Payment identity trust verification protection method and system

Abstract

This invention relates to a payment identity trusted verification and protection method, belonging to the field of electronic digital data processing, and more specifically to the field of identity recognition. The method includes: using an intelligent trusted verification model to perform intelligent verification of the payment identity based on multiple environmental parameters corresponding to the current financial payment, various target-related information of the target person, and visual trusted verification content. This invention also relates to a payment identity trusted verification and protection system. Through this invention, the technical problem of existing payment identity trusted verification technologies failing to simultaneously meet the increasingly diverse needs of users is addressed. By selectively filtering various visual screening data within the payment identity trusted verification environment of the current financial payment, and combining this with multiple environmental parameters and target-related information corresponding to the current financial payment as auxiliary methods, intelligent verification of whether the person being verified at the financial payment terminal is the target person is completed, thereby solving the aforementioned technical problem.

Description

Technical Field

[0001] The present invention relates to the processing of electronic digital data, more specifically to the field of identity recognition, and particularly to a method and system for secure verification of payment identity. Background Technology

[0002] In common applications of electronic data processing, identity verification, especially in financial payment processes, requires the processing of relevant electronic data to confirm whether the current user of the financial payment terminal is the legitimate user corresponding to the current financial payment, i.e., the target user. The type and method of using the electronic data for identity verification determine the reliability and stability of the payment identity verification process in each financial payment, and indirectly determine whether corresponding payment protection operations are needed subsequently. In addition, the type and method of using the electronic data for identity verification also determine the speed and efficiency of the payment identity verification process in each financial payment.

[0003] For example, Chinese invention patent publication CN110223073A discloses a payment authentication method and apparatus. According to this method and apparatus, a user can send a cardholder authentication request for a specific bank card to a payment platform. The payment platform obtains a dynamic code string, includes this dynamic code string in a deduction request, and requests a deduction from the cardholder's bank. After the cardholder deducts the amount according to the deduction request, a statement is generated, which includes the aforementioned code string. Thus, the user can obtain the dynamic code string by querying their bank statement. The user can then input the code string obtained from the statement into the payment platform. By comparing the user-input code string with the previously generated code string, the system can determine whether the user is the cardholder, thereby verifying the user's payment identity.

[0004] For example, Chinese invention patent publication CN108052997A discloses an intelligent identity verification device. This device includes a first touchscreen device, a second touchscreen device, a first support, a second support, and a payment module. The first touchscreen device includes a first touchscreen, a camera module, and a first smart chip. The first smart chip generates identity verification request data based on identity verification request information received from the first touchscreen and facial images captured by the camera module. The second touchscreen device includes a second touchscreen and a second smart chip. The second smart chip receives identity verification request data from the first touchscreen device, performs facial matching, and generates identity verification data based on the facial matching results. The payment module is used to pay for identity verification services. This invention improves the accuracy, real-time performance, convenience, and efficiency of identity verification.

[0005] However, the aforementioned technical solutions in the existing technologies either only involve conventional payment identity verification and protection mechanisms, such as the former based on dynamic code strings and the latter based on user facial image recognition. The former lacks reliability and stability, while the latter requires a complex and cumbersome operation process. A classic example is the cumbersome and complicated screen recording identity verification process. It is difficult to obtain a payment identity verification that balances reliability, stability, speed, and efficiency. As a result, the existing payment identity verification technologies cannot simultaneously meet the growing and diverse needs of users. Summary of the Invention

[0006] To address the technical problems in existing technologies, this invention provides a payment identity trusted verification and protection method and system. This system can selectively filter various visual screening data sets within the current financial payment's trusted verification environment, combining multiple environment-related parameters and target-related information as auxiliary tools. Employing a customized intelligent trusted verification model, it intelligently verifies whether the person awaiting verification at the financial payment terminal initiating the current payment is indeed the target person. This eliminates the need for cumbersome and complex screen recording and identity verification processes, enabling trusted verification of the payment identity for each financial payment. This provides crucial information for determining whether to execute subsequent payment protection measures, achieving a balance between reliability, stability, speed, and efficiency in trusted payment identity verification.

[0007] According to a first aspect of the present invention, a payment identity trusted verification and protection method is provided, the method comprising:

[0008] The light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment are used as multiple environment-related parameters for the current financial payment.

[0009] Obtain the target-related information of the target personnel corresponding to the current financial payment;

[0010] The continuous imaging device at the financial payment terminal that triggers the current financial payment is used to obtain each frame of real-shot imaging of the payment identity trust verification environment. The human face imaging area that occupies the largest area in each frame of real-shot imaging is used as each reference face imaging area, and each set of visual screening data corresponding to each reference face imaging area is used as the visual trust verification content corresponding to the current financial payment.

[0011] The intelligent and trustworthy verification model uses the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target-related information of the target personnel corresponding to the current financial payment, and the visual trustworthy verification content corresponding to the current financial payment to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target personnel.

[0012] The number of real-shot images in each frame is inversely related to the image resolution of the continuous imaging device.

[0013] Among them, the intelligent trustworthy verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device.

[0014] According to a second aspect of the present invention, a payment identity trusted verification and protection system is provided, the system comprising:

[0015] The first data collection agency is used to collect the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment as multiple environment-related parameters corresponding to the current financial payment.

[0016] The second data collection agency is used to obtain various target-related information of the target personnel corresponding to the current financial payment.

[0017] The third acquisition unit is used to trigger the continuous imaging device at the financial payment terminal of the current financial payment to obtain each frame of real-shot imaging of the payment identity trust verification environment. The human face imaging area that occupies the largest area in each frame of real-shot imaging is used as each reference face imaging area, and each set of visual screening data corresponding to each reference face imaging area is used as the visual trust verification content corresponding to the current financial payment.

[0018] The trusted verification agency is connected to the first acquisition agency, the second acquisition agency, and the third acquisition agency respectively. It is used to use the intelligent trusted verification model to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment.

[0019] The number of real-shot images in each frame is inversely related to the image resolution of the continuous imaging device.

[0020] Among them, the intelligent trustworthy verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device.

[0021] Compared with the prior art, the present invention has at least the following four key inventive points:

[0022] Invention Point A: By selectively filtering visual data from various parts of the current financial payment identity verification environment and combining them with multiple environment-related parameters and target-related information corresponding to the current financial payment as auxiliary, a customized intelligent trusted verification model is adopted to complete the intelligent verification of whether the person to be verified at the financial payment terminal that initiates the current financial payment is the target person. This eliminates the need for a cumbersome and complex screen recording identity verification process, enabling trusted verification of the payment identity for each financial payment and providing key information for deciding whether to implement subsequent payment protection.

[0023] Invention Point B: To perform intelligent verification of whether the person to be verified at the financial payment terminal that initiates the current financial payment is the target person, an intelligent and reliable verification model with a customized structure designed for the financial payment terminal is adopted. The intelligent and reliable verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely correlated with the imaging frame rate of the continuous imaging device, thereby providing a model basis for the accuracy and stability of intelligent verification.

[0024] Invention Point C: To perform intelligent verification of whether the person to be verified at the financial payment terminal initiating the current financial payment is the target person, several basic data points are introduced. These basic data points include visual screening data from the payment identity verification environment in which the current financial payment is being made, multiple environment-related parameters corresponding to the current financial payment, and various target-related information corresponding to the current financial payment. Specifically, a continuous imaging device at the financial payment terminal initiating the current financial payment is triggered to obtain real-time images of each frame of the payment identity verification environment. The human face imaging area occupying the largest area in each frame of the real-time imaging is used as a reference face imaging area, and each reference face imaging area is respectively assigned to... Each set of visual screening data serves as the visual credibility verification content corresponding to the current financial payment. The light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment identity credibility verification environment are used as multiple environment-related parameters corresponding to the current financial payment. The age information, gender identifier, number of recent successful payments, average payment amount, commonly used payment terminal category information, and frequently used payment address coding information of the target personnel of the current financial payment identity credibility verification are used as various target association information of the target personnel corresponding to the current financial payment. The specific data structure design of the above multiple basic data provides a data foundation for the accuracy and stability of intelligent verification.

[0025] Invention Point D: In each training iteration of the Hofit Neural Network, the reliable verification result of whether the person to be verified at the financial payment terminal of a known past financial payment is the target person is used as a single output of the Hofit Neural Network. The imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the past financial payment, multiple environmental parameters corresponding to the past financial payment, various target association information of the target person corresponding to the past financial payment, and the visual reliable verification content corresponding to the past financial payment are used as multiple inputs of the Hofit Neural Network to complete the training, thereby ensuring the training effect of the Hofit Neural Network in each iteration. Attached Figure Description

[0026] The embodiments of the present invention will now be described with reference to the accompanying drawings, wherein:

[0027] Figure 1 This is a schematic diagram of the working scenario of the payment identity trusted verification and protection method and system according to the present invention.

[0028] Figure 2 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 1 of the present invention.

[0029] Figure 3 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 2 of the present invention.

[0030] Figure 4 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 3 of the present invention.

[0031] Figure 5 This is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 4 of the present invention.

[0032] Figure 6 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 5 of the present invention.

[0033] Figure 7 This is a schematic diagram of the payment identity trusted verification and protection system according to Embodiment 6 of the present invention. Detailed Implementation

[0034] like Figure 1The diagram illustrates the working scenario of the payment identity verification and protection method and system according to the present invention. It primarily addresses the cumbersome and complex screen recording identity verification process with light-changing properties in existing financial payment technologies, proposing a convenient and accurate screen recording identity verification mechanism that does not require light-changing, providing targeted reference information for subsequent payment protection. The electronic digital data processing of this invention more specifically relates to the field of identity recognition.

[0035] The specific technical process of this invention is as follows:

[0036] Technical Process 1: To perform intelligent verification of whether the person being verified at the financial payment terminal initiating the current financial payment is the target person, an intelligent and reliable verification model with a customized structure designed for the financial payment terminal is adopted, such as... Figure 1 As shown;

[0037] Specifically, the intelligent trusted verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device.

[0038] Specifically, in each training iteration of the Hofit Neural Network, the trusted verification result of whether the person to be verified at the financial payment terminal of a known past financial payment is the target person is used as a single output of the Hofit Neural Network. The imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the past financial payment, multiple environmental parameters corresponding to the past financial payment, various target association information of the target person corresponding to the past financial payment, and the visual trusted verification content corresponding to the past financial payment are used as multiple inputs of the Hofit Neural Network to complete the training, thereby ensuring the training effect of the Hofit Neural Network in each training iteration.

[0039] In this way, through the above-mentioned targeted customized structural designs, intelligent and reliable verification models with different customized structures have been designed for different financial payment terminals, thereby providing a model foundation for the accuracy and stability of intelligent verification.

[0040] Technical Process 2: To perform intelligent verification of whether the person to be verified at the financial payment terminal that initiates the current financial payment is the target person, a number of basic data points are introduced;

[0041] Specifically, the aforementioned basic data includes various visual screening data sets of the payment identity trusted verification environment in which the current financial payment takes place, multiple environment-related parameters corresponding to the current financial payment, and various target-related information corresponding to the current financial payment, such as... Figure 1 As shown;

[0042] More specifically, the continuous imaging device at the financial payment terminal of the current financial payment is triggered to obtain each frame of real-shot imaging of the payment identity trust verification environment. The human face imaging area occupying the largest area in each frame of real-shot imaging is used as each reference face imaging area. The visual screening data corresponding to each reference face imaging area is used as the visual trust verification content corresponding to the current financial payment. The light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the payment identity trust verification environment where the current financial payment is located are used as multiple environment-related parameters corresponding to the current financial payment. The age information, gender identification, number of recent successful payments, average payment amount, commonly used payment terminal category information, and frequently used payment address encoding information of the target person of the current financial payment trust verification are used as various target association information of the target person corresponding to the current financial payment.

[0043] In this way, the specific data structure design of the above-mentioned basic data provides a data foundation for the accuracy and stability of intelligent verification;

[0044] Technical Process 3: Using the intelligent and reliable verification model with a customized structure designed for the financial payment terminal currently used by the user, based on the multiple basic data introduced in Technical Process 2, the intelligent verification of whether the person to be verified at the front of the financial payment terminal is the target person is completed.

[0045] Obviously, the person to be verified here can be the current user or not.

[0046] Technical Process Four: Based on the intelligent verification results of Technical Process Three, execute automatic protection operations for the current financial payment process, i.e., Figure 1 As shown, determine the corresponding automatic protection strategy;

[0047] Specifically, if the person to be verified at the financial payment terminal that initiated the current financial payment is not the target person, the current financial payment process will be automatically terminated.

[0048] Specifically, when the person to be verified at the financial payment terminal that initiates the current financial payment is the target person, the current financial payment is successfully processed through trusted verification.

[0049] Therefore, through the coordinated operation of the above-mentioned multiple technical processes, the intelligent verification of whether the person to be verified at the financial payment terminal initiating the current financial payment is the target person can be completed. This eliminates the need for a cumbersome and complicated screen recording and identity verification process, and enables reliable verification of the payment identity for each financial payment, providing key information for deciding whether to implement subsequent payment protection.

[0050] The key points of this invention are: the specific data structure design of each visual screening data in the current financial payment identity trust verification environment; the targeted screening of multiple environment-related parameters corresponding to the current financial payment as auxiliary data and the target-related information corresponding to the current financial payment; the design of intelligent trust verification models with different customized structures for different financial payment terminals; and the optimization and improvement of the cumbersome and complex screen recording identity verification process.

[0051] The payment identity trusted verification and protection method and system of the present invention will be specifically described below by way of embodiments.

[0052] Example 1

[0053] Figure 2 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 1 of the present invention.

[0054] like Figure 2 As shown, the payment identity trusted verification and protection method includes the following specific steps:

[0055] Step 201: Use the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment as multiple environment-related parameters for the current financial payment.

[0056] For example, the photometric quantities involved in common lighting systems mainly include luminous intensity, luminance, illuminance, and luminous flux. Luminous intensity generally refers to luminous intensity, which is abbreviated as luminous intensity or luminance in photometry. Luminous intensity is a physical quantity used to express the luminous flux per unit solid angle in a given direction of a light source. The SI unit is the candela, which is one of the seven base units of the International System of Units (SI).

[0057] Step 202: Obtain the target-related information of the target personnel corresponding to the current financial payment;

[0058] For example, obtaining various target-related information of the target personnel corresponding to the current financial payment includes: selecting various information collection units to obtain various target-related information of the target personnel corresponding to the current financial payment respectively;

[0059] Step 203: Trigger the continuous imaging device at the financial payment terminal of the current financial payment to obtain each frame of real-shot imaging of the payment identity trust verification environment. Take the human face imaging area that occupies the largest area in each frame of real-shot imaging as each reference face imaging area, and take each set of visual screening data corresponding to each reference face imaging area as the visual trust verification content corresponding to the current financial payment.

[0060] For example, the continuous imaging device at the financial payment terminal of the current financial payment is activated, that is, the continuous imaging device at the financial payment terminal currently used by the user. The resolution of the continuous imaging device can be 8K, 4K, 2K or high definition.

[0061] As a further example, when the resolution of the continuous imaging device is 8K resolution of 7680×4320, the horizontal resolution of the continuous imaging device, that is, the total number of pixel rows in the image, is 7680, and the vertical resolution of the continuous imaging device, that is, the total number of pixel columns in the image, is 4320.

[0062] Step 204: Use the intelligent trusted verification model to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment.

[0063] Specifically, the result of intelligent verification can be a verification identifier represented by a binary value. The verification identifier uses different specific values ​​to indicate whether the person being verified at the current financial payment terminal is the target person.

[0064] The number of real-shot images in each frame is inversely related to the image resolution of the continuous imaging device.

[0065] Specifically, the inverse relationship between the number of real-shot images in each frame and the image resolution of the continuous imaging device includes: the higher the image resolution of the continuous imaging device, the fewer real-shot images are selected in each frame.

[0066] Among them, the intelligent trustworthy verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device.

[0067] For example, the intelligent trust verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device, including: when the imaging frame rate of the continuous imaging device is 20 frames per second, the selected Hofit neural network is trained 1000 times; when the imaging frame rate of the continuous imaging device is 30 frames per second, the selected Hofit neural network is trained 900 times; when the imaging frame rate of the continuous imaging device is 40 frames per second, the selected Hofit neural network is trained 800 times; when the imaging frame rate of the continuous imaging device is 50 frames per second, the selected Hofit neural network is trained 700 times, and so on.

[0068] The visual screening data corresponding to each reference facial imaging region is used as the visual credibility verification content for the current financial payment. The visual screening data corresponding to each reference facial imaging region includes the grayscale gradient value, imaging depth value, vertical coordinate value and horizontal coordinate value of each pixel in the reference facial imaging region.

[0069] Specifically, the facial contour imaging characteristics of the human face can be used to identify the largest facial imaging area in each real-shot image as a reference facial imaging area in the real-shot image.

[0070] Among them, obtaining the target association information of the target person corresponding to the current financial payment includes: obtaining the age information, gender identifier, number of recent successful payments, average payment amount, commonly used payment terminal category information, and frequently used payment address coding information of the target person whose identity is trusted for the current financial payment as the target association information of the target person corresponding to the current financial payment;

[0071] The parameters used for the current financial payment's trusted identity verification environment, including light intensity, content change, content redundancy, background area ratio, and nearest target distance, are as follows: A continuous imaging device is used at the financial payment terminal to acquire a preview of the trusted identity verification environment in preview acquisition mode; the mean square error of each grayscale value corresponding to each pixel in the preview is used as the content change value of the trusted identity verification environment; deduplication is performed on each grayscale value corresponding to each pixel in the preview to obtain the remaining grayscale values; the difference between the number of remaining grayscale values ​​and the total number of grayscale values ​​is divided by the total number of grayscale values ​​to obtain the content redundancy value of the trusted identity verification environment; and the ratio of the number of background pixels in the preview to the total number of pixels in the preview is used as the background area ratio of the trusted identity verification environment.

[0072] Specifically, the grayscale value of each pixel ranges from 0 to 255;

[0073] In each training iteration of the Hofit Neural Network, the trusted verification result of whether the person being verified at the financial payment terminal of a known past financial payment is the target person is used as a single output of the Hofit Neural Network. The imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the past financial payment, multiple environmental parameters corresponding to the past financial payment, various target association information of the target person corresponding to the past financial payment, and the visual trusted verification content corresponding to the past financial payment are used as multiple inputs of the Hofit Neural Network to complete the training.

[0074] The method of using the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted identity verification environment as multiple environment-related parameters for the current financial payment also includes: using a photometer installed at the financial payment terminal that initiated the current financial payment to measure the light intensity value of the current financial payment's trusted identity verification environment; and using an ultrasonic ranging device installed at the financial payment terminal to measure the nearest target distance of the current financial payment's trusted identity verification environment.

[0075] For example, a photometer installed at the financial payment terminal that initiates the current financial payment is used to measure the light intensity value of the payment identity verification environment where the current financial payment is located. An ultrasonic ranging device installed at the financial payment terminal is used to measure the nearest target distance in the payment identity verification environment where the current financial payment is located. The ultrasonic ranging device installed at the financial payment terminal includes an ultrasonic transmitting unit, an ultrasonic receiving unit, and a microcontroller, and the microcontroller is connected to the ultrasonic transmitting unit and the ultrasonic receiving unit respectively.

[0076] Example 2

[0077] Figure 3 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 2 of the present invention.

[0078] like Figure 3 As shown, with Figure 2 Unlike the previous embodiment, after using the intelligent trusted verification model to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment, that is, after step S204, the method further includes:

[0079] Step S205: If the person to be verified at the financial payment terminal that initiated the current financial payment is not the target person, the payment process of the current financial payment will be automatically terminated.

[0080] For example, when the person to be verified at the financial payment terminal that initiates the current financial payment is not the target person, automatically terminating the current financial payment process includes: the financial payment terminal being a handheld payment terminal, a portable laptop, or a PDA terminal;

[0081] Specifically, when the person to be verified at the financial payment terminal that initiates the current financial payment is the target person, the current financial payment is successfully processed through trusted verification.

[0082] Example 3

[0083] Figure 4 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 3 of the present invention.

[0084] like Figure 4 As shown, with Figure 2 Unlike the previous embodiment, after the continuous imaging device at the financial payment terminal that triggers the current financial payment obtains each frame of the real-world image of the payment identity trust verification environment, and takes the human face imaging area that occupies the largest area in each frame of the real-world image as each reference face imaging area, and takes each set of visual screening data corresponding to each reference face imaging area as the visual trust verification content corresponding to the current financial payment, that is, after step S203, the method further includes:

[0085] Step S206: Perform multiple training operations on the Hofit neural network to obtain a Hofit neural network after multiple training operations and output it as an intelligent trustworthy verification model;

[0086] Specifically, the MATLAB toolbox can be used to test and simulate the modeling process of performing multiple trainings on the Hofit neural network to obtain the Hofit neural network after multiple trainings and outputting it as an intelligent trustworthy verification model.

[0087] The process of training the Hofit neural network multiple times to obtain a trained Hofit neural network and outputting it as an intelligent trustworthy verification model includes: completing the model representation of the intelligent trustworthy verification model through various model parameters of the intelligent trustworthy verification model.

[0088] Example 4

[0089] Figure 5 This is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 4 of the present invention.

[0090] like Figure 5 As shown, with Figure 2 Unlike the previous embodiment, before using the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment as multiple environment-related parameters for the current financial payment, i.e., before step S201, the method further includes:

[0091] Step S207: Set the mode parameters for the continuous imaging mode and preview acquisition mode respectively for the continuous imaging device at the financial payment terminal that initiates the current financial payment;

[0092] Specifically, the mode parameter settings for the continuous imaging mode and preview acquisition mode of the continuous imaging device at the financial payment terminal that initiates the current financial payment include: each mode has different specific values ​​for various mode parameters;

[0093] In the preview acquisition mode, the continuous imaging device captures only one preview frame, while in the continuous imaging mode, the continuous imaging device captures multiple real-world images, and the real-world images and the preview images have the same resolution.

[0094] Example 5

[0095] Figure 6 The following is a flowchart illustrating the steps of a payment identity trusted verification and protection method according to Embodiment 5 of the present invention.

[0096] like Figure 6 As shown, with Figure 2 Unlike the previous embodiment, after using the intelligent trusted verification model to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment, that is, after step S204, the method further includes:

[0097] Step S208: The intelligent verification result of the intelligent trusted verification model is wirelessly transmitted to the remote payment monitoring server via the mobile communication network;

[0098] For example, wirelessly transmitting the intelligent verification results of the intelligent trusted verification model to a remote payment monitoring server via a mobile communication network includes: the mobile communication network being based on a time-division duplex communication link or a frequency-division duplex communication link;

[0099] Among them, the intelligent verification results of the intelligent trusted verification model are wirelessly transmitted to the remote payment monitoring server through the mobile communication network. The remote payment monitoring server is a cloud computing server, a blockchain server, or a big data server.

[0100] Next, the various method embodiments of the present invention will be described in detail.

[0101] In the payment identity trusted verification and protection method according to various method embodiments of the present invention:

[0102] The target information obtained for the current financial payment identity trusted verification target person includes age, gender, number of recent successful payments, average payment amount, commonly used payment terminal type information, and frequently used payment address coding information. This information is used as the target association information for the target person corresponding to the current financial payment. The total number of successful payment orders of the target person within a set time range before the current time is taken as the number of recent successful payments of the target person. The average payment amount of each of the successful payment orders of the target person within the set time range before the current time is taken as the average payment amount of the target person.

[0103] For example, the total number of successfully paid orders within a set time period before the current time of the target person is taken as the target person's most recent successful payment number, and the average of the payment amounts corresponding to each successfully paid order within the set time period before the current time of the target person is taken as the target person's average payment amount, including: the set time period can be 1 year;

[0104] The acquisition of the target personnel's age, gender, number of recent successful payments, average payment amount, commonly used payment terminal type information, and frequently used payment address encoding information as the target personnel corresponding to the current financial payment also includes: taking the financial payment terminal type that appears most frequently among the financial payment terminal types corresponding to each order successfully paid within a set time range before the current time of the target personnel as the commonly used payment terminal type, and taking the category information corresponding to the commonly used payment terminal type as the commonly used payment terminal category information of the target personnel.

[0105] And in the payment identity trusted verification protection method according to various method embodiments of the present invention:

[0106] The intelligent trusted verification model uses the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person. This includes inputting the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment into the intelligent trusted verification model in parallel.

[0107] The parallel input of the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiates the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target personnel corresponding to the current financial payment, and visual credibility verification content corresponding to the current financial payment into the intelligent credibility verification model includes: using programmable logic devices to implement the parallel input of the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiates the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target personnel corresponding to the current financial payment, and visual credibility verification content corresponding to the current financial payment into the intelligent credibility verification model;

[0108] For example, the parallel input of the intelligent trust verification model to the continuous imaging device at the financial payment terminal that initiates the current financial payment, including the imaging frame rate and image resolution, multiple environmental parameters corresponding to the current financial payment, various target association information of the target personnel corresponding to the current financial payment, and the visual trust verification content corresponding to the current financial payment, is as follows: the programmable logic device is an FPGA chip designed with VHDL language.

[0109] Among them, the intelligent trusted verification model is used to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment. It also includes: executing the intelligent trusted verification model to obtain the trusted verification identifier output by the intelligent trusted verification model.

[0110] Furthermore, the trusted verification identifier output by the intelligent trusted verification model is used to indicate whether the person to be verified at the financial payment terminal that initiated the current financial payment is the target person.

[0111] Example 5

[0112] Figure 6 This is a schematic diagram of the payment identity trusted verification and protection system according to Embodiment 5 of the present invention.

[0113] like Figure 6 As shown, the payment identity trusted verification and protection system includes the following components:

[0114] The first data collection agency is used to collect the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment as multiple environment-related parameters corresponding to the current financial payment.

[0115] For example, the photometric quantities involved in common lighting systems mainly include luminous intensity, luminance, illuminance, and luminous flux. Luminous intensity generally refers to luminous intensity, which is abbreviated as luminous intensity or luminance in photometry. Luminous intensity is a physical quantity used to express the luminous flux per unit solid angle in a given direction of a light source. The SI unit is the candela, which is one of the seven base units of the International System of Units (SI).

[0116] The second data collection agency is used to obtain various target-related information of the target personnel corresponding to the current financial payment.

[0117] For example, obtaining various target-related information of the target personnel corresponding to the current financial payment includes: selecting various information collection units to obtain various target-related information of the target personnel corresponding to the current financial payment respectively;

[0118] The third acquisition unit is used to trigger the continuous imaging device at the financial payment terminal of the current financial payment to obtain each frame of real-shot imaging of the payment identity trust verification environment. The human face imaging area that occupies the largest area in each frame of real-shot imaging is used as each reference face imaging area, and each set of visual screening data corresponding to each reference face imaging area is used as the visual trust verification content corresponding to the current financial payment.

[0119] For example, the continuous imaging device at the financial payment terminal of the current financial payment is activated, that is, the continuous imaging device at the financial payment terminal currently used by the user. The resolution of the continuous imaging device can be 8K, 4K, 2K or high definition.

[0120] As a further example, when the resolution of the continuous imaging device is 8K resolution of 7680×4320, the horizontal resolution of the continuous imaging device, that is, the total number of pixel rows in the image, is 7680, and the vertical resolution of the continuous imaging device, that is, the total number of pixel columns in the image, is 4320.

[0121] The trusted verification agency is connected to the first acquisition agency, the second acquisition agency, and the third acquisition agency respectively. It is used to use the intelligent trusted verification model to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment.

[0122] Specifically, the result of intelligent verification can be a verification identifier represented by a binary value. The verification identifier uses different specific values ​​to indicate whether the person being verified at the current financial payment terminal is the target person.

[0123] The number of real-shot images in each frame is inversely related to the image resolution of the continuous imaging device.

[0124] Specifically, the inverse relationship between the number of real-shot images in each frame and the image resolution of the continuous imaging device includes: the higher the image resolution of the continuous imaging device, the fewer real-shot images are selected in each frame.

[0125] Among them, the intelligent trustworthy verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device.

[0126] For example, the intelligent trust verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device, including: when the imaging frame rate of the continuous imaging device is 20 frames per second, the selected Hofit neural network is trained 1000 times; when the imaging frame rate of the continuous imaging device is 30 frames per second, the selected Hofit neural network is trained 900 times; when the imaging frame rate of the continuous imaging device is 40 frames per second, the selected Hofit neural network is trained 800 times; when the imaging frame rate of the continuous imaging device is 50 frames per second, the selected Hofit neural network is trained 700 times, and so on.

[0127] The visual screening data corresponding to each reference facial imaging region is used as the visual credibility verification content for the current financial payment. The visual screening data corresponding to each reference facial imaging region includes the grayscale gradient value, imaging depth value, vertical coordinate value and horizontal coordinate value of each pixel in the reference facial imaging region.

[0128] Specifically, the facial contour imaging characteristics of the human face can be used to identify the largest facial imaging area in each real-shot image as a reference facial imaging area in the real-shot image.

[0129] Among them, obtaining the target association information of the target person corresponding to the current financial payment includes: obtaining the age information, gender identifier, number of recent successful payments, average payment amount, commonly used payment terminal category information, and frequently used payment address coding information of the target person whose identity is trusted for the current financial payment as the target association information of the target person corresponding to the current financial payment;

[0130] The parameters used for the current financial payment's trusted identity verification environment, including light intensity, content change, content redundancy, background area ratio, and nearest target distance, are as follows: A continuous imaging device is used at the financial payment terminal to acquire a preview of the trusted identity verification environment in preview acquisition mode; the mean square error of each grayscale value corresponding to each pixel in the preview is used as the content change value of the trusted identity verification environment; deduplication is performed on each grayscale value corresponding to each pixel in the preview to obtain the remaining grayscale values; the difference between the number of remaining grayscale values ​​and the total number of grayscale values ​​is divided by the total number of grayscale values ​​to obtain the content redundancy value of the trusted identity verification environment; and the ratio of the number of background pixels in the preview to the total number of pixels in the preview is used as the background area ratio of the trusted identity verification environment.

[0131] Specifically, the grayscale value of each pixel ranges from 0 to 255;

[0132] In each training iteration of the Hofit Neural Network, the trusted verification result of whether the person being verified at the financial payment terminal of a known past financial payment is the target person is used as a single output of the Hofit Neural Network. The imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the past financial payment, multiple environmental parameters corresponding to the past financial payment, various target association information of the target person corresponding to the past financial payment, and the visual trusted verification content corresponding to the past financial payment are used as multiple inputs of the Hofit Neural Network to complete the training.

[0133] The method of using the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted identity verification environment as multiple environment-related parameters for the current financial payment also includes: using a photometer installed at the financial payment terminal that initiated the current financial payment to measure the light intensity value of the current financial payment's trusted identity verification environment; and using an ultrasonic ranging device installed at the financial payment terminal to measure the nearest target distance of the current financial payment's trusted identity verification environment.

[0134] For example, a photometer installed at the financial payment terminal that initiates the current financial payment is used to measure the light intensity value of the payment identity verification environment where the current financial payment is located. An ultrasonic ranging device installed at the financial payment terminal is used to measure the nearest target distance in the payment identity verification environment where the current financial payment is located. The ultrasonic ranging device installed at the financial payment terminal includes an ultrasonic transmitting unit, an ultrasonic receiving unit, and a microcontroller, and the microcontroller is connected to the ultrasonic transmitting unit and the ultrasonic receiving unit respectively.

[0135] Furthermore, the present invention may also reference the following technical contents to further demonstrate the outstanding substantial progress of the present invention:

[0136] The acquisition of age information, gender identifier, number of recent successful payments, average payment amount, commonly used payment terminal type information, and frequently used payment address coding information of the target personnel for current financial payment identity verification also includes: the coding information of the most frequently occurring payment address among the payment addresses corresponding to each order successfully paid within a set time range before the current time of the target personnel as the coding information of the target personnel's frequently used payment addresses;

[0137] Among them, the value of the set duration within the set duration range has the same numerical trend as the frequency of financial payment orders appearing at the current financial payment address;

[0138] For example, the value of the set duration within the set duration range has the same numerical trend as the frequency of financial payment orders appearing at the current financial payment address, including: the higher the frequency of financial payment orders appearing at the current financial payment address, the larger the value of the set duration within the corresponding set duration range.

[0139] Although the invention has been described in conjunction with specific embodiments, it will be apparent to those skilled in the art that various modifications and variations can be made to the invention without departing from its scope and spirit. Therefore, it should be understood that the above embodiments are not restrictive in any respect, but rather illustrative.

Claims

1. A payment identity trusted verification and protection method, characterized in that, The method includes: The light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment are used as multiple environment-related parameters for the current financial payment. Obtain the target-related information of the target personnel corresponding to the current financial payment; The continuous imaging device at the financial payment terminal that triggers the current financial payment is used to obtain each frame of real-shot imaging of the payment identity trust verification environment. The human face imaging area that occupies the largest area in each frame of real-shot imaging is used as each reference face imaging area, and each set of visual screening data corresponding to each reference face imaging area is used as the visual trust verification content corresponding to the current financial payment. The intelligent and trustworthy verification model uses the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target-related information of the target personnel corresponding to the current financial payment, and the visual trustworthy verification content corresponding to the current financial payment to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target personnel. The number of real-shot images in each frame is inversely related to the image resolution of the continuous imaging device. Among them, the intelligent trustworthy verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device. Among them, the visual filtering data corresponding to each reference facial imaging area is the grayscale gradient value, imaging depth value, vertical coordinate value and horizontal coordinate value of each pixel in the reference facial imaging area. Among them, the age information, gender identifier, number of recent successful payments, average payment amount, commonly used payment terminal category information, and frequently used payment address coding information of the target personnel for current financial payment identity verification are obtained as various target association information of the target personnel corresponding to the current financial payment. The system employs a continuous imaging device installed at the financial payment terminal to acquire a preview image of the payment identity verification environment in preview acquisition mode. The mean square error of each grayscale value corresponding to each pixel in the preview image is used as the content change value of the payment identity verification environment. Deduplication is performed on each grayscale value corresponding to each pixel in the preview image to obtain the remaining grayscale values. The difference between the number of remaining grayscale values ​​and the total number of grayscale values ​​is divided by the total number of grayscale values ​​to obtain the content redundancy value of the payment identity verification environment. Finally, the proportion of each background pixel in the preview image to the total number of pixels in the preview image is used as the background area ratio of the payment identity verification environment.

2. The payment identity trusted verification and protection method as described in claim 1, characterized in that: In each training iteration of the Hofit Neural Network, the trusted verification result of whether the person being verified at the financial payment terminal of a known past financial payment is the target person is used as a single output of the Hofit Neural Network. The imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the past financial payment, multiple environmental parameters corresponding to the past financial payment, various target association information of the target person corresponding to the past financial payment, and the visual trusted verification content corresponding to the past financial payment are used as multiple inputs of the Hofit Neural Network to complete the training. The parameters used for the current financial payment include light intensity, content change, content redundancy, background area ratio, and nearest target distance within the payment identity verification environment. These parameters include: using a photometer installed at the financial payment terminal that initiated the current financial payment to measure the light intensity of the payment identity verification environment; and using an ultrasonic ranging device installed at the financial payment terminal to measure the nearest target distance within the payment identity verification environment.

3. The payment identity trusted verification and protection method as described in claim 2, characterized in that, After using an intelligent trusted verification model to intelligently verify whether the person to be verified at the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment, the method further includes: If the person being verified at the financial payment terminal that initiated the current financial payment is not the target person, the current financial payment process will be automatically terminated. Specifically, when the person to be verified at the financial payment terminal that initiates the current financial payment is the target person, the current financial payment is successfully processed through trusted verification.

4. The payment identity trusted verification and protection method as described in claim 2, characterized in that, After obtaining real-time images of each frame of the payment identity verification environment from the continuous imaging device at the financial payment terminal that triggers the current financial payment, the method further includes: taking the human face imaging area occupying the largest area in each frame of real-time imaging as each reference face imaging area, and taking the visual screening data corresponding to each reference face imaging area as the visual verification content corresponding to the current financial payment. The Hofitter neural network is trained multiple times to obtain a multi-trained Hofitter neural network, which is then output as an intelligent and trustworthy verification model. The process of training the Hofit neural network multiple times to obtain a trained Hofit neural network and outputting it as an intelligent trustworthy verification model includes: completing the model representation of the intelligent trustworthy verification model through various model parameters of the intelligent trustworthy verification model.

5. The payment identity trusted verification and protection method as described in claim 2, characterized in that, Before using the light intensity, content change, content redundancy, background area ratio, and nearest target distance of the current financial payment's trusted identity verification environment as multiple environment-related parameters, the method further includes: Perform mode parameter settings for continuous imaging mode and preview acquisition mode respectively on the continuous imaging device at the financial payment terminal that initiates the current financial payment; In the preview acquisition mode, the continuous imaging device captures only one preview frame, while in the continuous imaging mode, the continuous imaging device captures multiple real-world images, and the real-world images and the preview images have the same resolution.

6. The payment identity trusted verification and protection method as described in claim 2, characterized in that, After using an intelligent trusted verification model to intelligently verify whether the person to be verified at the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment, the method further includes: The intelligent verification results of the intelligent trusted verification model are wirelessly transmitted to a remote payment monitoring server via a mobile communication network. Among them, the intelligent verification results of the intelligent trusted verification model are wirelessly transmitted to the remote payment monitoring server through the mobile communication network. The remote payment monitoring server is a cloud computing server, a blockchain server, or a big data server.

7. The payment identity trusted verification and protection method as described in claim 2, characterized in that: The target information obtained for the current financial payment identity trusted verification target person includes age, gender, number of recent successful payments, average payment amount, commonly used payment terminal type information, and frequently used payment address coding information. This information is used as the target association information for the target person corresponding to the current financial payment. The total number of successful payment orders of the target person within a set time range before the current time is taken as the number of recent successful payments of the target person. The average payment amount of each of the successful payment orders of the target person within the set time range before the current time is taken as the average payment amount of the target person. The acquisition of the target personnel's age, gender, number of recent successful payments, average payment amount, commonly used payment terminal type information, and frequently used payment address encoding information as the target personnel corresponding to the current financial payment also includes: taking the financial payment terminal type that appears most frequently among the financial payment terminal types corresponding to each order successfully paid within a set time range before the current time of the target personnel as the commonly used payment terminal type, and taking the category information corresponding to the commonly used payment terminal type as the commonly used payment terminal category information of the target personnel.

8. The payment identity trusted verification and protection method as described in claim 2, characterized in that: The intelligent trusted verification model uses the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person. This includes inputting the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment into the intelligent trusted verification model in parallel. The parallel input of the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiates the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target personnel corresponding to the current financial payment, and visual credibility verification content corresponding to the current financial payment into the intelligent credibility verification model includes: using programmable logic devices to implement the parallel input of the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiates the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target personnel corresponding to the current financial payment, and visual credibility verification content corresponding to the current financial payment into the intelligent credibility verification model; Among them, the intelligent trusted verification model is used to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment. It also includes: executing the intelligent trusted verification model to obtain the trusted verification identifier output by the intelligent trusted verification model. Among them, the trusted verification identifier output by the intelligent trusted verification model is used to indicate whether the person to be verified at the financial payment terminal that initiates the current financial payment is the target person.

9. A payment identity trusted verification and protection system, characterized in that, The system includes: The first data collection agency is used to collect the light intensity value, content change value, content redundancy value, background area ratio, and nearest target distance of the current financial payment's trusted verification environment as multiple environment-related parameters corresponding to the current financial payment. The second data collection agency is used to obtain various target-related information of the target personnel corresponding to the current financial payment. The third acquisition unit is used to trigger the continuous imaging device at the financial payment terminal of the current financial payment to obtain each frame of real-shot imaging of the payment identity trust verification environment. The human face imaging area that occupies the largest area in each frame of real-shot imaging is used as each reference face imaging area, and each set of visual screening data corresponding to each reference face imaging area is used as the visual trust verification content corresponding to the current financial payment. The trusted verification agency is connected to the first acquisition agency, the second acquisition agency, and the third acquisition agency respectively. It is used to use the intelligent trusted verification model to intelligently verify whether the person to be verified in front of the financial payment terminal that initiated the current financial payment is the target person, based on the imaging frame rate and image resolution of the continuous imaging device at the financial payment terminal that initiated the current financial payment, multiple environmental parameters corresponding to the current financial payment, various target association information of the target person corresponding to the current financial payment, and the visual trusted verification content corresponding to the current financial payment. The number of real-shot images in each frame is inversely related to the image resolution of the continuous imaging device. Among them, the intelligent trustworthy verification model is a Hofit neural network that has been trained multiple times, and the number of training times is inversely related to the imaging frame rate of the continuous imaging device. Among them, the visual filtering data corresponding to each reference facial imaging area is the grayscale gradient value, imaging depth value, vertical coordinate value and horizontal coordinate value of each pixel in the reference facial imaging area. Among them, the age information, gender identifier, number of recent successful payments, average payment amount, commonly used payment terminal category information, and frequently used payment address coding information of the target personnel for current financial payment identity verification are obtained as various target association information of the target personnel corresponding to the current financial payment. The system employs a continuous imaging device installed at the financial payment terminal to acquire a preview image of the payment identity verification environment in preview acquisition mode. The mean square error of each grayscale value corresponding to each pixel in the preview image is used as the content change value of the payment identity verification environment. Deduplication is performed on each grayscale value corresponding to each pixel in the preview image to obtain the remaining grayscale values. The difference between the number of remaining grayscale values ​​and the total number of grayscale values ​​is divided by the total number of grayscale values ​​to obtain the content redundancy value of the payment identity verification environment. Finally, the proportion of each background pixel in the preview image to the total number of pixels in the preview image is used as the background area ratio of the payment identity verification environment.

Images