failure, and the total feature probability is used to characterize the proportion of verification success data to verification failure data in the sample population. The device also includes a first feature probability determination unit, configured to determine a first feature probability corresponding to the first sample set, the first feature probability is used to characterize the proportion of verification success data to verification failure data in the first sample set. In addition, the device also includes a second sample set acquisition unit, configured to acquire a second sample set from the sample population by a second sampling method in response to the first feature probability and the total feature probability satisfying a probability deviation condition, the second sampling method is different from the first sampling method.

[0008] In a third aspect of the present specification, a computing device is provided, comprising: at least one processing unit; and at least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the computing device to execute the method as described in the first aspect of the present specification.

[0009] In a fourth aspect of the present specification, a computer program product is provided, comprising machine executable instructions, which, when executed by a device, cause the device to perform the method according to the first aspect of the present specification.

[0010] It should be understood that the contents described in the Summary of the Invention are not intended to limit the key or important features of the embodiments of this specification, nor are they intended to limit the scope of this disclosure. Other features of this specification will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above and other purposes, features and advantages of the various embodiments of the present specification will become more easily understood through the following detailed description with reference to the accompanying drawings. In the accompanying drawings, various embodiments of the present specification will be described in an exemplary and non-limiting manner, wherein:

[0012] Figure 1 A schematic diagram showing an example environment in which some embodiments according to the present specification may be implemented;

[0013] Figure 2 A flowchart showing a method for sampling biometric verification data according to some embodiments of the present specification is shown;

[0014] Figure 3 A schematic diagram of the structure of a data sampler according to some embodiments of the present specification is shown;

[0015] Figure 4 A block diagram showing an apparatus for sampling biometric verification data according to some embodiments of the present specification; and

[0016] Figure 5 A block diagram of an electronic device according to some embodiments of the present specification is shown.

[0017] Throughout the drawings, the same or similar reference numbers denote the same or similar elements. DETAILED DESCRIPTION

[0018] The embodiments of the present specification will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present specification are shown in the accompanying drawings, it should be understood that the present specification can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present specification. It should be understood that the drawings and embodiments of the present specification are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] In the description of the embodiments of this specification, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0020] As mentioned above, in a big data environment, using sampling samples for simulation testing has many benefits, including reducing data processing complexity, saving computing resources, improving simulation efficiency, etc. However, in biometric verification data, since multiple biometrics may be uploaded for one verification, and only one of them is a successful verification, the sample distribution obtained by sampling the biometric verification data is often inconsistent with the sample population, resulting in inaccurate final simulation results.

[0021] To this end, an embodiment of the present specification provides a sampling method for biometric verification data. It can obtain a first sample set from a total sample set containing biometric verification data through a first sampling method, compare the deviation between the feature probabilities of the two, and then determine whether to change the sampling method and re-sample based on the deviation. In this way, the distribution consistency of the sampled samples and the total sample can be improved, and the accuracy of the simulation results based on the sampled samples can be further improved.

[0022] Figure 1 1 shows a schematic diagram of an environment 100 in which various embodiments of the present specification can be implemented. Figure 1 As shown, the environment 100 may include a sample population 102, a data sampler 104, and a target sample set 106. The sample population 102 may include multiple biometric verification data, wherein the biometrics may include a face, fingerprint, iris, voiceprint, etc. that are unique and stable. In the process of biometric verification, verification failures are often encountered. Therefore, a biometric verification process may generate several verification failure data and one verification success data.

[0023] In some embodiments, the data generated by the biometric verification process can be uploaded to a network server for storage. Optionally, the biometric verification data can also be directly stored in a local terminal, or stored in a hybrid storage manner. Optionally, the biometric verification data can be obtained by offline extraction to serve as the sample population 102. For example, 24 hours of face verification data in the server can be extracted offline on a daily basis to serve as the sample population 102. It is to be understood that the embodiments of this specification do not limit the storage and retrieval methods of the sample population 102.

[0024] like Figure 1As shown, the sample population 102 can be sampled by the data sampler 104 to obtain the target sample set 106. The target sample set 106 also includes verification success data and verification failure data, and its sample distribution (i.e., the proportion of verification success data and verification failure data) is basically consistent with the sample population 102. Therefore, the target sample set 106 can be used for simulation testing to improve simulation efficiency. Optionally, the target sample set 106 can be sampled according to a percentage (e.g., 10%) of the sample population 102. The specific structure of the data sampler 104 will be described below. Figure 3 Provide detailed explanation.

[0025] It should be understood that the architecture and functions in the example environment 100 are described for exemplary purposes only and do not imply any limitation on the scope of the present description. The embodiments of the present description may also be applied to other environments with different structures and / or functions.

[0026] The following will combine Figures 2 to 3 The process according to the embodiment of this specification is described in detail. For ease of understanding, the specific data mentioned in the following description are exemplary and are not intended to limit the scope of protection of the present disclosure. It should be understood that the embodiments described below may also include additional actions not shown and / or may omit the actions shown, and the scope of the present disclosure is not limited in this respect.

[0027] Figure 2 2 is a schematic flow chart of a method 200 for sampling biometric authentication data according to some embodiments of the present specification. In some embodiments, the method 200 may be performed by, for example, Figure 1 It should be understood that the method 200 may also include additional actions not shown and / or may omit the actions shown, and the scope of the present disclosure is not limited in this respect.

[0028] like Figure 2 As shown, in box 210, method 200 may include: obtaining a first sample set from a sample population through a first sampling method, the sample population includes at least two biometric verification data and a population characteristic probability, the biometric verification data is marked with a sample characteristic value, the sample characteristic value includes verification success and verification failure, and the population characteristic probability is used to characterize the proportion of verification success data to verification failure data in the sample population.

[0029] In some embodiments, the biometric verification data may include face verification data. The face verification data may be generated by face recognition behavior of the user terminal and stored in a network server. Optionally, the face verification data may be obtained by offline extraction.

[0030] In block 220, method 200 may include: determining a first feature probability corresponding to the first sample set, the first feature probability being used to characterize the proportion of successful verification data to failed verification data in the first sample set. For example, in face verification data, the first feature probability may be reflected as a face scanning success rate. When the face scanning success rate corresponding to the first sample set is substantially consistent with the face scanning success rate of the sample population, the result obtained by simulation using the first sample set will be more accurate.

[0031] In box 230, method 200 may include: in response to the first feature probability and the overall feature probability satisfying the probability deviation condition, obtaining a second sample set from the sample population by a second sampling method, the second sampling method being different from the first sampling method. When it is detected that the first sample set obtained by the first sampling method is significantly inconsistent with the sample distribution of the sample population, a new sample set may be obtained by changing the sampling method, and further sample distribution inspection may be performed. Optionally, the sampling method may include stratified sampling, random sampling, linear sampling, systematic sampling, etc.

[0032] In some embodiments, the data sampler 104 may first determine a deviation test method based on the overall feature probability and the number of samples in the first sample set. The deviation test method includes a small sample exact test (which may be referred to as a first deviation test method) and a large sample approximate test (which may be referred to as a second deviation test method). In response to the overall feature probability and the number of samples in the first sample set satisfying the first condition, the data sampler 104 may determine the deviation between the first feature probability and the overall feature probability by the first deviation test method; in response to the overall feature probability and the number of samples in the first sample set satisfying the second condition, the data sampler 104 may determine the deviation between the first feature probability and the overall feature probability by the second deviation test method. In this way, the accuracy of the deviation test can be improved. The first deviation test method and the second deviation test method will be described in detail below.

[0033] In the first deviation test method, the sample population only has successful verification data and failed verification data, and therefore obeys the binomial distribution. The data sampler 104 can first establish the null hypothesis and the alternative hypothesis, wherein the null hypothesis is that the deviation between the first feature probability and the overall feature probability is 0, and the alternative hypothesis is that the deviation between the first feature probability and the overall feature probability is not 0. Then, the data sampler 104 can determine the probability value under the null hypothesis. In response to the probability value being less than the preset significance level, the data sampler 104 can accept the alternative hypothesis, that is, it is considered that the first sample set and the sample population are not significantly correlated, and it is necessary to change the sampling method and resample; in response to the probability value being greater than the preset significance level, the data sampler 104 can accept the null hypothesis, that is, it is considered that the first sample set and the sample population are significantly correlated.

[0034] In the second deviation test method, the normal approximation can be used to determine the deviation. The data sampler 104 can first establish the null hypothesis and the alternative hypothesis, the null hypothesis is that the deviation between the first characteristic probability and the overall characteristic probability is 0, and the alternative hypothesis is that the deviation between the first characteristic probability and the overall characteristic probability is not 0. Then, the data sampler 104 can determine the observed value, the overall mean and the overall standard deviation. Based on the observed value, the overall mean, the overall standard deviation and the continuity correction coefficient, the data sampler 104 can determine the standard score. Optionally, the value of the continuity correction coefficient can be 0.5. Based on the standard score, the data sampler 104 can determine the probability value under the null hypothesis. In response to the probability value being less than the preset significance level, the data sampler 104 can accept the alternative hypothesis, that is, it is considered that the first sample set and the sample population are not significantly related, and it is necessary to change the sampling method and resample; in response to the probability value being greater than the preset significance level, the data sampler 104 can accept the null hypothesis, that is, it is considered that the first sample set and the sample population are significantly related.

[0035] The confidence interval is an estimated interval of a population parameter constructed by a sample statistic, which reflects the probability that the true value falls around the measurement result. The smaller the width of the interval, the more accurate the estimate of the true value, but the smaller the confidence (credibility), that is, the smaller the grasp. In some embodiments, the data sampler 104 can also determine the confidence interval based on the standard score, the sample size of the sample population, the sample size of the first sample set, the probability value and the continuity correction coefficient. In this way, the estimation of the population parameter by the sample data can be understood more accurately.

[0036] In some embodiments, the data sampler 104 may determine the minimum sample size based on the sample size, standard score, and continuity correction coefficient of the sample population. In this way, the sampling steps can be further simplified, thereby improving the simulation efficiency. Further, in some embodiments, the data sampler 104 may also determine the target sample size based on the minimum sample size and the error level to ensure a balance between sampling efficiency and sample distribution consistency.

[0037] In some embodiments, the data sampler 104 may determine a second feature probability corresponding to the second sample set, and the second feature probability is used to characterize the proportion of the verification success data and the verification failure data in the second sample set. In response to the second feature probability and the sample feature probability not satisfying the probability deviation condition, the data sampler 104 may determine that the second sample set is a simulation data set, and use the second sample set for simulation to obtain accurate and sufficient simulation results.

[0038] In some embodiments, the data sampler 104 may further sample the second sample set and also use the above-mentioned deviation test to obtain a sample set with a smaller sample size and substantially consistent sample distribution, thereby further improving simulation efficiency.

[0039] In this way, when it is detected that the sampled sample is significantly uncorrelated with the sample population, the sampling method can be changed to obtain a new sample set, thereby improving the distribution consistency of the sampled sample and the sample population, and further improving the accuracy of the simulation results based on the sampled sample.

[0040] Figure 3 FIG. 3 shows a schematic diagram of the structure of a data sampler 300 according to some embodiments of the present specification. The data sampler 300 can be regarded as Figure 1 A specific implementation of the data sampler 102 is shown. Figure 3 As shown, the data sampler 300 may firstly use a first sampling method to sample the sample population 302 in step 304 to obtain a first sample set. The sample population 302 may include multiple biometric verification data, including verification success data and verification failure data. For example, the sample population 302 may be face verification data, and includes face scanning success data and face scanning failure data. The first sampling method may include random sampling, systematic sampling, stratified sampling, etc.

[0041] In step 306, the data sampler 300 can determine the feature probability in the first sample set, such as the face recognition success rate in the face verification data, to perform a deviation test with the feature probability of the sample population, thereby determining the consistency of the sample distribution. In step 308, the data sampler 300 can determine whether the number of samples is sufficient, thereby deciding whether to use the first deviation test method 310 (which can be called a small sample accurate test) or the second deviation test method 312 (which can be called a large sample approximate test) to test the deviation. Specifically, the data sampler 300 can select the first deviation test method when the following formula is satisfied, otherwise select the second deviation test method:

[0042] n*p*(1-p)<5 (1)

[0043] Where n represents the number of samples in the first sample set, and p represents the verification success rate of the sample population.

[0044] In the small sample precision test, if the number of successful verification data in the first sample set is k, the null hypothesis can be set as the verification success rate of the first sample set p(k) = p, that is, the first sample set has a significant correlation with the sample population. At the same time, the alternative hypothesis is set as p(k) <> p, that is, the first sample set has no significant correlation with the sample population. Since the sample conforms to the pass / fail binomial distribution, the probability of drawing k successful verification data in the first sample set can be calculated by the following formula:

[0045]

[0046] If P rIf (x=k) is less than the preset significance level (for example, 0.05), the null hypothesis can be rejected, and it is considered that the first sample set has no significant correlation with the sample population; otherwise, it cannot be rejected, and it is considered that there is a significant correlation between them.

[0047] In the large sample approximation test, similarly, if the number of successful verification data in the first sample set is k, the null hypothesis can be set as the verification success rate of the first sample set p(k) = p, that is, the first sample set has a significant correlation with the sample population. At the same time, the alternative hypothesis is set as p(k) <> p, that is, the first sample set has no significant correlation with the sample population. For a binomial distribution random variable X with parameters p and n, the mean u and standard deviation σ of the population can be determined as follows:

[0048] u=np (3)

[0049]

[0050] Assuming that the sample size is large or p is infinitely close to 1 / 2, and X approximately obeys a normal distribution, the following formula for calculating the Z value (also known as the standard score) can be obtained:

[0051]

[0052] Where 1 / 2 is the continuity correction coefficient, a is the minimum value of probability, b is the maximum value of probability, X normal represents a random variable that satisfies the normal distribution, Z normal It represents the Z value that satisfies the normal distribution. After calculating the Z value, the probability value p can be determined by looking up the Z table of the standard normal distribution and comparing it with the preset significance level. If the probability value is less than the preset significance level (for example, 0.05), the null hypothesis can be rejected, and it is believed that the first sample set has no significant correlation with the sample population. Otherwise, it cannot be rejected, and it is believed that they are significantly correlated.

[0053] When using large sample approximate tests to determine deviations, you can further obtain confidence intervals and minimum sample sizes to more accurately understand the sample data's estimate of the population parameter and improve sampling efficiency. Based on the Z value calculated above, the confidence interval Range probability The calculation formula is as follows:

[0054]

[0055] p sample -c≤Range probability ≤p sample +c (7)

[0056] Where c is the standard error adjustment term, N is the sample size of the sample population, and p sampleis the sampling inspection pass rate, is the continuity correction coefficient. Referring to the logic of the above large sample approximate test, the calculation formula for the minimum sample size is:

[0057] n sample =Z 2 p(1-p) (8)

[0058] Taking into account the continuity correction and the degrees of freedom of sample loss, the improved minimum sample size calculation formula is:

[0059]

[0060] Furthermore, considering the error level MoE, the following formula can be substituted into formula (9) to obtain the target sample size n revised :

[0061]

[0062] In step 320, when the data sampler 300 determines that the deviation is too large, it can enter step 324, update the sampling method, re-sample the sample population 302 to obtain a different sample set, and re-test the deviation; if the data sampler 300 determines that the deviation is within an acceptable range, it can directly use the current sample set for simulation in step 322 to obtain accurate and sufficient simulation results.

[0063] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment and the computing device embodiment, since they are basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0064] Figure 4 FIG. 4 is a schematic block diagram of a sampling device 400 for biometric verification data according to some embodiments of the present specification. Figure 4 As shown, the apparatus 400 includes: a first sample set acquisition unit 402 , a first feature probability determination unit 404 , and a second sample set acquisition unit 406 .

[0065] In some embodiments, the first sample set acquisition unit 402 is configured to acquire a first sample set from a sample population through a first sampling method, the sample population including at least two biometric verification data and a total feature probability, the biometric verification data is annotated with a sample feature value, the sample feature value includes verification success and verification failure, and the total feature probability is used to characterize the proportion of verification success data to verification failure data in the sample population; the first feature probability determination unit 404 is configured to determine a first feature probability corresponding to the first sample set, the first feature probability is used to characterize the proportion of verification success data to verification failure data in the first sample set; the second sample set acquisition unit 406 is configured to acquire a second sample set from the sample population through a second sampling method in response to the first feature probability and the total feature probability satisfying a probability deviation condition, the second sampling method being different from the first sampling method.

[0066] In some embodiments, the device 400 also includes a simulation data set determination unit, which is configured to determine a second feature probability corresponding to the second sample set, the second feature probability is used to characterize the proportion of verification success data and verification failure data in the second sample set; and in response to the second feature probability and the sample feature probability not satisfying the probability deviation condition, determine that the second sample set is a simulation data set.

[0067] In some embodiments, the second sample set acquisition unit 406 is also configured to determine a deviation verification method based on the overall feature probability and the number of samples in the first sample set, the deviation verification method including a first deviation verification method and a second deviation verification method; in response to the overall feature probability and the number of samples in the first sample set satisfying a first condition, determine the deviation between the first feature probability and the overall feature probability through the first deviation verification method; and in response to the overall feature probability and the number of samples in the first sample set satisfying a second condition, determine the deviation between the first feature probability and the overall feature probability through the second deviation verification method.

[0068] In some embodiments, the second sample set acquisition unit 406 is further configured to establish a null hypothesis and an alternative hypothesis, the null hypothesis being that the deviation between the first feature probability and the overall feature probability is 0, and the alternative hypothesis being that the deviation between the first feature probability and the overall feature probability is not 0; determine the probability value under the null hypothesis; accept the alternative hypothesis in response to the probability value being less than a preset significance level; and accept the null hypothesis in response to the probability value being greater than a preset significance level.

[0069] In some embodiments, the second sample set acquisition unit 406 is also configured to establish a null hypothesis and an alternative hypothesis, the null hypothesis being that the deviation between the first feature probability and the overall feature probability is 0, and the alternative hypothesis being that the deviation between the first feature probability and the overall feature probability is not 0; determine the observed value, the overall mean, and the overall standard deviation; determine the standard score based on the observed value, the overall mean, the overall standard deviation, and the continuity correction coefficient; determine the probability value under the null hypothesis based on the standard score; accept the alternative hypothesis in response to the probability value being less than a preset significance level; and accept the null hypothesis in response to the probability value being greater than a preset significance level.

[0070] In some embodiments, the device 400 also includes a confidence interval determination unit configured to determine the confidence interval based on the standard score, the sample size of the sample population, the sample size of the first sample set, the probability value and the continuity correction coefficient.

[0071] In some embodiments, the apparatus 400 further comprises a minimum sample size determination unit configured to determine the minimum sample size based on the sample size, the standard score and the continuity correction coefficient of the sample population.

[0072] In some embodiments, the apparatus 400 further includes a target sample size determination unit configured to determine the target sample size based on the minimum sample size and the error level.

[0073] In some embodiments, the biometric verification data includes face verification data, and the face verification data is obtained by offline extraction.

[0074] It should be noted that the reference Figures 1 to 3 Further actions or steps shown can be performed by Figure 4 For example, the device 400 may include more modules or units to implement the actions or steps described above, or Figure 4 Some of the units or modules shown may be further configured to implement the actions or steps described above, which will not be repeated here.

[0075] Figure 5 Schematic block diagram of an example device 500 that can be used to implement some embodiments of the present specification is shown. Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 502 or computer program instructions loaded from a storage unit 506 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0076] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0077] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a CPU, a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as method 200. For example, in some embodiments, the method 200 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the method 200 described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the method 200 in any other appropriate manner (e.g., by means of firmware).

[0078] In some embodiments, the methods and processes described above may be implemented as a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present specification.

[0079] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0080] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0081] The computer program instructions for performing the operation of each embodiment of this specification can be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data or source code or object code written in any combination of one or more programming languages, and the programming languages ​​include object-oriented programming languages, and conventional procedural programming languages. Computer-readable program instructions can be executed completely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer by any type of network-including local area network (LAN) or wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect by the Internet). In certain embodiments, by using the state information of computer-readable program instructions to customize electronic circuits, such as programmable logic circuits, field programmable gate arrays (FPGAs) or programmable logic arrays (PLAs), the electronic circuits can execute computer-readable program instructions, thereby realizing the various aspects of this specification.

[0082] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0083] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0084] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The embodiments of the present specification have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

[0086] The above are only optional embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A sampling method for biometric verification data, comprising: Obtaining a first sample set from a sample population by a first sampling method, the sample population including at least two biometric verification data and a population characteristic probability, the biometric verification data being annotated with a sample characteristic value, the sample characteristic value including verification success and verification failure, and the population characteristic probability being used to characterize the proportion of verification success data to verification failure data in the sample population; Determine a first feature probability corresponding to the first sample set, where the first feature probability is used to characterize a proportion of verification success data to verification failure data in the first sample set; as well as In response to the first feature probability and the overall feature probability satisfying a probability deviation condition, a second sample set is obtained from the sample population by a second sampling method, where the second sampling method is different from the first sampling method.

2. The method according to claim 1, wherein the method further comprises: Determine a second feature probability corresponding to the second sample set, where the second feature probability is used to characterize a proportion of verification success data to verification failure data in the second sample set; as well as In response to the second feature probability and the sample feature probability not satisfying a probability deviation condition, determining that the second sample set is a simulation data set.

3. The method of claim 1 , wherein determining the deviation between the first feature probability and the overall feature probability comprises: Determining a deviation test method based on the overall characteristic probability and the number of samples in the first sample set, the deviation test method including a first deviation test method and a second deviation test method; In response to the overall feature probability and the number of samples in the first sample set satisfying a first condition, determining a deviation between the first feature probability and the overall feature probability by the first deviation detection method; as well as In response to the overall feature probability and the number of samples in the first sample set satisfying a second condition, the deviation between the first feature probability and the overall feature probability is determined by the second deviation detection method.

4. The method according to claim 3, wherein determining the deviation between the first feature probability and the overall feature probability by the first deviation test method comprises: Establishing a null hypothesis and an alternative hypothesis, wherein the null hypothesis is that the deviation between the first feature probability and the overall feature probability is 0, and the alternative hypothesis is that the deviation between the first feature probability and the overall feature probability is not 0; Determine the probability value under the null hypothesis; In response to the probability value being less than a preset significance level, accepting the alternative hypothesis; as well as In response to the probability value being greater than a preset significance level, the null hypothesis is accepted.

5. The method according to claim 3, wherein determining the deviation between the first feature probability and the overall feature probability by the second deviation test method comprises: Establishing a null hypothesis and an alternative hypothesis, wherein the null hypothesis is that the deviation between the first feature probability and the overall feature probability is 0, and the alternative hypothesis is that the deviation between the first feature probability and the overall feature probability is not 0; Determine the observed values, population mean, and population standard deviation; Determining a standard score based on the observed value, the population mean, the population standard deviation, and a continuity correction factor; Based on the standard score, determining the probability value under the null hypothesis; In response to the probability value being less than a preset significance level, accepting the alternative hypothesis; as well as In response to the probability value being greater than a preset significance level, the null hypothesis is accepted.

6. The method according to claim 5, wherein the method further comprises: A confidence interval is determined based on the standard score, the sample size of the sample population, the sample size of the first sample set, the probability value, and the continuity correction coefficient.

7. The method according to claim 5, wherein the method further comprises: A minimum sample size is determined based on the sample size of the sample population, the standard score, and the continuity correction factor.

8. The method according to claim 7, wherein the method further comprises: Based on the minimum sample size and the error level, a target sample size is determined.

9. The method according to claim 1, wherein the biometric verification data includes face verification data, and the face verification data is obtained by offline extraction.

10. A sampling system for biometric verification data, comprising: A first sample set acquisition unit is configured to acquire a first sample set from a sample population by a first sampling method, wherein the sample population includes at least two biometric verification data and a population characteristic probability, wherein the biometric verification data is annotated with a sample characteristic value, wherein the sample characteristic value includes verification success and verification failure, and the population characteristic probability is used to characterize the proportion of verification success data to verification failure data in the sample population; A first feature probability determination unit is configured to determine a first feature probability corresponding to the first sample set, wherein the first feature probability is used to characterize a proportion of verification success data to verification failure data in the first sample set; as well as The second sample set acquisition unit is configured to acquire a second sample set from the sample population by a second sampling method in response to the first feature probability and the population feature probability satisfying a probability deviation condition, where the second sampling method is different from the first sampling method.

11. A computing device comprising: at least one processing unit; At least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the computing device to perform the method as claimed in any one of claims 1 to 9.

12. A computer program product comprising machine executable instructions which, when executed by a device, cause the device to perform the method of any one of claims 1 to 9.

Images