Method and apparatus for updating face image data, device, and storage medium

Abstract

The application is a face image data updating method, device, equipment and storage medium, and relates to the technical field of face recognition. The method comprises the following steps: receiving first face image data; obtaining a first comparison result between the first face image data and second face image data; updating face recognition statistical information of a first user account in response to the first comparison result meeting statistical updating conditions; sending an image resetting entry to the first terminal in response to the face recognition statistical information meeting image resetting conditions; receiving third face image data sent by the first terminal based on the image resetting entry; and updating the second face image data in the face template database based on the third face image data. According to the above scheme, the background server can send the image resetting entry to the terminal, so that the terminal uploads the updated face image data, and the face recognition error caused by the old base map in the face recognition process is avoided as much as possible, and the accuracy of face recognition is improved.

Description

Technical Field

[0001] This application relates to the field of face recognition, and in particular to a method, apparatus, device and storage medium for updating face image data. Background Technology

[0002] Facial recognition is a series of related technologies that use cameras or webcams to capture images or video streams containing human faces, automatically detect human faces in the images, and then perform facial recognition on the detected faces.

[0003] In current technologies, facial recognition requires acquiring a facial image through a terminal device with an image acquisition component and saving the image to a facial template database. When facial recognition is needed, the terminal acquires the facial image to be recognized and compares it with the facial images in the facial template database to determine the identity information of the facial image to be recognized.

[0004] The above technical solutions may result in facial recognition errors due to changes in the user's face (such as changes in weight), affecting the accuracy of facial recognition. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for updating facial image data, which can improve the accuracy of facial recognition. The technical solution is as follows:

[0006] On one hand, a method for updating facial image data is provided, the method being executed by a backend server, the backend server containing a facial template database for facial recognition, the method comprising:

[0007] Receive the first face image data;

[0008] Obtain a first comparison result between the first face image data and the second face image data; the second face image data is the face image data in the face template database that has the highest similarity to the first face image data;

[0009] In response to the first comparison result satisfying the statistical update condition, the facial recognition statistics of the first user account are updated; the first user account is the user account corresponding to the second facial image data.

[0010] In response to the face recognition statistics meeting the image reset conditions, an image reset entry is sent to the first terminal; the first terminal is a terminal logged into the first user account;

[0011] Receive third face image data sent by the first terminal based on the image reset entry;

[0012] Based on the third face image data, the second face image data is updated in the face template database.

[0013] On another front, a method for updating facial image data is provided, the method being executed by a first terminal, which is a terminal logged into a first user account, the method comprising:

[0014] An image reset entry is displayed; the image reset entry is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions; the face recognition statistics are updated by the backend server when it receives the first face image data and the first comparison result between the first face image data and the second face image data meets the statistical update conditions; the second face image data is the face image data corresponding to the first user account in the face template database, and the second face image data is the face image data in the face template database with the highest similarity to the first face image data;

[0015] In response to receiving a specified operation on the image reset entry, third-party face image data is acquired;

[0016] The third face image data is sent to the backend server so that the backend server updates the second face image data in the face template database based on the third face image data.

[0017] Furthermore, a facial image data updating device is provided. This device is used in a backend server, which contains a facial template database for facial recognition. The device includes:

[0018] The first image receiving module is used to receive the first face image data;

[0019] The first comparison result acquisition module is used to acquire a first comparison result between the first face image data and the second face image data; the second face image data is the face image data in the face template database that has the highest similarity to the first face image data;

[0020] The first information update module is used to update the face recognition statistics of the first user account in response to the first comparison result meeting the statistical update conditions; the first user account is the user account corresponding to the second face image data.

[0021] The first entry sending module is used to send an image reset entry to the first terminal in response to the face recognition statistics meeting the image reset conditions; the first terminal is a terminal logged into the first user account;

[0022] The second image receiving module is used to receive third face image data sent by the first terminal based on the image reset entry;

[0023] The image data update module is used to update the second face image data in the face template database based on the third face image data.

[0024] In one possible implementation, the first information update module is further configured to:

[0025] In response to the first comparison result satisfying the statistical update condition, the statistical count corresponding to the first user account is updated, and the statistical count is used to indicate the number of times that the comparison results corresponding to the second face image data have satisfied the statistical update condition.

[0026] In one possible implementation, the first ingress sending module includes:

[0027] The account status setting unit is used to set the first user account to a pending update status in response to the face recognition statistics meeting the data reset conditions.

[0028] The first entry sending unit is used to send the image reset entry to the first terminal in response to detecting that the first user account is in a pending update state at a specified time.

[0029] In one possible implementation, the image reset condition includes at least one of the following:

[0030] In the comparison results corresponding to the second face image data, the number of times the statistical update condition is met consecutively reaches the first threshold.

[0031] In the comparison results corresponding to the second face image data, the cumulative number of times the statistical update condition is met reaches the second threshold.

[0032] In the comparison results corresponding to the second face image data, the ratio of the first number of times the statistical update condition is met reaches the third threshold.

[0033] In one possible implementation, the statistical update condition includes at least one of a first condition and a second condition;

[0034] The first condition includes: the first comparison result indicates that the first face image data and the second face image data do not match;

[0035] The second condition includes: the first comparison result indicates that the first face image data matches the second face image data, and the similarity between the first face image data and the second face image data is less than the first similarity threshold.

[0036] In one possible implementation, the first ingress sending module further includes:

[0037] The frequency ratio acquisition unit is used to, in response to the face recognition statistics information satisfying the image reset condition, acquire the second frequency ratio between the number of times the first condition is satisfied and the number of times the second condition is satisfied in the previous comparison results corresponding to the second face image data;

[0038] The second entry sending unit is used to send the image reset entry to the first terminal based on the second count ratio; the reminder strength of the first prompt information in the image reset entry is positively correlated with the second count ratio; the prompt information is used to prompt the user to update the face image data.

[0039] In one possible implementation, the first face image data is face image data collected by the second terminal based on a face recognition interface, and the device further includes:

[0040] In response to the face recognition statistics meeting the image reset conditions and the face recognition interface not being closed on the second terminal, a second prompt message is sent to the second terminal; the second prompt message is used to prompt the user to view the image reset entry on the first terminal.

[0041] Furthermore, a facial image data updating device is provided. This device is used on a first terminal, which is a terminal logged into a first user account. The device includes:

[0042] A reset entry display module is used to display the image reset entry; the image reset entry is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions; the face recognition statistics are updated by the backend server when it receives the first face image data and the first comparison result between the first face image data and the second face image data meets the statistical update conditions; the second face image data is the face image data corresponding to the first user account in the face template database, and the second face image data is the face image data in the face template database with the highest similarity to the first face image data;

[0043] A face image acquisition module is used to acquire third face image data in response to receiving a specified operation on the image reset entry;

[0044] A face image sending module is used to send the third face image data to the backend server, so that the backend server updates the second face image data in the face template database based on the third face image data.

[0045] In one possible implementation, the face image acquisition module includes:

[0046] An image component invocation unit is used to invoke the image acquisition component in the first terminal in response to receiving a specified operation on the image reset entry.

[0047] A face image acquisition unit is used to acquire the third face image data acquired by the image acquisition component.

[0048] In one possible implementation, the device further includes:

[0049] The facial recognition interface display module is used to display the facial recognition interface in response to a received facial recognition operation.

[0050] The first face image data acquisition module is used to call the image acquisition component in the first terminal to acquire the first face image data;

[0051] The first face image data sending module is used to send the first face image data to the backend server;

[0052] The face recognition result display module is used to display the face recognition result of the backend server on the first face image data in the face recognition interface.

[0053] In one possible implementation, the device further includes:

[0054] The prompt information display module is used to display a second prompt information in the face recognition interface. The second prompt information is used to prompt the user to view the image reset entry in the first terminal. The second prompt information is sent by the backend server when the face recognition statistics of the first user account meet the image reset conditions.

[0055] In another aspect, a computer device is provided, the computer device comprising a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, the code set or instruction set being loaded and executed by the processor to implement the above-described face image data update method.

[0056] In another aspect, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, code set, or instruction set is stored therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the above-described face image data update method.

[0057] In another aspect, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned face image data update method.

[0058] The beneficial effects of the technical solutions provided in this application include at least the following:

[0059] During the face recognition process, the backend server maintains statistical information based on the face recognition records of the first user account. When the backend server detects that the face recognition records of the first user account meet the image reset conditions based on the statistical information, it can send an image reset entry to the terminal corresponding to the first user account so that the terminal can upload new face image data and update the face data in the face template database. This minimizes the occurrence of face recognition errors caused by outdated base images and improves the accuracy of face recognition. Attached Figure Description

[0060] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0061] Figure 1 A schematic diagram of a computer system provided in an exemplary embodiment of this application is shown.

[0062] Figure 2 This is a flowchart illustrating a face image data update method according to an exemplary embodiment.

[0063] Figure 3 This is a flowchart illustrating a face image data update method according to an exemplary embodiment.

[0064] Figure 4 This is a flowchart illustrating a method for updating face image data according to an exemplary embodiment.

[0065] Figure 5 It shows Figure 4 The illustrated embodiment is a schematic diagram of the intensity of a prompting message.

[0066] Figure 6 It shows Figure 4 The illustrated embodiment is a schematic diagram of a prompt message.

[0067] Figure 7 It shows Figure 4 The illustrated embodiment is a schematic diagram of a prompt message.

[0068] Figure 8 It shows Figure 4 The illustrated embodiment is a schematic diagram of the overall architecture of a face image data update system.

[0069] Figure 9 This is a schematic diagram illustrating a face image data update method according to an exemplary embodiment.

[0070] Figure 10 This is a structural block diagram of a face image data updating device according to an exemplary embodiment.

[0071] Figure 11 This is a structural block diagram of a face image data updating device according to an exemplary embodiment.

[0072] Figure 12 This is a schematic diagram of the structure of a computer device according to an exemplary embodiment. Detailed Implementation

[0073] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0074] First, the terms used in the embodiments of this application will be introduced.

[0075] 1) Artificial Intelligence (AI)

[0076] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0077] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0078] 2) Computer Vision (CV) technology

[0079] Computer vision is the science that studies how to enable machines to "see." More specifically, it refers to machine vision, which uses cameras and computers to replace human eyes in recognizing and measuring targets, and then performs image processing to create images more suitable for human observation or transmission to instruments. As a scientific discipline, computer vision studies related theories and technologies, attempting to build artificial intelligence systems capable of extracting information from images or multidimensional data. Computer vision technologies typically include image processing, image recognition, image semantic understanding, image retrieval, OCR (optical character recognition), video processing, video semantic understanding, video content / behavior recognition, 3D object reconstruction, 3D technology, virtual reality, augmented reality, and map building.

[0080] Please refer to Figure 1 The diagram illustrates a computer system provided in an exemplary embodiment of this application. The computer system 200 includes a terminal 110 and a server 120, wherein the terminal 110 and the server 120 communicate via a communication network. Optionally, the communication network can be a wired network or a wireless network, and the communication network can be at least one of a local area network (LAN), a metropolitan area network (MAN), and a wide area network (WAN).

[0081] The terminal 110 is equipped with an application with facial recognition function. This application can be a payment application, a communication application, other applications that require identity verification, or an artificial intelligence (AI) application that needs to call facial recognition function. This application embodiment does not limit this.

[0082] Optionally, terminal 110 may have a data transmission interface for receiving facial image data with input from other computer devices.

[0083] Optionally, the terminal 110 may also include an image acquisition device, which can be used to directly acquire facial image data, or the image acquisition device can be used to acquire video data within a specified time period, and acquire facial image data based on the video data within the specified time period.

[0084] Optionally, the terminal 110 can be a mobile terminal such as a smartphone, tablet computer, or laptop computer, or a terminal such as a desktop computer or projector computer, or a smart terminal with data processing components and image acquisition components. This application embodiment does not limit this.

[0085] Server 120 can be implemented as a single server or as a server cluster consisting of a group of servers. It can be a physical server or a cloud server. In one possible implementation, server 120 is the backend server for the application in terminal 110.

[0086] In one possible implementation of this embodiment, terminal 110 acquires target face image data through an image acquisition component and uploads the target face image data to a backend server 120. The backend server 120 compares the target face image data with each face image data in the face template database to obtain the identity information corresponding to the target face image data and returns the identity information corresponding to the target face image data to terminal 110, thereby realizing a complete face recognition process.

[0087] Figure 2 This is a flowchart illustrating a method for updating face image data according to an exemplary embodiment. The method can be executed by a computer device, wherein the computer device can be the one described above. Figure 1 The backend server 120 in the illustrated embodiment. For example... Figure 2 As shown, the process of this face image data update method may include the following steps:

[0088] Step 201: Receive the first face image data.

[0089] Step 202: Obtain the first comparison result between the first face image data and the second face image data; the second face image data is the face image data in the face template database that has the highest similarity to the first face image data.

[0090] Step 203: In response to the first comparison result satisfying the statistical update conditions, update the face recognition statistics of the first user account; the first user account is the user account corresponding to the second face image data.

[0091] Step 204: In response to the face recognition statistics meeting the image reset conditions, an image reset entry is sent to the first terminal; the first terminal is the terminal logged into the first user account.

[0092] Step 205: Receive the third face image data sent by the first terminal based on the image reset entry.

[0093] Step 206: Based on the third face image data, update the second face image data in the face template database.

[0094] In summary, the solution shown in this application embodiment allows the backend server to maintain statistical information based on the face recognition records of the first user account during the face recognition process. When the backend server detects that the face recognition records of the first user account meet the image reset conditions based on the statistical information, it can send an image reset entry to the terminal corresponding to the first user account so that the terminal can upload new face image data and update the face data in the face template database. This minimizes the occurrence of face recognition errors caused by outdated background images during the face recognition process and improves the accuracy of face recognition.

[0095] Figure 3 This is a flowchart illustrating a method for updating face image data according to an exemplary embodiment. The method can be executed by a computer device, wherein the computer device can be the one described above. Figure 1 Terminal 110 in the illustrated embodiment. (As shown) Figure 3 As shown, the process of this face image data update method may include the following steps:

[0096] Step 301: Display the image reset entry.

[0097] Specifically, the image reset entry is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions; the face recognition statistics are updated by the backend server when it receives the first face image data and the first comparison result between the first face image data and the second face image data meets the statistical update conditions; the second face image data is the face image data corresponding to the first user account in the face template database, and the second face image data is the face image data in the face template database with the highest similarity to the first face image data.

[0098] Step 302: In response to receiving a specified operation to reset the image entry, obtain third-person face image data.

[0099] Step 303: The third face image data is sent to the backend server so that the backend server can update the second face image data in the face template database based on the third face image data.

[0100] In summary, the solution shown in this application embodiment allows the backend server to maintain statistical information based on the face recognition records of the first user account during the face recognition process. When the backend server detects that the face recognition records of the first user account meet the image reset conditions based on the statistical information, it can send an image reset entry to the terminal corresponding to the first user account so that the terminal can upload new face image data and update the face data in the face template database. This minimizes the occurrence of face recognition errors caused by outdated background images during the face recognition process and improves the accuracy of face recognition.

[0101] Figure 4 This is a flowchart illustrating a method for updating face image data according to an exemplary embodiment. The method described above... Figure 1 In the illustrated embodiment, server 120 and terminal 110 work together. For example... Figure 4 As shown, the process of this face image data update method may include the following steps:

[0102] Step 401: The backend server receives the first face image data.

[0103] In one possible implementation, the first face image data is the face image data to be identified collected by a second terminal corresponding to the backend server.

[0104] After the second terminal corresponding to the backend server collects the facial image data of the target object to be identified, it needs to upload the facial image data to the corresponding backend server so that the backend server can perform facial recognition on the facial image data to be identified and return the recognition result to the second terminal so that the second terminal can perform the corresponding operation process based on the recognition result.

[0105] In one possible implementation, the second terminal sends the first face image data to the backend server, and the backend server receives the first face image data sent by the second terminal.

[0106] The second terminal is the terminal device corresponding to the client that logged into the first user account. When the backend server receives the first face image data sent by the second terminal, it can recognize the first face image data and return the recognition result directly to the second terminal through the client of the first user account.

[0107] In one possible implementation, the second terminal is a mobile terminal, which hosts a client application corresponding to the backend server. The client application is logged into the first user account. When the user needs to perform facial recognition to implement a function in the client application, the client application calls the image acquisition component (i.e., camera device) in the mobile terminal to obtain the user's facial image data to be recognized (i.e., the first facial image data) and uploads the facial image data to be recognized to the backend server. The backend server performs facial recognition on the uploaded facial image based on a facial template database, obtains the facial recognition result, and returns it to the mobile terminal so that the mobile terminal can implement a function (such as a payment function) in the client application.

[0108] In one possible implementation, the second terminal is an IoT (Internet of Things) terminal. This IoT terminal hosts an application that requires facial recognition (e.g., a payment application). When a user needs to perform facial recognition to achieve a certain function within the application, the application calls the image acquisition component (i.e., the camera device) in the IoT terminal to acquire the user's facial image data to be recognized (i.e., the first facial image data) and uploads it to a backend server. The backend server performs facial recognition on the uploaded facial image based on a facial template database, obtains the facial recognition result, and returns it to the IoT terminal so that the IoT terminal can perform a certain function within the application (e.g., a payment function).

[0109] In one possible implementation, in response to receiving the first recognition operation, the image acquisition component in the second terminal is invoked to acquire the first face image data; the first face image data is then sent to the backend server.

[0110] The first recognition operation is an operation performed by the user corresponding to the second terminal on the recognition trigger control on the second terminal. When the recognition trigger control is displayed on the client interface of the second terminal, the user can perform the first recognition operation on the recognition trigger control. In response to receiving the first recognition operation, the terminal calls the image acquisition component in the second terminal to obtain the first face image data.

[0111] In one possible implementation, in response to receiving a first recognition operation, the image acquisition component in the second terminal is invoked to acquire at least two acquired images; based on the acquisition scores corresponding to each of the at least two acquired images, the first face image data is determined; the acquisition scores are used to indicate the acquisition quality of the acquired images.

[0112] When the first recognition operation is received, the second terminal can acquire at least two images through the image acquisition component in the second terminal. The acquired images are facial images obtained by the second terminal from the user's facial information. The second terminal selects the image with the highest quality from the at least two acquired images as the first facial image data and uploads it to the backend server according to the quality of each acquired image.

[0113] In one possible implementation, in response to receiving a first recognition operation, the image acquisition component in the second terminal is invoked to acquire a first video segment; based on each video frame in the first video segment, at least two acquired images are acquired.

[0114] When the first recognition operation is received, the second terminal can also call the image acquisition component in the second terminal to obtain the first video segment, and then determine at least two acquired images obtained by the second terminal from the user's facial information in each video frame of the first video segment.

[0115] In one possible implementation, the acquisition score corresponding to each of the at least two acquired images is determined based on the image parameters corresponding to each of the at least two acquired images.

[0116] The image parameters include at least one of the following: face size, face angle, image contrast, image brightness, and image sharpness.

[0117] When the image acquisition component in the second terminal is invoked and at least two acquired images are obtained, the acquisition parameters of each acquired image can be determined based on at least one of the following: face size, face angle, image contrast, image brightness, and image clarity. The face image with the best overall evaluation among the acquired images can then be selected as the first face image data.

[0118] In one possible implementation, in response to receiving a first recognition operation, the image acquisition component in the second terminal is invoked to acquire video images in real time. In response to receiving a first video frame with a quality score greater than a quality threshold, the acquisition of video images is stopped, and the image corresponding to the first video frame is identified as the first face image data.

[0119] Upon receiving the first recognition operation, the image acquisition component in the second terminal can be directly invoked to acquire video images in real time and perform quality analysis on the acquired video frames in real time to obtain the quality score of each video frame. When the first video frame with a quality score greater than the quality threshold is received, it indicates that the first video frame meets the quality requirements for uploading to the server for face recognition. Therefore, the first video frame can be directly identified as the first face image data, and the invocation of the image acquisition component can be stopped.

[0120] In one possible implementation, the first face image data includes at least one of RGB (Red Green Blue) image data, depth image data, and infrared image data.

[0121] The image acquisition component in the second terminal may include at least one of an RGB image acquisition component, a depth image acquisition component, and an infrared image acquisition component; the first face image data can be obtained by calling at least one of the RGB image acquisition component, the depth image acquisition component, and the infrared image acquisition component in the second terminal.

[0122] In one possible implementation, in response to the second terminal sending the first face image data to the backend server, a waiting screen is displayed on the second terminal.

[0123] That is, after the second terminal sends the first face image data to the backend server, the second terminal needs to receive the face recognition result of the first face image data from the backend server before it can perform subsequent operations based on the face recognition result. At this time, the terminal interface of the second terminal can display a waiting screen, such as "loading" or "recognizing", to prompt the user that the second terminal and the corresponding backend server are performing the corresponding face recognition operation.

[0124] Step 402: The backend server obtains the first comparison result between the first face image data and the second face image data.

[0125] The second face image data is the face image data in the face template database that has the highest similarity to the first face image data.

[0126] In one possible implementation, feature extraction is performed on the first face image data to obtain first feature data corresponding to the first face image data; based on the first feature data, the second face image data and the first user account corresponding to the second face image data are determined in the face template database.

[0127] In one possible implementation, the face template database also includes feature data corresponding to each face image data. Therefore, based on the similarity between the first face image data and the feature data corresponding to each of the face image data, the face image data whose feature data is most similar to the first feature data among the face image data can be determined as the first face image data.

[0128] In one possible implementation, the feature data can be the feature vectors corresponding to each face image data. In this case, the face image data closest to the first face image data can be determined based on the vector distance between the first face image data and the feature vectors of each face image data, and this face image data can be identified as the second face image data.

[0129] In one possible implementation, the vector distance can be at least one of Euclidean distance and cosine distance.

[0130] In one possible implementation, first feature data corresponding to the first face image data is obtained; in response to the face template database, if the similarity between the first feature data and each feature data is greater than a similarity threshold, the face image data corresponding to the feature data with the highest similarity to the first feature data is determined as the second face image data.

[0131] When face recognition is required on the first face image data, it is necessary to first search for the template face image corresponding to the first face image data in the face template database. That is, to search for the template face image that the user corresponding to the first face image data has uploaded in advance in the face template database. At this time, the backend server can compare the first feature data corresponding to the first face image data with the feature data of each face image data in the face template database to determine the second face image data and the first user account corresponding to the second face image data in the face template database, so as to determine the identity information of the first face image data.

[0132] In one possible implementation, in response to the face template database, if the maximum similarity between the first feature data and each feature data is less than a similarity threshold, an identification error message is generated and the identification error message is returned to the terminal corresponding to the first face image data.

[0133] When the maximum similarity between the first feature data and each feature data in the face template database is less than the similarity threshold, it means that no face image data similar to the first face image data can be found in the face template database. Therefore, the identity of the first face image data cannot be recognized normally, and a recognition error message is generated and returned to the terminal corresponding to the first face image data. After receiving the recognition error message, the terminal displays the recognition error message on the terminal's display interface to prompt the user that the uploaded face information cannot be recognized normally.

[0134] Step 403: The backend server responds to the fact that the first comparison result meets the statistical update conditions and updates the face recognition statistics of the first user account.

[0135] Among them, the facial recognition statistics are used to indicate the historical information records of facial recognition corresponding to the first user account.

[0136] For example, the facial recognition statistics may include records of failed facial recognition for the first user account, or the facial recognition statistics may also include records of risky facial recognition for the first user account. Specifically, the failed recognition records indicate where the backend server failed to compare the facial template (i.e., the second facial image data) corresponding to the first user account with the facial image data uploaded to the backend server; the risky recognition records indicate where the backend server successfully compared the facial template corresponding to the first user account with the facial image data uploaded to the backend server, and the similarity is less than a security threshold.

[0137] In one possible implementation, in response to the first comparison result satisfying the statistical update condition, the statistical count corresponding to the first user account is updated, and the statistical count is used to indicate the number of times that the comparison results corresponding to the second face image data have satisfied the statistical update condition.

[0138] In one possible implementation, the statistical update condition includes at least one of the first condition and the second condition;

[0139] The first condition includes: the first comparison result indicates that the first face image data does not match the second face image data;

[0140] The second condition includes: the first comparison result indicates that the first face image data matches the second face image data, and the similarity between the first face image data and the second face image data is less than the first similarity threshold.

[0141] In one possible implementation, the number of statistics includes at least one of the number of failure statistics and the number of danger statistics;

[0142] When the statistical update condition includes the first condition, that is, when the first comparison result indicates that the first face image data does not match the second face image data, it means that the face recognition result corresponding to the first user account is a failure, and the failure count corresponding to the first user account is updated (i.e. incremented by one).

[0143] When the statistical update condition includes the second condition, that is, when the first comparison result indicates that the first face image data matches the second face image data, and the similarity between the first face image data and the second face image data is less than the first similarity threshold (i.e., the safety threshold), although the first comparison result indicates that the first face image data and the second face image data are a match (e.g., the similarity is 95%), the similarity between the first face image data and the second face image data is less than the first similarity threshold (e.g., the similarity is 99%). At this time, the match between the first face image data and the second face image data is an unsafe match, so the number of dangerous statistics corresponding to the first user account is updated (i.e., incremented by one).

[0144] Step 404: In response to the face recognition statistics meeting the image reset conditions, the backend server sends an image reset entry to the first terminal.

[0145] In one possible implementation, in response to the face recognition statistics corresponding to the first user account meeting the image reset conditions, the backend server sends an image reset entry to the first terminal corresponding to the first user account based on the first user account.

[0146] In one possible implementation, in response to the face recognition statistics meeting the image reset conditions, the first user account is set to a pending update state; in response to detecting that the first user account is in a pending update state at a specified time, the image reset entry is sent to the first terminal.

[0147] When the face recognition statistics meet the image reset conditions, the first user account can be set to the pending update state. When the backend server detects that the first user account is in the pending update state during the specified time, it can send the image reset entry to the first terminal based on the first user account.

[0148] In one possible implementation, the backend server checks each user account stored in the backend server at a specified period, and sends an image reset entry to the terminal of the user account that needs to update the face template data, so as to realize the timed update of the face template data of each user account.

[0149] In one possible implementation, the image reset condition includes at least one of the following:

[0150] In the comparison results corresponding to the second face image data, the number of times the statistical update condition is met consecutively reaches the first threshold.

[0151] In the comparison results corresponding to the second face image data, the cumulative number of times the statistical update condition is met reaches the second threshold.

[0152] In the comparison results corresponding to the second face image data, the cumulative ratio of the first number of times that the statistical update condition is met reaches the third threshold.

[0153] When the number of consecutive comparison results corresponding to the second face image data, i.e. the face recognition results corresponding to the first user account, reaches the first threshold, for example, when there are 5 consecutive comparison failures in the comparison results corresponding to the second face image data (the first threshold is 4), it can be considered that there may be a problem with the second face image data. Therefore, it can be considered that the number of face recognition statistics for the first user account meets the image reset condition, and the first user account is set to the pending update state. When the backend server detects that the first user account is in the pending update state, it sends the image reset entry to the first terminal.

[0154] When the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data, i.e. the face recognition results corresponding to the first user account, reaches the first threshold, for example, in the comparison results corresponding to the second person Alin's image data, there are a cumulative 11 comparison failure records (i.e., 11 times the statistical update condition is met and the face recognition statistics are updated), the second threshold is 10 times. This means that the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data reaches the second threshold. At this time, it can be considered that the second face image data has a large number of errors, and the face template data of the first user account can be updated. Therefore, the first user account is set to the pending update state. When the backend server detects that the first user account is in the pending update state, it sends the image reset entry to the first terminal.

[0155] In one possible implementation, the first ratio is the ratio of the cumulative number of times the statistical update condition is met in all comparison results corresponding to the second face image data to the total number of comparisons corresponding to the second face image data. For example, if the number of comparisons corresponding to the second face image data is 100, and the number of times the statistical update condition is met is 5, then the first ratio is 5%.

[0156] In one possible implementation, in response to the update of the second face image data corresponding to the first user account, the number of comparisons corresponding to the second face image data and the number of times the statistical update conditions are met are set to zero.

[0157] In one possible implementation, in response to the face recognition statistics meeting the image reset condition, a second ratio is obtained between the number of times the first condition is met and the number of times the second condition is met in the previous comparison results corresponding to the second face image data; based on the second ratio, the image reset entry is sent to the first terminal; the reminder strength of the first prompt information in the image reset entry is positively correlated with the second ratio; the prompt information is used to prompt the user to update the face image data.

[0158] When the statistical update conditions include a first condition and a second condition, the number of times the first condition is met is the number of failure statistics, and the number of times the second condition is met is the number of dangerous statistics. The ratio between the number of failure statistics and the number of dangerous statistics is determined as the second ratio. When the second ratio is high, it means that the proportion of failure statistics in the total number of statistics is high, that is, there are more recognition failures. In this case, a stronger first prompt message can be used. When the second ratio is low, it means that the proportion of failure statistics in the total number of statistics is low, that is, there are fewer recognition failures, but more insecure recognitions (or dangerous recognitions). In this case, although normal face recognition can be achieved, the accuracy of face recognition is low, and it is also necessary to update the face data corresponding to the first user account in the face template database. Therefore, a weaker first prompt message can be used.

[0159] Please refer to Figure 5 This illustration shows a schematic diagram of the intensity of a prompting message according to an embodiment of this application. For example... Figure 5 As shown, when the ratio of the second count is small, the alerting effect of the first prompt is also weak. For example, when the ratio of the second count is less than the first threshold, the first prompt is a weak alert. Figure 5 As shown in section 501, the terminal's display interface shows a message stating "Face recognition may fail; face identity information needs to be updated," prompting the user to update their face image data. When the second count ratio is larger, the first prompt becomes more forceful; for example, when the second count ratio exceeds a first threshold, the first prompt is a stronger warning. Figure 5 As shown in section 502, the terminal's display interface shows the first prompt message: "Face recognition carries a great risk. Please update your facial identity information!" This first prompt message can achieve a high level of alertness through text content, text size (or bolding), and display color (e.g., highlighted in red, not shown in the figure).

[0160] In one possible implementation, the first face image data is face image data collected by the second terminal based on the face recognition interface. In response to the face recognition statistics meeting the image reset conditions and the face recognition interface not being closed on the second terminal, a second prompt message is sent to the second terminal. The second prompt message is used to prompt the user to view the image reset entry on the first terminal.

[0161] The first facial image data is facial image data collected by the second terminal based on the facial recognition interface. When the facial recognition statistics meet the image reset conditions, and the facial recognition interface is not closed in the second terminal, the backend server sends a second prompt message to the second terminal to prompt the user that the backend server has sent the image reset entry to the first terminal. The user can view the image reset entry in the first terminal to complete the reset of the facial image data corresponding to the first user account.

[0162] Step 405: The first terminal displays the image reset entry.

[0163] In one possible implementation, in response to receiving an image reset entry sent by a backend server, a push message for the image reset entry is displayed in the notification bar of the terminal; in response to receiving a trigger operation of the push message for the image reset entry, the image reset entry is displayed on the display interface of the first terminal.

[0164] In one possible implementation, in response to receiving a face recognition operation, a face recognition interface is displayed; the image acquisition component in the first terminal is invoked to acquire the first face image data; the first face image data is sent to the backend server; and the face recognition result of the backend server on the first face image data is displayed in the face recognition interface.

[0165] Step 406: The first terminal responds to receiving the specified operation for resetting the image entry and obtains the third face image data.

[0166] In one possible implementation, in response to receiving a specified operation on the image reset entry, the image acquisition component in the first terminal is invoked; and the third face image data acquired by the image acquisition component is obtained.

[0167] In one possible implementation, in response to receiving a specified operation to reset the image entry, the image acquisition component in the first terminal is invoked to acquire at least two acquired images; based on the acquisition scores corresponding to each of the at least two acquired images, the third face image data is determined from the at least two acquired images.

[0168] When a specified operation to reset the image entry is received, the first terminal can acquire at least two images through the image acquisition component in the first terminal. The acquired image is a face image obtained by the first terminal from the face information of the image. The first terminal selects the image with the highest quality from the at least two acquired images as the first image data and uploads it to the backend server according to the quality of each acquired image.

[0169] In one possible implementation, in response to receiving a specified operation to reset the image entry, the image acquisition component in the first terminal is invoked to acquire a second video segment; based on each video frame in the second video segment, at least two acquired images are acquired.

[0170] When a specified operation to reset the image entry is received, the first terminal can also call the image acquisition component in the first terminal to obtain the second video segment, and then determine at least two captured images obtained by the first terminal from the user's facial information in each video frame of the second video segment.

[0171] In one possible implementation, the acquisition score corresponding to each of the at least two acquired images is determined based on the image parameters corresponding to each of the at least two acquired images.

[0172] The image parameters include at least one of the following: face size, face angle, image contrast, image brightness, and image sharpness.

[0173] When the image acquisition component in the first terminal is invoked and at least two acquired images are obtained, the acquisition parameters of each acquired image can be determined based on at least one of the following: face size, face angle, image contrast, image brightness, and image clarity. The face image with the best overall evaluation among the acquired images can then be selected as the third face image data.

[0174] In one possible implementation, in response to receiving a specified operation on the image reset entry, the image acquisition component in the first terminal is invoked to acquire video images in real time. In response to receiving a second video frame with a quality score greater than the quality threshold, the acquisition of video images is stopped, and the image corresponding to the second video frame is identified as the third face image data.

[0175] Upon receiving a specified operation to reset the image entry, the image acquisition component in the first terminal can be directly invoked to acquire video frames in real time and perform quality analysis on the acquired video frames in real time to obtain the quality score of each video frame. When a second video frame with a quality score greater than the quality threshold is received, it indicates that the second video frame meets the quality requirements for uploading to the server for face recognition. Therefore, the second video frame can be directly identified as the third face image data, and the invocation of the image acquisition component can be stopped.

[0176] Step 407: The first terminal sends the third face image data to the backend server so that the backend server can update the second face image data in the face template database based on the third face image data; correspondingly, the backend server receives the third face image data sent by the first terminal based on the image reset entry.

[0177] When the third face image data is sent to the backend server, the backend server can update the second face image data based on the third face image data to obtain the updated second face image. In the subsequent face recognition process, when it is necessary to verify the identity corresponding to the first user account, face recognition operation can be performed based on the updated second face image to correct the inaccuracy of face data in the face template database caused by the long time.

[0178] Step 408: The backend server updates the second face image data in the face template database based on the third face image data.

[0179] In one possible implementation, the backend server deletes the second face image data and identifies the third face image data as the updated second face image data.

[0180] In one possible implementation, after the third face image data is sent to the backend server, the backend server compares the third face image data with the second face image data to obtain an updated similarity. In response to the updated similarity being greater than a first update threshold, the backend server deletes the second face image data and identifies the third face image data as the updated second face image data.

[0181] When the backend server sends an image reset entry to the first terminal and receives the third face image data uploaded by the first terminal for updating the face template data corresponding to the first user account, it can first compare the second face image corresponding to the first user account with the third face image data uploaded by the user through the first terminal. Since the facial changes of the same user are usually not significant, if the second face image data stored in the face template database differs greatly from the third face image data uploaded by the user through the first terminal, there may be a risk that the user account has been stolen. In this case, the update of the second face image data based on the third face image data will be refused.

[0182] In one possible implementation, after the third face image data is sent to the backend server, the backend server compares the third face image data with the second face image data to obtain an updated similarity. If the updated similarity is less than a first update threshold, the server obtains the age information corresponding to the second face image data through an age prediction model. If the age information corresponding to the second face image data is less than the age threshold and the updated similarity is greater than the second update threshold, the server determines the third face image data as the updated second face image data.

[0183] Since younger users may experience significant changes in facial features within a short period, when comparing the third and second facial image data, if the updated similarity is less than the first update threshold, the age information corresponding to the second facial image data (e.g., 14 years old) is first determined based on the age prediction model. If the age information corresponding to the second facial image data is less than the age threshold (e.g., 16 years old), it indicates that the user may have experienced changes in facial features due to growth and development. However, the facial features of the user after growth and development should still have a certain degree of similarity to the facial features saved before growth and development. Therefore, the updated similarity can be compared with the second update threshold, which is smaller than the first update threshold. When the updated similarity is greater than the second update threshold, it indicates that the user is highly likely to be the same person. At this time, the third facial image is determined as the updated second facial image data.

[0184] The age prediction model mentioned above can be a machine learning model trained based on different face images and the age annotation information corresponding to each face image.

[0185] In one possible implementation, the third face image data is fused with the second face image data to obtain updated second face image data.

[0186] In one possible implementation, the third face image data and the second face image data are weighted and fused together based on the fusion weight to obtain the updated second face image data. The fusion weight is used to indicate the proportion of the third face image data in the weighted fusion process.

[0187] When the third face image data is weighted and fused with the second face image data, the updated second face image data contains features of the second face image data previously uploaded by the user, as well as features of the third face image data. Therefore, the updated second face image data obtained after fusion contains the user's facial features at different time periods, resulting in better recognition performance.

[0188] In one possible implementation, a first time difference is obtained between the second face image data and the third face image data, and the fusion weight is obtained based on the first time difference.

[0189] For example, when the first time difference is large, it means that the time between the uploaded third face image and the second face image data is long, and the credibility of the second face image data stored in the face template database is low. Therefore, the fusion weight can be set to a larger value, which increases the proportion of the third face image data in the image fusion process.

[0190] In one possible implementation, a second prompt message is displayed on the face recognition interface. This second prompt message is used to prompt the user to view the image reset entry on the first terminal. The second prompt message is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions.

[0191] When the first face image data is sent from the first terminal to the backend server, the first terminal is performing face recognition operations, and a face recognition interface is displayed on the terminal interface of the first terminal. A second prompt message is displayed on the face recognition interface to prompt the user to view the image reset entry on the first terminal.

[0192] Please refer to Figure 6 This illustration shows a schematic diagram of a prompt message according to an embodiment of this application. Figure 6 As shown, when the first terminal sends the first face image data to the backend server, a face recognition interface is displayed on the image display component of the first terminal, such as... Figure 6As shown in section 610, after the face recognition statistics of the first user account meet the image reset conditions, the backend server can send a second prompt message to the first terminal. The user can click on this second prompt message to directly jump to a page via the corresponding navigation control. Figure 6 The client interface shown in section 620 has an image reset entry.

[0193] In another possible implementation, in response to the first face image data being sent to the backend server by the second terminal, a third prompt message is displayed on the face recognition interface of the second terminal, which prompts the user to view the image reset entry on the first terminal.

[0194] That is, when the first face image data is sent to the backend server by the second terminal, a third prompt message can be displayed on the face recognition interface of the second terminal so that the user can confirm from the third prompt message displayed on the second terminal that the backend server has sent the image reset entry to the first terminal.

[0195] Please refer to Figure 7 This illustration shows a schematic diagram of a prompt message according to an embodiment of this application. Figure 7 As shown, when the first facial image data is sent from the second terminal 710 to the backend server 700, the backend server can send a third prompt message to the second terminal, and the backend server can send an image reset entry to the first terminal 720. At this time, when the user performs facial recognition through the second terminal, they can obtain the message "Facial recognition is at risk of failure; please update your facial identity information on the first terminal" shown in the third prompt message. Based on the image reset entry in the client interface of the second terminal according to the third prompt message, the user can update the facial image information corresponding to the first user account.

[0196] In one possible implementation, in response to sending an image reset entry to the first terminal and not receiving third face image data sent by the first terminal based on the image reset entry within a first specified time, the first user account is determined to be in an automatic update state; in response to the first user account being in an automatic update state, when receiving fourth face image data sent by the first terminal and the fourth face image data meets the image update conditions, the second face image data is updated in the face template database based on the fourth face image data.

[0197] When the server sends an image reset entry to the first terminal, and the server does not receive third face image data sent by the first terminal based on the image reset entry within a first specified time, it indicates that the first user did not trigger the image reset entry within the first specified time, or triggered the image reset entry but failed to upload the third face image data. At this time, the server can set the first user account to automatic update status so that it can update the data based on the face data uploaded later corresponding to the first user account, thus avoiding the failure of the user to perform the trigger operation on the image reset entry, which would result in the second face image data corresponding to the first user account not being updated in a timely manner.

[0198] If the first user account is in an automatic update state, the server receives the fourth face image data sent by the first terminal. If the fourth face image data meets the image update conditions, it means that the fourth face image data can be used to update the second face image data. At this time, the second face image data is updated directly based on the fourth face image data, thereby realizing the automatic update of the face template image (i.e., the second face image data) corresponding to the first user account.

[0199] In one possible implementation, the image update condition is that the similarity to the second face image data is greater than a third update threshold.

[0200] That is, when the similarity between the fourth face image data uploaded by the first terminal and the second face image data is greater than the third update threshold, the confidence level of the fourth face image data is high, the difference between the fourth face image data and the second face image data is not large, and the security risk of updating the fourth face image data to the new second face image data is small. At this time, the second face image data can be updated according to the fourth face image data.

[0201] In one possible implementation, in response to sending an image reset entry to the first terminal and not receiving third face image data sent by the first terminal based on the image reset entry within the first specified time period, historical face image data uploaded by the first terminal within the first specified time period is obtained; based on the historical face image data uploaded by the first terminal within the first specified time period, the second face image data is updated in the face template database.

[0202] When an image reset entry is sent to the first terminal, and no third face image data is received from the first terminal based on the image reset entry within the first specified time, the server can obtain the historical face image data uploaded by the first terminal from an interface other than the image reset entry after the image reset entry is sent to the first terminal, and update the second face image data in the face template database based on the historical face image data.

[0203] In one possible implementation, the second face image data is updated based on the historical face image data with the highest quality score among the various historical face image data uploaded by the first terminal within the first specified time period.

[0204] In one possible implementation, the second face image data is updated based on the historical face image data with the highest similarity to the second face image data among the various historical face image data uploaded by the first terminal within the first specified time period.

[0205] When the server obtains the historical face image data uploaded by the first terminal within a first specified time, it can further obtain the similarity between each historical face image data and the second face image data, and update the second face image data based on the historical face image data with the highest similarity.

[0206] In one possible implementation, in response to the first terminal uploading historical face image data within a first specified time period, if there is historical face image data whose similarity to the second face image data is higher than a third update threshold, the second face image data is updated based on the historical face image data with the highest similarity to the second face image data among the historical face image data.

[0207] If the server finds no historical face image data uploaded by the first terminal within a specified time period that has a similarity higher than the third update threshold with the second face image data, it indicates that the historical face image data uploaded by the first terminal within the specified time period is not suitable for updating the face data corresponding to the first user account. If the server finds historical face image data uploaded by the first terminal within the specified time period that has a similarity higher than the third update threshold with the second face image data, it indicates that the historical face image data uploaded by the first terminal within the specified time period is suitable for updating the face data corresponding to the first user account. In this case, the server updates the second face image data by adding the historical face image data uploaded by the first terminal within the specified time period that has a similarity higher than the third update threshold with the second face image data.

[0208] In one possible implementation, if an image reset entry is sent to the first terminal and no third face image data is received from the first terminal through the image reset entry, in response to receiving the fifth face image data corresponding to the first user account, the image reset entry is resent to the first terminal. The fifth face image data is used to trigger the face recognition service corresponding to the first user account.

[0209] When the server receives the fifth face image data uploaded by the first terminal or IoT device, and recognizes that the fifth face image data is used to trigger the face recognition service corresponding to the first user account, it means that after the first terminal sends the image reset entry, the user did not trigger the second face image data update process through the image reset entry sent by the first terminal, but continued to implement the face recognition service corresponding to the first user account through the first terminal or IoT device. When the server receives the fifth face image data corresponding to the first user account, it can resend the image reset entry to the first terminal so as to prompt the user again to trigger the second face image data update process through the image reset entry.

[0210] In one possible implementation, in response to sending the image reset entry to the first terminal a first specified number of times, the first user account is determined to be in an automatic update state.

[0211] When the server sends the image reset entry to the first terminal a first specified number of times, and the user still does not trigger the update process of the second face image data through the image reset entry after receiving the first specified number of image reset entries, in order to ensure the security and accuracy of the face recognition service of the first user account, the first user account can be set to automatic update status so as to realize the automatic update of the second face image data corresponding to the first user account.

[0212] In another possible implementation, in response to sending the image reset entry to the first terminal a first specified number of times, the second face image data is updated in the face template database based on a first specified number of historical face image data recently received by the server.

[0213] When the server sends the image reset entry to the first terminal a first specified number of times, the server can update the second face image data in the face template database based on the first specified number of recently received historical face image data, so as to realize the automatic update of the second face image data corresponding to the first user account.

[0214] In one possible implementation, in response to sending the image reset entry to the first terminal a first specified number of times, the second face image data is updated based on the historical face image data recently received by the server.

[0215] In another possible implementation, in response to sending the image reset entry to the first terminal a first specified number of times, the second face image data is updated based on historical face image data recently received by the server that has a similarity greater than a third update threshold with the second face image data.

[0216] In another possible implementation, in response to sending the image reset entry to the first terminal a first specified number of times, the second face image data is updated based on the historical face image data with the highest similarity to the second face image data from the first specified number of historical face image data recently received by the server.

[0217] Please refer to Figure 8 This diagram illustrates the overall architecture of a face image data update system according to an embodiment of this application. Figure 8 As shown, the overall architecture of the face image data update system includes a user mobile phone 820, an IoT face terminal device 810, and a backend server 800.

[0218] The IoT face terminal device 810 includes a 3D camera. In one possible implementation of this application embodiment, in order to enhance the security of user face recognition, a 3D camera is used on the IoT face terminal. The data output by the camera includes not only RGB images but also depth images and other related information.

[0219] The IoT facial recognition terminal device 810 includes a facial recognition app. The core of this app includes a facial recognition module and a user facial payment status display module. The facial recognition module mainly comprises a facial acquisition section and a facial selection section. The facial acquisition section uses a 3D camera to capture RGB image streams, depth image streams, and infrared image streams. The facial selection section uses a comprehensive evaluation based on factors such as face size, face angle, image contrast, image brightness, and sharpness to select the optimal facial image.

[0220] After successful data collection and selection, the IoT face recognition terminal will send the data to the backend for recognition via the network module. The APP front-end interface will enter the LOADING state, waiting for the backend to query the final payment result.

[0221] The IoT facial recognition terminal device 810 also includes a user facial recognition payment status display module, which is mainly used to display the current status information of the user after facial recognition payment, such as payment success or payment failure. When certain backend policies are met, the user will be notified that a reset entry will be pushed to the user's mobile APP, where the user can reset the background image (i.e., upload facial image data) on the mobile phone.

[0222] The core modules of the IoT face recognition backend server 800 include face payment service, basic account service, scheduled service, push service, policy control service, and face image update service.

[0223] The facial recognition service receives facial data uploaded from the terminal, extracts features from the current image, compares these features with features in the database, identifies the feature data with the highest score, and then compares the relevant facial data with the facial data in the backend database. Within the payment system, this ultimately returns the user's account or payment code information. When a user recognition anomaly occurs, such as a persistent failure in facial comparison, the backend policy service is queried. If the threshold set by the policy service is exceeded, the basic account service is invoked to mark the user's facial image in the user's basic account as needing an update.

[0224] The basic account system refers to the ability to retrieve basic information about a user through that user account, including the user's current facial payment status.

[0225] Scheduled service refers to the backend periodically retrieving basic user information, querying user information that marks the user's facial background image as needing to be updated, retrieving the user, and then calling the push service to push the background image reset entry to the user's mobile phone.

[0226] Push service refers to the ability of the backend to push notifications for the background map reset entry through a long-term connection between the app and the backend.

[0227] The strategy control service refers to the backend adjusting the threshold information for controlling face comparison failures based on the actual situation, such as 3 consecutive failures or 5 intermittent failures (for example, adjusting the threshold for controlling face comparison failures when the user updates the background image a lot within a specified time).

[0228] The Face Background Image Update Service is used to receive background image update requests from users' mobile devices and update the underlying face image corresponding to the account.

[0229] The user's mobile phone has the First APP installed. After logging into the First APP, the user's account information is accessed. The APP includes a notification reminder for facial recognition payment reset and a module for resetting using a facial image.

[0230] The user's mobile app 820 also has a facial recognition payment reset message module, which is used to receive the backend push of the user's facial background image abnormal reset entry, and can be notified through the payment account.

[0231] The 820 mobile app also includes a face background image reset module. When the user clicks on it, the phone's camera is used to capture and select a face. The selected image is then sent to the backend face background image update service, which updates the image in the face database.

[0232] This architecture enables the facial recognition backend to dynamically control and periodically retrieve user information whose facial images do not match. It then pushes a background image reset option to the user's mobile device, allowing users to proactively resolve issues related to outdated background images and improving user experience. In facial recognition payment scenarios, combined with the facial recognition payment service and account system, the backend periodically scans and identifies abnormal user information and pushes a background image reset option to the user's mobile device. Users can then re-collect and upload their background images for subsequent offline facial recognition payment comparisons, effectively solving the problem of users with outdated background images being unable to use facial recognition payments offline.

[0233] In summary, the solution shown in this application embodiment allows the backend server to maintain statistical information based on the face recognition records of the first user account during the face recognition process. When the backend server detects that the face recognition records of the first user account meet the image reset conditions based on the statistical information, it can send an image reset entry to the terminal corresponding to the first user account so that the terminal can upload new face image data and update the face data in the face template database. This minimizes the occurrence of face recognition errors caused by outdated background images during the face recognition process and improves the accuracy of face recognition.

[0234] Figure 9 This is a schematic diagram illustrating a face image data update method according to an exemplary embodiment. In this embodiment, the face image data update method is jointly executed by a first terminal, a second terminal, and a backend server, and the face image data update method may include the following steps:

[0235] S901, the second terminal sends the first face image data to the backend server. The first user triggers the face recognition function on the second terminal. The second terminal calls the image acquisition component to obtain the first face image data corresponding to the first user and uploads the first face image data to the backend server. S902, the backend server performs face recognition. After receiving the first face image data uploaded by the second terminal, the backend server determines the second face image data most similar to the first face image data in its face template database and determines the first user account corresponding to the second face image data. Then, it compares the first face image data with the second face image data. When the comparison result between the first face image data and the second face image data meets the statistical update conditions, it updates the face recognition statistics information corresponding to the first user account. S903, the backend server sends an image reset entry to the first terminal. When the face recognition statistics information of the backend server meets the image reset conditions, it sends an image reset entry to the first terminal. S904, the first terminal sends the third face image data to the backend server. When the first terminal receives the image reset entry sent by the backend server and receives notification that the first user has triggered the image reset entry, the first terminal calls the image acquisition component to obtain the third face image data corresponding to the first user and uploads the third face image data to the backend server. S905, the backend server updates the second face image data. The backend server updates the second face image data based on the third face image data uploaded by the first terminal.

[0236] Figure 10 This is a structural block diagram illustrating a face image data updating device according to an exemplary embodiment. This target region determination device can realize the function of... Figure 2 or Figure 4 The method provided in the illustrated embodiment includes all or part of the steps, and the face image data updating device includes:

[0237] The first image receiving module 1001 is used to receive first face image data;

[0238] The first comparison result acquisition module 1002 is used to acquire a first comparison result between the first face image data and the second face image data; the second face image data is the face image data in the face template database that has the highest similarity to the first face image data;

[0239] The first information update module 1003 is used to update the face recognition statistics of the first user account in response to the first comparison result meeting the statistical update conditions; the first user account is the user account corresponding to the second face image data.

[0240] The first entry sending module 1004 is used to send an image reset entry to the first terminal in response to the face recognition statistics meeting the image reset conditions; the first terminal is a terminal logged into the first user account;

[0241] The second image receiving module 1005 is used to receive third face image data sent by the first terminal based on the image reset entry;

[0242] The image data update module 1006 is used to update the second face image data in the face template database based on the third face image data.

[0243] In one possible implementation, the first information update module 1003 is further used to,

[0244] In response to the first comparison result satisfying the statistical update condition, the statistical count corresponding to the first user account is updated, and the statistical count is used to indicate the number of times that the comparison results corresponding to the second face image data have satisfied the statistical update condition.

[0245] In one possible implementation, the first ingress sending module 1004 includes:

[0246] The account status setting unit is used to set the first user account to a pending update status in response to the face recognition statistics meeting the data reset conditions.

[0247] The first entry sending unit is used to send the image reset entry to the first terminal in response to detecting that the first user account is in a pending update state at a specified time.

[0248] In one possible implementation, the image reset condition includes at least one of the following:

[0249] In the comparison results corresponding to the second face image data, the number of times the statistical update condition is met consecutively reaches the first threshold.

[0250] In the comparison results corresponding to the second face image data, the cumulative number of times the statistical update condition is met reaches the second threshold.

[0251] In the comparison results corresponding to the second face image data, the ratio of the first number of times the statistical update condition is met reaches the third threshold.

[0252] In one possible implementation, the statistical update condition includes at least one of a first condition and a second condition;

[0253] The first condition includes: the first comparison result indicates that the first face image data and the second face image data do not match;

[0254] The second condition includes: the first comparison result indicates that the first face image data matches the second face image data, and the similarity between the first face image data and the second face image data is less than the first similarity threshold.

[0255] In one possible implementation, the first ingress sending module 1004 further includes:

[0256] The frequency ratio acquisition unit is used to, in response to the face recognition statistics information satisfying the image reset condition, acquire the second frequency ratio between the number of times the first condition is satisfied and the number of times the second condition is satisfied in the previous comparison results corresponding to the second face image data;

[0257] The second entry sending unit is used to send the image reset entry to the first terminal based on the second count ratio; the reminder strength of the first prompt information in the image reset entry is positively correlated with the second count ratio; the prompt information is used to prompt the user to update the face image data.

[0258] In one possible implementation, the first face image data is face image data collected by the second terminal based on a face recognition interface, and the device further includes:

[0259] In response to the face recognition statistics meeting the image reset conditions and the face recognition interface not being closed on the second terminal, a second prompt message is sent to the second terminal; the second prompt message is used to prompt the user to view the image reset entry on the first terminal.

[0260] In summary, the solution shown in this application embodiment allows the backend server to maintain statistical information based on the face recognition records of the first user account during the face recognition process. When the backend server detects that the face recognition records of the first user account meet the image reset conditions based on the statistical information, it can send an image reset entry to the terminal corresponding to the first user account so that the terminal can upload new face image data and update the face data in the face template database. This minimizes the occurrence of face recognition errors caused by outdated background images during the face recognition process and improves the accuracy of face recognition.

[0261] Figure 11 This is a structural block diagram illustrating a face image data updating device according to an exemplary embodiment. This face image data updating device can realize the following: Figure 3 or Figure 4The method provided in the illustrated embodiment includes all or part of the steps, and the face image data updating device includes:

[0262] The reset entry display module 1101 is used to display the image reset entry; the image reset entry is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions; the face recognition statistics are updated by the backend server when it receives the first face image data and the first comparison result between the first face image data and the second face image data meets the statistical update conditions; the second face image data is the face image data corresponding to the first user account in the face template database, and the second face image data is the face image data with the highest similarity to the first face image data in the face template database;

[0263] The face image acquisition module 1102 is used to acquire third face image data in response to receiving a specified operation on the image reset entry;

[0264] The face image sending module 1103 is used to send the third face image data to the backend server, so that the backend server updates the second face image data in the face template database based on the third face image data.

[0265] In one possible implementation, the face image acquisition module includes:

[0266] An image component invocation unit is used to invoke the image acquisition component in the first terminal in response to receiving a specified operation on the image reset entry.

[0267] A face image acquisition unit is used to acquire the third face image data acquired by the image acquisition component.

[0268] In one possible implementation, the device further includes:

[0269] The facial recognition interface display module is used to display the facial recognition interface in response to a received facial recognition operation.

[0270] The first face image data acquisition module is used to call the image acquisition component in the first terminal to acquire the first face image data;

[0271] The first face image data sending module is used to send the first face image data to the backend server;

[0272] The face recognition result display module is used to display the face recognition result of the backend server on the first face image data in the face recognition interface.

[0273] In one possible implementation, the device further includes:

[0274] The prompt information display module is used to display a second prompt information in the face recognition interface. The second prompt information is used to prompt the user to view the image reset entry in the first terminal. The second prompt information is sent by the backend server when the face recognition statistics of the first user account meet the image reset conditions.

[0275] In summary, the solution shown in this application embodiment allows the backend server to maintain statistical information based on the face recognition records of the first user account during the face recognition process. When the backend server detects that the face recognition records of the first user account meet the image reset conditions based on the statistical information, it can send an image reset entry to the terminal corresponding to the first user account so that the terminal can upload new face image data and update the face data in the face template database. This minimizes the occurrence of face recognition errors caused by outdated background images during the face recognition process and improves the accuracy of face recognition.

[0276] Figure 12 This is a schematic diagram illustrating the structure of a computer device according to an exemplary embodiment. The computer device can be implemented as a model training device and / or signal processing device in the various method embodiments described above. The computer device 1200 includes a central processing unit (CPU) 1201, a system memory 1204 including random access memory (RAM) 1202 and read-only memory (ROM) 1203, and a system bus 1205 connecting the system memory 1204 and the central processing unit 1201. The computer device 1200 also includes a basic input / output system 1206 to facilitate information transfer between various devices within the computer, and a mass storage device 1207 for storing an operating system 1213, application programs 1214, and other program modules 1215.

[0277] The mass storage device 1207 is connected to the central processing unit 1201 via a mass storage controller (not shown) connected to the system bus 1205. The mass storage device 1207 and its associated computer-readable media provide non-volatile storage for the computer device 1200. That is, the mass storage device 1207 may include computer-readable media (not shown), such as a hard disk or a compact disc read-only memory (CD-ROM) drive.

[0278] Without loss of generality, the computer-readable medium may include computer storage media and communication media. Computer storage media include volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include RAM, ROM, flash memory or other solid-state storage technologies, CD-ROM, or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will recognize that the computer storage media are not limited to the above-mentioned types. The system memory 1204 and the mass storage device 1207 described above can be collectively referred to as memory.

[0279] Computer device 1200 can be connected to the Internet or other network devices via network interface unit 1211 connected to the system bus 1205.

[0280] The memory also includes one or more programs, which are stored in the memory, and the central processing unit 1201 implements these programs by executing them. Figure 2 , Figure 3 or Figure 4 All or part of the steps of the method shown.

[0281] In exemplary embodiments, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory including a computer program (instructions) that can be executed by a processor of a computer device to perform the methods shown in the various embodiments of this application. For example, the non-transitory computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, floppy disk, and optical data storage device, etc.

[0282] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods shown in the various embodiments described above.

[0283] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0284] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for updating facial image data, characterized in that, The method is executed by a backend server, which contains a face template database for face recognition. The method includes: Receive the first face image data; Obtain a first comparison result between the first face image data and the second face image data; the second face image data is the face image data in the face template database that has the highest similarity to the first face image data; In response to the first comparison result satisfying the statistical update condition, the facial recognition statistics of the first user account are updated; the first user account is the user account corresponding to the second facial image data. In response to the face recognition statistics meeting the image reset conditions, an image reset entry is sent to the first terminal; the first terminal is a terminal logged into the first user account; Receive third face image data sent by the first terminal based on the image reset entry; Based on the third face image data, the second face image data is updated in the face template database; The image reset condition includes at least one of the following: the number of times the statistical update condition is continuously met in the comparison results corresponding to the second face image data reaches a first threshold; the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data reaches a second threshold; the cumulative first number ratio of the statistical update condition being met in the comparison results corresponding to the second face image data reaches a third threshold; the first number ratio is the ratio of the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data to the number of comparisons corresponding to the second face image data. The statistical update conditions include at least one of a first condition and a second condition; the first condition includes: the first comparison result indicates that the first face image data and the second face image data do not match; the second condition includes: the first comparison result indicates that the first face image data and the second face image data match, and the similarity between the first face image data and the second face image data is less than a first similarity threshold.

2. The method according to claim 1, characterized in that, The step of updating the facial recognition statistics of the first user account in response to the first comparison result satisfying the statistical update condition includes: In response to the first comparison result satisfying the statistical update condition, the statistical count corresponding to the first user account is updated, and the statistical count is used to indicate the number of times that the comparison results corresponding to the second face image data have satisfied the statistical update condition.

3. The method according to claim 1, characterized in that, The step of sending an image reset input to the first terminal in response to the face recognition statistics meeting the image reset conditions includes: In response to the face recognition statistics meeting the image reset conditions, the first user account is set to a pending update state; In response to detecting that the first user account is in a pending update state at a specified time, the image reset entry is sent to the first terminal.

4. The method according to claim 1, characterized in that, The step of sending an image reset input to the first terminal in response to the face recognition statistics meeting the image reset conditions includes: In response to the face recognition statistics meeting the image reset condition, the second ratio between the number of times the first condition is met and the number of times the second condition is met is obtained in the comparison results corresponding to the second face image data. Based on the second ratio of the number of times, the image reset entry is sent to the first terminal; the reminder strength of the first prompt information in the image reset entry is positively correlated with the second ratio of the number of times; the prompt information is used to prompt the user to update the face image data.

5. The method according to claim 1, characterized in that, The first face image data is face image data collected by the second terminal based on a face recognition interface, and the method further includes: In response to the face recognition statistics meeting the image reset conditions and the face recognition interface not being closed on the second terminal, a second prompt message is sent to the second terminal; the second prompt message is used to prompt the user to view the image reset entry on the first terminal.

6. A method for updating facial image data, characterized in that, The method is executed by a first terminal, which is a terminal logged into a first user account, and the method includes: An image reset entry is displayed; the image reset entry is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions; the face recognition statistics are updated by the backend server when it receives the first face image data and the first comparison result between the first face image data and the second face image data meets the statistical update conditions; the second face image data is the face image data corresponding to the first user account in the face template database, and the second face image data is the face image data with the highest similarity to the first face image data in the face template database; In response to receiving a specified operation on the image reset entry, third-party face image data is acquired; The third face image data is sent to the backend server so that the backend server updates the second face image data in the face template database based on the third face image data; The image reset condition includes at least one of the following: the number of times the statistical update condition is continuously met in the comparison results corresponding to the second face image data reaches a first threshold; the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data reaches a second threshold; the cumulative first number ratio of the statistical update condition being met in the comparison results corresponding to the second face image data reaches a third threshold; the first number ratio is the ratio of the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data to the number of comparisons corresponding to the second face image data. The statistical update conditions include at least one of a first condition and a second condition; the first condition includes: the first comparison result indicates that the first face image data and the second face image data do not match; the second condition includes: the first comparison result indicates that the first face image data and the second face image data match, and the similarity between the first face image data and the second face image data is less than a first similarity threshold.

7. The method according to claim 6, characterized in that, The step of obtaining third-party face image data in response to receiving a specified operation on the image reset entry includes: In response to receiving a specified operation on the image reset entry, the image acquisition component in the first terminal is invoked; The third face image data acquired by the image acquisition component is obtained.

8. The method according to claim 6, characterized in that, Before the image reset entry is displayed, the method further includes: In response to receiving a face recognition operation, display the face recognition interface; The image acquisition component in the first terminal is invoked to obtain the first face image data; The first face image data is sent to the backend server; The face recognition interface displays the face recognition results of the backend server on the first face image data.

9. The method according to claim 8, characterized in that, The method further includes: A second prompt message is displayed on the face recognition interface, which prompts the user to view the image reset entry on the first terminal. The second prompt message is sent by the backend server when the face recognition statistics of the first user account meet the image reset conditions.

10. A facial image data updating device, characterized in that, The device is used as a backend server, the backend server containing a face template database for face recognition, and the device includes: The first image receiving module is used to receive the first face image data; The first comparison result acquisition module is used to acquire a first comparison result between the first face image data and the second face image data; the second face image data is the face image data in the face template database that has the highest similarity to the first face image data; The first information update module is used to update the face recognition statistics of the first user account in response to the first comparison result meeting the statistical update conditions; the first user account is the user account corresponding to the second face image data. The first entry sending module is used to send an image reset entry to the first terminal in response to the face recognition statistics meeting the image reset conditions; the first terminal is a terminal logged into the first user account; The second image receiving module is used to receive third face image data sent by the first terminal based on the image reset entry; An image data update module is used to update the second face image data in the face template database based on the third face image data; The image reset condition includes at least one of the following: the number of times the statistical update condition is continuously met in the comparison results corresponding to the second face image data reaches a first threshold; the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data reaches a second threshold; the cumulative first number ratio of the statistical update condition being met in the comparison results corresponding to the second face image data reaches a third threshold; the first number ratio is the ratio of the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data to the number of comparisons corresponding to the second face image data. The statistical update conditions include at least one of a first condition and a second condition; the first condition includes: the first comparison result indicates that the first face image data and the second face image data do not match; the second condition includes: the first comparison result indicates that the first face image data and the second face image data match, and the similarity between the first face image data and the second face image data is less than a first similarity threshold.

11. A facial image data updating device, characterized in that, The device is used for a first terminal, which is a terminal logged into a first user account, and the device includes: A reset entry display module is used to display the image reset entry; the image reset entry is sent by the backend server in response to the face recognition statistics of the first user account meeting the image reset conditions; the face recognition statistics are updated by the backend server when it receives the first face image data and the first comparison result between the first face image data and the second face image data meets the statistical update conditions; the second face image data is the face image data corresponding to the first user account in the face template database, and the second face image data is the face image data with the highest similarity to the first face image data in the face template database; A face image acquisition module is used to acquire third face image data in response to receiving a specified operation on the image reset entry; A face image sending module is used to send the third face image data to the backend server, so that the backend server updates the second face image data in the face template database based on the third face image data; The image reset condition includes at least one of the following: the number of times the statistical update condition is continuously met in the comparison results corresponding to the second face image data reaches a first threshold; the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data reaches a second threshold; the cumulative first number ratio of the statistical update condition being met in the comparison results corresponding to the second face image data reaches a third threshold; the first number ratio is the ratio of the cumulative number of times the statistical update condition is met in the comparison results corresponding to the second face image data to the number of comparisons corresponding to the second face image data. The statistical update conditions include at least one of a first condition and a second condition; the first condition includes: the first comparison result indicates that the first face image data and the second face image data do not match; the second condition includes: the first comparison result indicates that the first face image data and the second face image data match, and the similarity between the first face image data and the second face image data is less than a first similarity threshold.

12. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, a code set, or an instruction set, the at least one instruction, the at least one program, the code set, or the instruction set being loaded and executed by the processor to implement the face image data update method as described in any one of claims 1 to 9.

13. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or instruction set is loaded and executed by a processor to implement the face image data update method as described in any one of claims 1 to 9.

14. A computer program product, characterized in that, The computer program product includes computer instructions that are executed by a processor to implement the face image data update method as described in any one of claims 1 to 9.

Images