A method and system for verifying customer identity
By calculating the difference between the face recognition matching threshold and the matching value, determining the risk category vector, and selecting alternative identity review methods, the problem of smooth transactions after the face recognition failure is solved, and the operational efficiency and risk control capabilities of bank branches are improved.
Patent Information
- Application Number
- CN202210490700.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-07
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, after the facial recognition fails, staff need manual intervention, which is inefficient and has poor customer experience, and the transaction risk is high in high risk scenarios. How to ensure the smooth progress of the transaction is an urgent problem to be solved.
By calculating the difference between the face recognition matching threshold and the matching value, determining the risk category vector, combining the risk category vectors of multiple identity audit methods, selecting an appropriate identity audit method to replace face recognition, ensuring the smooth progress of customer identity audits.
Effectively analyze the failure of facial recognition, improve the operational efficiency of bank branches, reduce operational risks, realize digital risk control, and ensure the safety of customer property.
Smart Images

Figure CN114840835B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of bank big data processing, and in particular to a customer identity verification method and system. Background Art
[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.
[0003] At present, when customers conduct business in banks, in order to reduce risks, they often need to verify their identities in a certain way. In existing transaction scenarios, face recognition is a commonly used identity verification method; because the parameter settings of face recognition are not suitable for some people, face recognition mismatch problems often occur.
[0004] Usually, when the above problems occur, manual intervention by staff is often required, which is inefficient and provides a poor customer experience. At the same time, in high-risk scenarios, there are many uncertainties, which makes customer transactions more risky. Therefore, after facial recognition fails, how to further ensure the smooth progress of transactions is an urgent problem to be solved.
[0005] In summary, there is an urgent need for a technical solution that can overcome the above-mentioned defects and verify the customer's identity after face recognition fails. Summary of the invention
[0006] In order to solve the problems existing in the prior art, the present invention proposes a customer identity verification method and system.
[0007] In a first aspect of an embodiment of the present invention, a method for verifying a customer's identity is provided, comprising:
[0008] When a customer fails to perform face recognition at a bank branch, the difference between the face recognition matching threshold and the current face matching value is calculated, and the difference is used as the matching difference corresponding to the current face recognition failure;
[0009] According to the matching difference corresponding to the current face recognition failure, determine the risk category vector corresponding to the current face recognition failure;
[0010] For multiple identity verification methods, determine the risk category vector corresponding to each identity verification method;
[0011] Based on the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to this face failure, determine the identity verification method corresponding to the customer after this face recognition failure.
[0012] In a second aspect of an embodiment of the present invention, a customer identity verification system is provided, comprising:
[0013] A matching difference calculation module is used to calculate the difference between the face recognition matching threshold and the current face matching value when the customer fails to perform face recognition at a bank branch, and use the difference as the matching difference corresponding to the current face recognition failure;
[0014] The face recognition processing module is used to determine the risk category vector corresponding to the face recognition failure according to the matching difference corresponding to the face recognition failure;
[0015] An identity verification method processing module is used to determine the risk category vector corresponding to each identity verification method for multiple identity verification methods;
[0016] The identity verification method determination module is used to determine the identity verification method corresponding to the customer after the face recognition failure based on the risk category vector corresponding to each identity verification method and the risk category vector corresponding to the current face failure.
[0017] In a third aspect of an embodiment of the present invention, a computer device is proposed, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements a customer identity verification method when executing the computer program.
[0018] In a fourth aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and the computer program implements a customer identity verification method when executed by a processor.
[0019] In a fifth aspect of an embodiment of the present invention, a computer program product is proposed. The computer program product includes a computer program. When the computer program is executed by a processor, a method for verifying a customer's identity is implemented.
[0020] The customer identity verification method and system proposed in the present invention calculate the difference between the face recognition matching threshold and the face matching value of this time when the face recognition of the customer fails, and use the difference as the matching difference corresponding to the face recognition failure of this time; determine the risk category vector corresponding to the face recognition failure of this time according to the matching difference corresponding to the face recognition failure of this time; for multiple identity verification methods, respectively determine the risk category vector corresponding to each identity verification method; determine the identity verification method corresponding to the customer after the face recognition failure of this time according to the risk category vector corresponding to each identity verification method and the risk category vector corresponding to the face failure of this time. The present invention can effectively analyze the situation of customer face recognition failure, add other identity verification methods to perform identity verification, ensure the safety of customer property, improve the operating efficiency of bank branches, realize the digital risk control of banks, and effectively reduce the operating risks of bank branches. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0022] Figure 1 The figure is a flow chart of a method for verifying a customer's identity according to an embodiment of the present invention.
[0023] Figure 2 It is a flow chart of a method for verifying a customer's identity according to another embodiment of the present invention.
[0024] Figure 3 It is a schematic diagram of a specific process of determining a risk category vector corresponding to the current face recognition failure according to an embodiment of the present invention.
[0025] Figure 4 It is a specific flow chart of determining the risk category vector corresponding to each identity verification method according to an embodiment of the present invention.
[0026] Figure 5 It is a specific flow chart of determining the identity verification method corresponding to the customer after the face recognition fails in one embodiment of the present invention.
[0027] Figure 6 It is a specific flow chart of determining a risk control model according to risk type according to an embodiment of the present invention.
[0028] Figure 7 It is a schematic diagram of the customer identity verification system architecture according to an embodiment of the present invention.
[0029] Figure 8 It is a schematic diagram of the customer identity verification system architecture of another embodiment of the present invention.
[0030] Fig. 9 It is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION
[0031] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. On the contrary, these embodiments are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.
[0032] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.
[0033] According to an embodiment of the present invention, a customer identity verification method and system are proposed, which relate to the technical field of bank big data processing.
[0034] The principle and spirit of the present invention are explained in detail below with reference to several representative embodiments of the present invention.
[0035] Figure 1 FIG. 1 is a flow chart of a method for verifying a customer's identity according to an embodiment of the present invention. Figure 1 As shown, the method includes:
[0036] S1, when a customer fails to perform face recognition at a bank branch, the difference between the face recognition matching threshold and the current face matching value is calculated, and the difference is used as the matching difference corresponding to the current face recognition failure;
[0037] S2, determining the risk category vector corresponding to the current face recognition failure according to the matching difference corresponding to the current face recognition failure;
[0038] S3, for multiple identity verification methods, determine the risk category vector corresponding to each identity verification method;
[0039] S4, based on the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to this face failure, determine the identity verification method corresponding to the customer after this face recognition failure.
[0040] In another embodiment, reference Figure 2 , the method further comprises:
[0041] S5, for each identity verification method corresponding to the customer after the face recognition failure, determine the risk type corresponding to the identity verification method, determine a risk control model according to the risk type, and determine the risk control model as the risk control model corresponding to the identity verification method;
[0042] S6. When the customer's identity verification method is passed, the risk control model corresponding to the passed identity verification method is used to perform risk control on the customer's transaction.
[0043] In order to explain the above customer identity verification method more clearly, each step is explained in detail below.
[0044] In S1, when a customer fails in face recognition at a bank branch, the difference between the face recognition matching threshold and the current face matching value is calculated, and the difference is used as the matching difference corresponding to the current face recognition failure.
[0045] In S2, reference Figure 3 , according to the matching difference corresponding to the face recognition failure, the specific process of determining the risk category vector corresponding to the face recognition failure is:
[0046] S201, obtaining the bank's historical face recognition failure data;
[0047] S202, selecting, according to the matching difference, from the bank's historical facial recognition failure data, facial recognition failure data having a corresponding matching difference equal to the matching difference;
[0048] S203, determining the risk coefficient of each risk type corresponding to the current face recognition failure according to the selected face recognition failure data;
[0049] S204, determining the risk category vector corresponding to the current face recognition failure based on the risk coefficients of each risk type corresponding to the current face recognition failure; wherein each component of the risk category vector corresponds one-to-one to a risk type of the bank, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the current face recognition failure.
[0050] In one embodiment, (S203) the risk coefficients of each risk type corresponding to the current face recognition failure are determined according to the selected face recognition failure data, specifically:
[0051] According to the selected face recognition failure data, determine the period range included in the face recognition failure data as the first period, and determine the risk coefficient of each risk type for each day of the matching difference corresponding to the face recognition failure in the first period;
[0052] For each risk type, based on the risk coefficients (multiple samples of the same probability distribution) of the risk type for all days in the first period for the matching difference corresponding to the current face recognition failure, determine the variance of the risk type corresponding to the current face recognition failure;
[0053] Select an acceptable risk coefficient error threshold ε, the probability P that the acceptable risk coefficient error is greater than ε, and determine the sample threshold corresponding to the face recognition failure as the maximum value of the sample thresholds of each risk type corresponding to the face recognition failure, where the sample threshold of the i-th risk type corresponding to the face recognition failure is where σ i is the variance of the i-th risk type corresponding to the face recognition failure;
[0054] Select a suitable setting value so that the absolute value of the difference with the matching difference corresponding to the current face recognition failure is less than the sum of the days corresponding to all matching differences of the setting value is greater than the sample threshold corresponding to the current face recognition failure, wherein the days corresponding to each matching difference is the number of days in the period range of the face recognition failure data of the bank whose corresponding matching difference is equal to the matching difference in the historical face recognition failure data;
[0055] Determine multiple matching differences whose absolute values of the differences from the matching differences corresponding to the current face recognition failure are smaller than the set value as similar matching differences;
[0056] For each similarity matching difference, select the face recognition failure data whose corresponding matching difference is equal to the similarity matching difference from the bank's historical face recognition failure data, determine the period range included in the selected face recognition failure data as the period corresponding to the similarity matching difference, and determine the risk coefficient of each risk type for each day of the similarity matching difference in the period corresponding to the similarity matching difference; for each risk type, use the determined risk coefficient of the similarity matching difference for all days in the period corresponding to the similarity matching difference as multiple samples of the similarity matching difference for the risk type;
[0057] For each risk type, the risk coefficient of the face recognition failure corresponding to the risk type is determined as the mean of all similar matching differences for all samples of the risk type.
[0058] Among them, for any day, the risk coefficient of each risk type for the matching difference on that day is determined as the proportion of face recognition failure data involving each risk type in the bank's historical face recognition failure data on that day and the corresponding matching difference is equal to the matching difference.
[0059] In S3, reference Figure 4 ,For multiple identity verification methods, the specific process of determining the risk category vector corresponding to each identity verification method is as follows:
[0060] S301, for each identity verification method, determine the risk coefficient of each risk type corresponding to the identity verification method according to the historical transaction data of the identity verification method;
[0061] S302, determine the risk category vector corresponding to the identity verification method according to the risk coefficients of each risk type corresponding to the identity verification method; wherein each component of the risk category vector corresponds one-to-one to a risk type of the bank, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the identity verification method.
[0062] In one embodiment, before determining the risk category vector corresponding to each identity verification method, bank customers are first classified to obtain multiple customer categories; after obtaining the multiple customer categories, for each identity verification method, based on the historical transaction data of each customer category regarding the identity verification method, the risk coefficient of each customer category and each risk type corresponding to the identity verification method is determined, as well as the risk category vector of each customer category corresponding to the identity verification method.
[0063] Among them, bank customers are classified to obtain multiple customer categories, which can be:
[0064] S31, determining multiple customer dimensions, for example, determining bank transactions as customer dimensions, the value of each customer in each customer dimension is equal to the transaction quantity of bank transactions corresponding to the customer dimension in the transaction data of the customer; for any customer dimension, determining a function corresponding to the customer dimension, the independent variables of the function are two customers, and the corresponding function value is the absolute value of the difference between the values of the two customers in the customer dimension;
[0065] S32, determining that the function corresponding to the bank customer is the P-th root of the sum of the P-th powers of the functions corresponding to all customer dimensions, where P is a positive integer greater than 1; and selecting a clustering algorithm to cluster the bank customers according to the function corresponding to the bank customer to obtain multiple customer categories;
[0066] S33, for each customer category obtained, determine the transaction with the largest transaction data in the historical transaction data of each customer in the customer category, and determine the transaction as the largest transaction of the customer. For each largest transaction corresponding to the customer category, determine the ratio of the number of customers of the customer category corresponding to the largest transaction to the number of customers of the customer category, select the largest ratio from all the ratios, and determine the largest ratio as the consistency index of the customer category; determine whether the customer category meets the following condition p: the consistency index of the customer category is greater than the set value, if not, continue to cluster all customers of the customer category using the clustering algorithm until each newly generated customer category meets the condition p.
[0067] In one embodiment, (S4) determines the identity verification method corresponding to the customer after the face recognition failure based on the risk category vector corresponding to each identity verification method and the risk category vector corresponding to the face failure. It can also be: determining the customer category of the customer; determining the identity verification method corresponding to the customer after the face recognition failure based on the risk category vector corresponding to the customer category of each identity verification method and the risk category vector corresponding to the face failure.
[0068] In S4, ref. Figure 5, based on the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to this face failure, the specific process of determining the identity verification method corresponding to the customer after this face recognition failure is as follows:
[0069] S401, determining a partial order of identity verification methods based on the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face recognition failure, wherein the partial order is used to determine whether a first identity verification method is better than a second identity verification method among any two identity verification methods;
[0070] S402, determining multiple maximum identity verification methods among all identity verification methods of the bank according to the partial order of the identity verification methods, wherein the maximum identity verification method is a maximum element of the partial order;
[0071] S403, determining the multiple maximum identity verification methods as the identity verification methods corresponding to the customer after the face recognition fails this time.
[0072] In one embodiment, (S401) the detailed method of determining the partial order of the identity verification methods according to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face failure is:
[0073] When determining the partial order of identity verification methods, for any two identity verification methods, calculate the vector difference between the risk category vector corresponding to the first of the two identity verification methods and the risk category vector corresponding to the second of the two identity verification methods, and calculate the first distance between the risk category vector corresponding to the first identity verification method and the risk category vector corresponding to the current face failure, and calculate the second distance between the risk category vector corresponding to the second identity verification method and the risk category vector corresponding to the current face failure. If each component of the vector difference is less than or equal to 0 and the first distance is greater than or equal to the second distance, then it is determined that the first identity verification method is better than the second identity verification method.
[0074] It should be noted that the maximum element of the partial order is that in the set corresponding to the partial order, there is no other element that is better than the maximum element. In this application, multiple maximum identity verification methods among all identity verification methods of the bank specifically refer to the absence of other identity verification methods among all identity verification methods, such that the other identity verification methods are better than the maximum identity verification method.
[0075] In one embodiment, (S402) a method for determining a plurality of maximum identity verification methods among all identity verification methods of a bank according to the partial order of the identity verification methods is as follows:
[0076] S4021, initializing the maximum flag corresponding to each identity verification method of the bank to possible, and initializing the corresponding comparison flag to yes;
[0077] S4022, for each identity verification method of the bank, if the maximum identifier corresponding to the identity verification method is not possible, then continue to determine whether the maximum identifier corresponding to the next identity verification method is possible, otherwise, select multiple other identity verification methods whose corresponding comparison identifiers are yes from all other identity verification methods except the identity verification method, and determine the selected multiple other identity verification methods as the identity verification methods to be compared corresponding to the identity verification method;
[0078] S4023, for each identity verification method to be compared corresponding to the identity verification method, if the identity verification method is better than the identity verification method to be compared, then update the maximum flag corresponding to the identity verification method to be compared to no, and determine the identity verification method to be compared as the secondary identity verification method of the identity verification method; if the identity verification method to be compared is better than the identity verification method, then update the maximum flag corresponding to the identity verification method to no, and then continue to execute the above step S4022 for the next identity verification method; otherwise, do not update the maximum flags corresponding to the identity verification method and the identity verification method to be compared;
[0079] S4024, if it is confirmed that all identity verification methods to be compared with the identity verification method are not better than the identity verification method, the identity verification method is determined as the largest identity verification method among all identity verification methods of the bank, and the comparison flag of each sub-identity verification method of the largest identity verification method is updated to no;
[0080] S4025, then continue to execute step S4022 for the next identity verification method until the above steps are completed for all identity verification methods of the bank.
[0081] In S5, ref. Figure 6 For each identity verification method corresponding to the customer after the face recognition failure, determine the risk type corresponding to the identity verification method, determine the risk control model according to the risk type, and determine the risk control model as the risk control model corresponding to the identity verification method. The specific process is as follows:
[0082] S501, for each component of the risk category vector, calculate the difference between the component value of the risk category vector corresponding to the identity verification method and the component value of the risk category vector corresponding to the current face failure, and use the difference as the difference corresponding to the component;
[0083] S502, determining the risk type corresponding to the component with the largest corresponding difference as the risk type corresponding to the identity verification method;
[0084] S503: Determine the risk control model corresponding to the risk type corresponding to the identity verification method as the risk control model corresponding to the identity verification method.
[0085] It should be noted that the risk control model corresponding to the risk type includes two parts: the prediction model corresponding to the risk type, and the control means corresponding to the prediction result. Among them, the prediction model corresponding to the risk type is a model obtained by training the transaction data involving the risk type with a machine learning model (such as a tree model, a probability model, a neural network model), which is used to predict the risk probability of each transaction data. Different prediction results can correspond to different control means (control means may include: limiting the customer's transaction amount, performing other forms of identity authentication on the customer, freezing the customer's account, rejecting the customer's current transaction, etc.). For example, when the prediction result is the probability of the occurrence of the risk type, the corresponding control means can be determined based on the probability.
[0086] In S6, when the identity verification method of the customer is passed, the risk control model corresponding to the passed identity verification method is used to perform risk control on the customer's transaction.
[0087] In one embodiment, after the facial recognition fails, all risk control models corresponding to all identity verification methods corresponding to the customer can be sent to the bank branch (such as the redis of the bank branch's edge system) in the form of an executable program. This can reduce the customer's transaction waiting time and ensure that the risk control model is not obtained by others.
[0088] It should be noted that, although the operations of the method of the present invention are described in a specific order in the above embodiments and the accompanying drawings, this does not require or imply that the operations must be performed in the specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.
[0089] After introducing the method of the exemplary embodiment of the present invention, next, refer to Figure 7 A customer identity verification system according to an exemplary embodiment of the present invention is introduced.
[0090] The implementation of the customer identity verification system can refer to the implementation of the above method, and the repeated parts will not be repeated. The term "module" or "unit" used below can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.
[0091] Based on the same inventive concept, the present invention also proposes a customer identity verification system, such as Figure 7 As shown, the system includes:
[0092] The matching difference calculation module 710 is used to calculate the difference between the face recognition matching threshold and the current face matching value when the face recognition of the customer fails at the bank branch, and use the difference as the matching difference corresponding to the current face recognition failure;
[0093] The current face recognition processing module 720 is used to determine the risk category vector corresponding to the current face recognition failure according to the matching difference corresponding to the current face recognition failure;
[0094] The identity verification mode processing module 730 is used to determine the risk category vector corresponding to each identity verification mode for each of the multiple identity verification modes;
[0095] The identity verification method determination module 740 is used to determine the identity verification method corresponding to the customer after the face recognition fails based on the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the face failure.
[0096] In one embodiment, reference Figure 8 , the system further comprises:
[0097] The risk control model determination module 750 is used to determine the risk type corresponding to each identity verification method corresponding to the customer after the face recognition failure, determine the risk control model according to the risk type, and determine the risk control model as the risk control model corresponding to the identity verification method;
[0098] The risk control module 760 is used to perform risk control on the transaction of the customer using the risk control model corresponding to the passed identity verification method after the customer's identity verification method is passed.
[0099] In one embodiment, the face recognition processing module is specifically used for:
[0100] Obtain the bank’s historical facial recognition failure data;
[0101] According to the matching difference, selecting the facial recognition failure data with the corresponding matching difference equal to the matching difference from the historical facial recognition failure data of the bank;
[0102] According to the selected face recognition failure data, the risk coefficient of each risk type corresponding to the face recognition failure is determined;
[0103] According to the risk coefficients of each risk type corresponding to the face recognition failure, the risk category vector corresponding to the face recognition failure is determined; wherein each component of the risk category vector corresponds one-to-one to a risk type of the bank, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the face recognition failure.
[0104] In one embodiment, the identity verification mode processing module is specifically used to:
[0105] For each identity verification method, determine the risk coefficient of each risk type corresponding to the identity verification method based on the historical transaction data of the identity verification method;
[0106] According to the risk coefficients of each risk type corresponding to the identity verification method, the risk category vector corresponding to the identity verification method is determined; wherein each component of the risk category vector corresponds one-to-one to a risk type of the bank, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the identity verification method.
[0107] In one embodiment, the identity verification method determination module is specifically used to:
[0108] Determine the partial order of identity verification methods according to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face recognition failure, wherein the partial order is used to determine whether the first identity verification method is better than the second identity verification method among any two identity verification methods;
[0109] According to the partial order of the identity verification methods, a plurality of maximum identity verification methods among all the identity verification methods of the bank are determined, wherein the maximum identity verification method is a maximum element of the partial order;
[0110] The multiple maximum identity verification methods are determined as the identity verification methods corresponding to the customer after the face recognition fails this time.
[0111] In one embodiment, the identity verification method determination module is specifically used to:
[0112] When determining the partial order of identity verification methods, for any two identity verification methods, calculate the vector difference between the risk category vector corresponding to the first of the two identity verification methods and the risk category vector corresponding to the second of the two identity verification methods, and calculate the first distance between the risk category vector corresponding to the first identity verification method and the risk category vector corresponding to the current face failure, and calculate the second distance between the risk category vector corresponding to the second identity verification method and the risk category vector corresponding to the current face failure. If each component of the vector difference is less than or equal to 0 and the first distance is greater than or equal to the second distance, then it is determined that the first identity verification method is better than the second identity verification method.
[0113] In one embodiment, the risk control model determination module is specifically used to:
[0114] For each component of the risk category vector, calculate the difference between the component value of the risk category vector corresponding to the identity verification method and the component value of the risk category vector corresponding to the current face failure, and use the difference as the difference corresponding to the component;
[0115] The risk type corresponding to the component with the largest corresponding difference is determined as the risk type corresponding to the identity verification method;
[0116] The risk control model corresponding to the risk type corresponding to the identity verification method is determined as the risk control model corresponding to the identity verification method.
[0117] It should be noted that although several modules of the customer identity verification system are mentioned in the above detailed description, such division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided into multiple modules for embodiment.
[0118] Based on the above invention concept, Fig. 9 As shown, the present invention also proposes a computer device 900, including a memory 910, a processor 920, and a computer program 930 stored in the memory 910 and executable on the processor 920, wherein the processor 920 implements the aforementioned customer identity verification method when executing the computer program 930.
[0119] Based on the aforementioned inventive concept, the present invention proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the aforementioned customer identity verification method is implemented.
[0120] Based on the aforementioned inventive concept, the present invention proposes a computer program product, wherein the computer program product includes a computer program, and the computer program implements a customer identity verification method when executed by a processor.
[0121] The customer identity verification method and system proposed in the present invention calculate the difference between the face recognition matching threshold and the face matching value of this time when the face recognition of the customer fails, and use the difference as the matching difference corresponding to the face recognition failure of this time; determine the risk category vector corresponding to the face recognition failure of this time according to the matching difference corresponding to the face recognition failure of this time; for multiple identity verification methods, respectively determine the risk category vector corresponding to each identity verification method; determine the identity verification method corresponding to the customer after the face recognition failure of this time according to the risk category vector corresponding to each identity verification method and the risk category vector corresponding to the face failure of this time. The present invention can effectively analyze the situation of customer face recognition failure, add other identity verification methods to perform identity verification, ensure the safety of customer property, improve the operating efficiency of bank branches, realize the digital risk control of banks, and effectively reduce the operating risks of bank branches.
[0122] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0123] The present invention is described with reference to flowcharts and / or block diagrams of methods and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0124] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.
[0125] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0126] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims
1. A customer identity verification method, characterized in that: include: When a customer fails to perform face recognition at a bank branch, the difference between the face recognition matching threshold and the current face matching value is calculated, and the difference is used as the matching difference corresponding to the current face recognition failure; According to the matching difference corresponding to the current face recognition failure, determine the risk category vector corresponding to the current face recognition failure; For multiple identity verification methods, determine the risk category vector corresponding to each identity verification method; According to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face recognition failure, determine the identity verification method corresponding to the customer after the current face recognition failure; Among them, according to the matching difference corresponding to the current face recognition failure, the risk category vector corresponding to the current face recognition failure is determined, including: Obtain the bank’s historical facial recognition failure data; According to the matching difference, selecting the facial recognition failure data with the corresponding matching difference equal to the matching difference from the historical facial recognition failure data of the bank; According to the selected face recognition failure data, the risk coefficient of each risk type corresponding to the face recognition failure is determined; According to the risk coefficients of each risk type corresponding to the current face recognition failure, determine the risk category vector corresponding to the current face recognition failure; wherein each component of the risk category vector corresponds to a risk type of the bank one by one, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the current face recognition failure; Among them, for multiple identity verification methods, the risk category vector corresponding to each identity verification method is determined respectively, including: For each identity verification method, determine the risk coefficient of each risk type corresponding to the identity verification method based on the historical transaction data of the identity verification method; According to the risk coefficients of each risk type corresponding to the identity verification method, the risk category vector corresponding to the identity verification method is determined; wherein each component of the risk category vector corresponds one-to-one to a risk type of the bank, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the identity verification method.
2. The method according to claim 1, characterized in that Also includes: For each identity verification method corresponding to the customer after the face recognition failure, determine the risk type corresponding to the identity verification method, determine the risk control model based on the risk type, and determine the risk control model as the risk control model corresponding to the identity verification method; When the customer's identity verification method is passed, the risk control model corresponding to the passed identity verification method is used to perform risk control on the customer's transaction.
3. The method according to claim 1, characterized in that According to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to this face failure, the identity verification method corresponding to the customer after this face recognition failure is determined, including: Determine the partial order of identity verification methods according to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face recognition failure, wherein the partial order is used to determine whether the first identity verification method is better than the second identity verification method among any two identity verification methods; According to the partial order of the identity verification methods, a plurality of maximum identity verification methods among all the identity verification methods of the bank are determined, wherein the maximum identity verification method is a maximum element of the partial order; The multiple maximum identity verification methods are determined as the identity verification methods corresponding to the customer after the face recognition fails this time.
4. The method according to claim 3, characterized in that: According to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face recognition failure, the partial order of the identity verification methods is determined, including: When determining the partial order of identity verification methods, for any two identity verification methods, calculate the vector difference between the risk category vector corresponding to the first of the two identity verification methods and the risk category vector corresponding to the second of the two identity verification methods, and calculate the first distance between the risk category vector corresponding to the first identity verification method and the risk category vector corresponding to the current face failure, and calculate the second distance between the risk category vector corresponding to the second identity verification method and the risk category vector corresponding to the current face failure. If each component of the vector difference is less than or equal to 0 and the first distance is greater than or equal to the second distance, then it is determined that the first identity verification method is better than the second identity verification method.
5. The method according to claim 2, characterized in that: For each identity verification method corresponding to the customer after the face recognition failure, determine the risk type corresponding to the identity verification method, determine the risk control model according to the risk type, and determine the risk control model as the risk control model corresponding to the identity verification method, including: For each component of the risk category vector, calculate the difference between the component value of the risk category vector corresponding to the identity verification method and the component value of the risk category vector corresponding to the current face failure, and use the difference as the difference corresponding to the component; The risk type corresponding to the component with the largest corresponding difference is determined as the risk type corresponding to the identity verification method; The risk control model corresponding to the risk type corresponding to the identity verification method is determined as the risk control model corresponding to the identity verification method.
6. A customer identity verification system, characterized in that: include: The matching difference calculation module is used to calculate the difference between the face recognition matching threshold and the current face matching value when the customer fails to perform face recognition at the bank branch, and use the difference as the matching difference corresponding to the current face recognition failure; The face recognition processing module is used to determine the risk category vector corresponding to the face recognition failure according to the matching difference corresponding to the face recognition failure; An identity verification method processing module is used to determine the risk category vector corresponding to each identity verification method for multiple identity verification methods; An identity verification method determination module is used to determine the identity verification method corresponding to the customer after the face recognition failure according to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the face failure; Among them, the face recognition processing module is specifically used for: Obtain the bank’s historical facial recognition failure data; According to the matching difference, selecting the facial recognition failure data with the corresponding matching difference equal to the matching difference from the historical facial recognition failure data of the bank; According to the selected face recognition failure data, the risk coefficient of each risk type corresponding to the face recognition failure is determined; According to the risk coefficients of each risk type corresponding to the current face recognition failure, determine the risk category vector corresponding to the current face recognition failure; wherein each component of the risk category vector corresponds to a risk type of the bank one by one, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the current face recognition failure; Among them, the identity verification method processing module is specifically used for: For each identity verification method, determine the risk coefficient of each risk type corresponding to the identity verification method based on the historical transaction data of the identity verification method; According to the risk coefficients of each risk type corresponding to the identity verification method, the risk category vector corresponding to the identity verification method is determined; wherein each component of the risk category vector corresponds one-to-one to a risk type of the bank, and the component value of each component is equal to the risk coefficient of the risk type corresponding to the component of the identity verification method.
7. The system according to claim 6, characterized in that Also includes: Risk control model determination module and risk control module; among them, A risk control model determination module is used to determine the risk type corresponding to each identity verification method corresponding to the customer after the face recognition failure, determine the risk control model according to the risk type, and determine the risk control model as the risk control model corresponding to the identity verification method; The risk control module is used to perform risk control on the customer's transaction using the risk control model corresponding to the passed identity verification method after the customer's identity verification method is passed.
8. The system according to claim 6, characterized in that The identity verification method determination module is specifically used for: Determine the partial order of identity verification methods according to the risk category vectors corresponding to each identity verification method and the risk category vector corresponding to the current face recognition failure, wherein the partial order is used to determine whether the first identity verification method is better than the second identity verification method among any two identity verification methods; According to the partial order of the identity verification methods, a plurality of maximum identity verification methods among all the identity verification methods of the bank are determined, wherein the maximum identity verification method is a maximum element of the partial order; The multiple maximum identity verification methods are determined as the identity verification methods corresponding to the customer after the face recognition fails this time.
9. The system according to claim 8, characterized in that The identity verification method determination module is specifically used for: When determining the partial order of identity verification methods, for any two identity verification methods, calculate the vector difference between the risk category vector corresponding to the first of the two identity verification methods and the risk category vector corresponding to the second of the two identity verification methods, and calculate the first distance between the risk category vector corresponding to the first identity verification method and the risk category vector corresponding to the current face failure, and calculate the second distance between the risk category vector corresponding to the second identity verification method and the risk category vector corresponding to the current face failure. If each component of the vector difference is less than or equal to 0 and the first distance is greater than or equal to the second distance, then it is determined that the first identity verification method is better than the second identity verification method.
10. The system according to claim 7, characterized in that The risk control model determination module is specifically used for: For each component of the risk category vector, calculate the difference between the component value of the risk category vector corresponding to the identity verification method and the component value of the risk category vector corresponding to the current face failure, and use the difference as the difference corresponding to the component; The risk type corresponding to the component with the largest corresponding difference is determined as the risk type corresponding to the identity verification method; The risk control model corresponding to the risk type corresponding to the identity verification method is determined as the risk control model corresponding to the identity verification method.
11. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.
12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.
13. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.
