Data processing method, device, electronic device and storage medium

Abstract

The present application provides a data processing method, including: obtaining target biometric data; establishing a correspondence between the target biometric data and target user information in a target terminal; labeling the target biometric data in the target terminal according to the correspondence, obtaining the labeling data corresponding to the target biometric data, and using the labeling data corresponding to the target biometric data as data for training a biometric data processing model, wherein the biometric data processing model is a network model used by a server to process biometric data. The data processing method provided by the present application ensures that, when the target user does not upload the user data to a server or a cloud, data labeling of the user's biometric data is achieved for use in training the biometric data processing model, thereby solving the problem of privacy leakage caused by uploading the user's biometric data to a server or a cloud.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more particularly to a data processing method. The present application also relates to a data processing device, an electronic device, and a storage medium. Background Art

[0002] With the rapid development of computer technology and the advent of the big data era, more and more network platforms and artificial intelligence devices have begun to emerge in large numbers. In order to better serve users, various network platforms and artificial intelligence devices often need to collect user data and provide users with accurate personalized services based on user data. For example, e-commerce platforms often collect user data to provide users with personalized product recommendations, and intelligent voice terminal devices often need to collect user biometric data to complete personalized biometric authentication, personalized content, and instruction recommendations for users. In the process of network platforms and artificial intelligence devices providing users with accurate personalized services, most of them rely on various deep neural network models. For example, intelligent voice terminal devices are generally based on voice recognition network models and image recognition network models to identify and process the collected user biometric data to achieve user identification and authentication.

[0003] In order to ensure that the deep neural network model has a high accuracy, it is necessary to continuously train and update the deep neural network model. In the process of training the deep neural network model, it is necessary to obtain user data through the server or cloud, and perform specific data annotation on the user data, such as data attribute annotation, to complete the training and updating of the network model. In many cases, user data, especially user biometric data, is likely to involve the user's personal privacy, which may cause users who are sensitive to security and privacy to refuse to share user data. In addition, according to the GDPR (General Data Protection Regulation) that has officially come into effect, no organization or company can upload and store user data without the explicit permission of EU users. Taking the above factors into consideration, how to achieve data annotation of user data to complete the training and updating of the deep network model when the user is unwilling to upload the user data to the server or cloud has become a very challenging and urgent problem to be solved.

[0004] In the prior art, the core idea for solving the problem of how to label user data to complete the training and updating of deep network models when users are unwilling to upload their data to the server or the cloud is to complete the training of the network model on the user terminal and upload some weights of the deep learning model. However, the prior art does not propose a solution to the problem of how to label user data when users are unwilling to upload their data to the server or the cloud. Summary of the invention

[0005] The present application provides a data processing method, device, electronic device and storage medium, so as to realize data labeling of user biometric data without the target user uploading the user data to a server or a cloud.

[0006] The present application provides a data processing method, comprising:

[0007] Obtain target biometric data;

[0008] Establishing a correspondence between the target biometric data and target user information in the target terminal;

[0009] According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

[0010] Optionally, establishing a correspondence between the target biometric data and the target user information in the target terminal includes:

[0011] Determining whether there is target historical biometric data matching the target biometric data in the target terminal;

[0012] If so, the user information corresponding to the target historical biometric data is determined to be the target user information, and the corresponding relationship is established.

[0013] Optionally, the method further includes: if there is no target historical biometric data matching the target biometric data in the target terminal, creating user information for the target biometric data in the target terminal as the target user information, and establishing the corresponding relationship.

[0014] Optionally, the determining whether there is target historical biometric data matching the target biometric data in the target terminal includes:

[0015] Obtaining historical biometric data in the target terminal;

[0016] Comparing the target biometric data with the historical biometric data to obtain a similarity between the target biometric data and the historical biometric data;

[0017] The historical biometric feature data whose similarity reaches a similarity threshold is used as the target historical biometric feature data.

[0018] Optionally, comparing the target biometric data with the historical biometric data to obtain the similarity between the target biometric data and the historical biometric data includes: comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object;

[0019] The taking the historical biometric data whose similarity reaches a similarity threshold as the target historical biometric data includes: taking the historical biometric data whose first similarity reaches a first similarity threshold or the second similarity reaches a second similarity threshold as the target historical biometric data.

[0020] Optionally, the labeling the target biometric data in the target terminal according to the corresponding relationship to obtain the labeling data corresponding to the target biometric data includes:

[0021] According to the corresponding relationship, the target biometric data is merged with the target historical biometric data in the target terminal to obtain the target biometric data corresponding to the target user information;

[0022] The target biometric data corresponding to the target user information is annotated in the target terminal to obtain annotated data corresponding to the target biometric data.

[0023] Optionally, based on the corresponding relationship, the target biometric data and the target historical biometric data are merged in the target terminal, including: obtaining a third similarity between the physiological characteristic data of the target object in the target biometric data and the physiological characteristic data of the object in the target historical biometric data, and merging the physiological characteristic data of the target object and the physiological characteristic data of the object whose third similarity is greater than a third similarity threshold.

[0024] Optionally, based on the corresponding relationship, the target biometric data and the target historical biometric data are merged in the target terminal, including: obtaining a fourth similarity between the behavior characteristic data of the target object in the target biometric data and the behavior characteristic data of the object in the target historical biometric data, and merging the behavior characteristic data of the target object and the behavior characteristic data of the object whose fourth similarity is greater than a fourth similarity threshold.

[0025] Optionally, the physiological characteristic data of the target object includes facial physiological characteristic data of the target object.

[0026] Optionally, the behavior characteristic data of the target object includes voice behavior characteristic data of the target object.

[0027] Optionally, obtaining target biometric data includes:

[0028] obtaining video data including a target object;

[0029] A target object in the video data is determined, and biometric feature extraction is performed on the target object to obtain the target biometric feature data corresponding to the target object.

[0030] Optionally, obtaining the video data including the target object includes: collecting the video data through the target terminal.

[0031] Optionally, the method further includes: after obtaining the video data, blurring the video data corresponding to objects other than the target object in the video data; and after obtaining the target biometric data, blurring the video data corresponding to the target object in the video data.

[0032] In another aspect, the present application provides a data processing device, applied to a target terminal, comprising:

[0033] A biometric data acquisition unit, used to acquire target biometric data;

[0034] A corresponding relationship establishing unit, used to establish a corresponding relationship between the target biometric data and the target user information;

[0035] A data labeling unit is used to label the target biometric data according to the corresponding relationship, obtain the labeled data corresponding to the target biometric data, and use the labeled data corresponding to the target biometric data as data for training a biometric data processing model, wherein the biometric data processing model is a network model used by the server to process biometric data.

[0036] In another aspect, the present application provides an electronic device, including:

[0037] Processor; and

[0038] The memory is used to store a program for the data processing method. After the device is powered on and the program of the data processing method is run by the processor, the following steps are performed:

[0039] Obtain target biometric data;

[0040] Establishing a correspondence between the target biometric data and target user information in the target terminal;

[0041] According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

[0042] In another aspect of the present application, a storage medium is provided, which stores a program for a data processing method, and the program is executed by a processor to perform the following steps:

[0043] Obtain target biometric data;

[0044] Establishing a correspondence between the target biometric data and target user information in the target terminal;

[0045] According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

[0046] In another aspect, the present application provides a data processing method, comprising:

[0047] Obtain target biometric data;

[0048] Establishing a correspondence between the target biometric data and target user information in the target terminal;

[0049] According to the corresponding relationship, the target biometric data is annotated in the target terminal to obtain annotated data corresponding to the target biometric data, and the annotated data corresponding to the target biometric data is used as data for training a biometric data recognition model, wherein the biometric data recognition model is a network model used by a server to process biometric data;

[0050] A target biometric data recognition model to be trained is obtained, and the labeled data corresponding to the target biometric data is used to perform model training on the target biometric data recognition model in the user terminal.

[0051] Compared with the prior art, this application has the following advantages:

[0052] The data processing method provided by the present application, after obtaining the target biometric data, first, establishes a correspondence between the target biometric data and the target user information in the target terminal, then, based on the correspondence, annotates the target biometric data in the target terminal, obtains the annotated data corresponding to the target biometric data, and uses the annotated data corresponding to the target biometric data as data for training the biometric data processing model. The data processing method provided by the present application, after establishing the correspondence between the target biometric data and the target user information in the target terminal, further annotates the target biometric data in the target terminal based on the correspondence between the target biometric data and the target user information, obtains the annotated data corresponding to the target biometric data, thereby ensuring that the data annotation of the user's biometric data is realized when the target user does not upload the user data to the server or the cloud, so as to be used for training the biometric data processing model, thereby solving the privacy leakage problem caused by uploading the user's biometric data to the server or the cloud. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic diagram of an application scenario of the data processing method provided in an embodiment of the present application.

[0054] Figure 2 This is a flowchart of a data processing method provided in the first embodiment of the present application.

[0055] Figure 3 This is a flow chart of a method for determining biometric data matching provided in the first embodiment of the present application.

[0056] Figure 4 This is a schematic diagram of a data processing device provided in the second embodiment of the present application.

[0057] Figure 5 A schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0058] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0059] In order to more clearly demonstrate the data processing method provided by the present application, the application scenario of the data processing method provided by the present application is first introduced.

[0060] The data processing method provided in the present application is generally applied to the scenario of generating annotated data for training a biometric data processing model at a target terminal. In practical applications, when training and updating models such as biometric data processing models used to assist artificial intelligence devices or network platforms in biometric identification or classification such as facial recognition and voice recognition, it is generally necessary to complete it on the server or cloud. At this time, the user terminal is required to upload the training data used for model training to the server or cloud. Since the training data will inevitably involve the biometric data of the object when training the biometric data processing model, such as: the facial physiological characteristics of the object, the voice behavior characteristics of the object, etc., uploading the training data used to train the biometric data processing model to the server or cloud will cause the corresponding privacy to be leaked. Therefore, the training of the network model is completed on the user terminal, and then the trained network model is uploaded to the server or cloud to complete the model. The method can better solve the above problems. When the network model is trained on the user terminal, it is necessary to first annotate the biometric data used for model training to obtain the annotated data corresponding to the biometric data for training the biometric data processing model. Among them, in the application scenario embodiment of the data processing method provided by the present application, the annotated data corresponding to the biometric data is generally the annotated data generated after data annotation for the attributes corresponding to the biometric data, which is used to indicate the type of biometric data, the relationship of the data, the capacity of the data and other information.

[0061] In the application scenario embodiment of the data processing method provided in the present application, the biometric data includes the physiological characteristic data of the object and the behavioral characteristic data of the object, wherein the physiological characteristic data of the object is generally the facial physiological characteristic data of the object, the fingerprint physiological characteristic data of the object and the iris data of the object, and the behavioral characteristic data of the object is generally the voice characteristic data of the object, the morphological characteristic data of the object, etc. The biometric data processing model is a model used for biometric identification or processing, and specifically can be a model that processes the physiological characteristic data of the object alone, or a model that processes the behavioral characteristic data of the object alone, or a model that processes the physiological characteristic data and the behavioral characteristic data of the object at the same time, such as: face recognition model, voice recognition model, audio classification model, etc.

[0062] In the application scenario embodiment of the data processing method provided in the present application, the target terminal is an electronic device loaded with software or programs for information or data transmission, information or data information processing to implement the corresponding supply, such as: personal handheld smart terminal devices, smart speakers, tablet computers, personal computers, etc. Figure 1 The application scenario of the data processing method provided by the embodiment of the present application is given. Figure 1 The corresponding application scenario is realized by installing a program for the data processing method in the target terminal, wherein Figure 1 In the corresponding application scenario, the target terminal is selected as a smartphone.

[0063] First, the video acquisition module 101 in the program for the data processing method will first collect video data including the target object through the audio acquisition device on the smart phone, and then separate the target audio data corresponding to the target object and the target image data corresponding to the target object in the video data, and provide the target audio data and the target image data to the speech feature extraction module 102 and the image feature extraction module 103 in the program for the data processing method respectively. The specific implementation method of separating the target audio data and image data corresponding to the target object in the video data includes: obtaining the target audio data through the first data separation model 101-1 according to the audio data and the image data in the video data at the same time. Specifically, first, the audio data and the image data in the video data are input into the first data separation model 101-1 to obtain the audio-related motion feature data in the image data; secondly, the original audio data corresponding to the audio data is obtained; finally, according to the audio-related motion feature data, the target audio data is analyzed from the original audio data to obtain the target audio data. The specific implementation method of separating the target audio data and image data corresponding to the target object in the video data includes: obtaining the target image data corresponding to the target object through the second data separation model 101-2 according to the image data in the video data. Specifically, the image data in the video data and the target object are input into the second data separation model 101-2 to obtain the target image data.

[0064] In addition, in the application scenario embodiment of the data processing method provided by the present application, when collecting video data through the target terminal, a sensor for target sensing can also be set on the audio acquisition device on the target terminal or on the audio acquisition device connected to the target terminal and capable of exchanging audio data. When the sensor for target sensing detects preset target information, such as facial information of the target object, infrared information of the object, etc., it can control the audio acquisition device to automatically collect video data. The specific implementation method is such as: the audio acquisition device on the target terminal is additionally provided with an infrared sensor for infrared sensing. After the infrared sensor detects the infrared information, the audio acquisition device on the target terminal can collect video data by itself; another example: the audio acquisition device on the target terminal is additionally provided with a millimeter wave radar sensor for target sensing. After the millimeter wave radar sensor detects the millimeter wave in the preset frequency domain, the audio acquisition device on the target terminal can collect video data.

[0065] Secondly, after receiving the target audio data provided by the video acquisition module 101, the speech feature extraction module 102 will extract the speech features of the target object corresponding to the target audio data, such as the voiceprint features of the target object, the audio frequency features of the target object, and the timbre features of the target object, and obtain the speech feature data of the target object corresponding to the target audio data; correspondingly, after receiving the target image data provided by the video acquisition module 101, the image feature extraction module 103 will extract the physiological feature data of the target object corresponding to the target image data, such as the facial physiological features of the target object, and obtain the physiological feature data of the target object corresponding to the target image data.

[0066] It should be noted that in order to protect the user's privacy data, after the video data is collected, it is necessary to first perform fuzzy processing on the video data corresponding to other objects other than the target object in the video data. That is, after the video data is collected, the video data in the video data that involves the user's privacy data and is not related to the target biometric data is first fuzzy processed, such as: fuzzy processing is performed on the video screen of other objects other than the target object in the video data. In addition, after obtaining the target biometric data, it is necessary to further perform fuzzy processing on the video data corresponding to the target object in the video data. That is, after the target biometric data, it is also necessary to perform fuzzy processing on the video data in the audio data that involves the user's privacy data of the target user object, such as: fuzzy processing is performed on the video screen of the target object in the video data. Fuzzy processing of the video data involving the user's privacy data in the video data can prevent the leakage of the user's privacy data in other scenarios involving the interaction of the video data, such as the dissemination and display of the video data.

[0067] In the application scenario embodiment of the data processing method provided in the present application, the specific implementation method of blurring the video data involving user privacy data in the video data can be: reducing the image quality of the video frame corresponding to the relevant video data, such as: reducing the clarity of the video frame, adding noise to the video frame, etc.; it can also be: coding the video frame or relevant parts of the video involving user privacy in the video data, etc.

[0068] Again, the corresponding relationship establishing unit 104 in the program for the data processing method will establish a corresponding relationship between the target biometric data and the target user information in the target terminal. The specific process is as follows: First, the historical biometric data in the target terminal is obtained. Second, the physiological characteristic data of the target object and the behavioral characteristic data of the target object are compared with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object. Third, the historical biometric data whose first similarity reaches the first similarity threshold or the second similarity reaches the second similarity threshold is used as the target historical biometric data. Fourth, the historical biometric data whose first similarity reaches the first similarity threshold or the second similarity reaches the second similarity threshold is used as the target historical biometric data. Fifth, the corresponding relationship between the target biometric data and the target user information is established. Sixth, the target biometric data and the corresponding relationship are provided to the data annotation unit 105 in the program for the data processing method.

[0069] In the application scenario embodiment of the data processing method provided by the present application, the user information of the target user includes the user's identification information. Generally, the user's identification is the user's ID (Identity Document, identity identification number) identification.

[0070] Finally, after obtaining the target biometric data and the corresponding relationship, the data annotation unit 105 in the program of the data processing method will perform data annotation on the target biometric data corresponding to the target user information in the target terminal to obtain the annotation data corresponding to the target biometric data. Specifically, the physiological characteristic data of the target object in the target biometric data with greater similarity and the physiological characteristic data of the object in the target historical biometric data, as well as the behavioral characteristic data of the target object with greater similarity and the behavioral characteristic data of the object in the target historical biometric data are merged to obtain the target biometric data corresponding to the target user information; and then the target biometric data corresponding to the target user information is data annotated in the target terminal to obtain the annotation data corresponding to the target biometric data. Among them, the target biometric data corresponding to the target user information is the target biometric data carrying the target user information. When the target user information is the ID identification information of the target user, the target biometric data corresponding to the target user information can be: the first target physiological characteristic data of user 1, the second target physiological characteristic data of user 1, the first target behavioral data of user 1, and the second target behavioral data of user 1….

[0071] In addition, in the application scenario embodiment of the data processing method provided by the present application, the data annotation unit 105 in the program for the data processing method will further provide the annotation data corresponding to the target biometric data to the model training unit 106 in the program for the data processing method for training the biometric data processing model. After receiving the annotation data corresponding to the target biometric data, the model training unit 106 will use the annotation data corresponding to the target biometric data to perform model training on the target biometric data recognition model in the user terminal, obtain the trained target biometric data recognition model, and upload the trained target biometric data recognition model to the server corresponding to the target terminal.

[0072] The application scenarios of the data processing method provided in this application are not specifically limited in the embodiments of this application, such as: the data processing method provided in this application can also be applied to other types of target terminals. The application scenario of the above-mentioned data processing method is only an embodiment of the application scenario of the data processing method provided in this application. The purpose of providing this application scenario embodiment is to facilitate the understanding of the data processing method provided in this application, and it is not used to limit the data processing method provided in this application. Other application scenarios of the data processing method provided in the embodiments of this application are not described one by one.

[0073] First embodiment

[0074] The first embodiment of the present application provides a data processing method, as follows: Figure 2-Figure 3 Provide explanation.

[0075] Please refer to Figure 2 , which is a flowchart of a data processing method provided in the first embodiment of the present application.

[0076] In step S201, target biometric data is obtained.

[0077] In the first embodiment of the present application, the target biometric data obtained is the biometric data of the target object, the target object includes but is not limited to the target person object, the target biometric data includes the physiological characteristic data of the target object, the behavioral characteristic data of the target object, and may include only the physiological characteristic data of the target object, or only the behavioral characteristic data of the target object, or may include both the physiological characteristic data of the target object and the behavioral characteristic data of the target object. Among them, the physiological characteristic data of the target object includes: the facial biometric data of the target object, the iris biometric data of the target object, the fingerprint characteristic data of the target object, and the palm print characteristic data of the target object, etc. The behavioral characteristic data of the target object includes: the voice characteristic data of the target object. This application specifically takes the example that the target biometric data includes both the physiological characteristic data of the target object and the behavioral characteristic data of the target object, and introduces the data processing method provided in the first embodiment of the present application in detail.

[0078] In practical applications, the video data including the target object is generally collected by a video acquisition device installed on the target terminal, so as to obtain the video data including the target object. The video data is generally a video clip, and can also be an image, and the image can be a video frame, or audio data, generally an audio clip. When the video data is a video clip, in the process of obtaining the target biometric data, it is necessary to first separate the video data, and obtain the image data in the video data and the audio data corresponding to the target object contained in the video data, that is, separate the video clip into an image set and a target audio clip corresponding to the target object. Then, for the image set and the target audio clip, the physiological feature data of the target object and the behavioral feature data of the target object are extracted respectively, so as to obtain the target biometric data corresponding to the target object. That is, the target object in the video data is determined, and the biometric feature extraction is performed on the target object to obtain the target biometric data corresponding to the target object.

[0079] It should be noted that before separating the video clip into an image set and a target audio clip, it is necessary to first confirm whether the audio clip in the video clip is the target audio clip corresponding to the target object, that is, the target audio clip. If so, the relevant behavior feature data extraction operation of "extracting the behavior feature data of the target object for the target audio clip" can be further performed. The so-called whether the audio clip in the video clip is the target audio clip corresponding to the target object includes: the audio clip in the video clip is a voice clip corresponding to an object other than the target object in the video.

[0080] The specific implementation method of determining whether the audio segment in the video segment is the target audio segment corresponding to the target object is as follows:

[0081] First, audio-related motion feature data in the video clip is obtained.

[0082] In the first embodiment of the present application, the so-called audio-related motion feature data in the video clip is the lip motion feature data of the target object in the video clip, and the lip motion feature data of the target object can be one or more of the lip vibration frequency data, lip vibration phase data, and lip vibration amplitude data of the target object. Among them, the lip vibration frequency data is used to indicate the speech speed of the vocal subject, the lip vibration phase data is used to indicate the time point of the lip opening and closing of the vocal subject, and the lip vibration phase data is used to indicate the amplitude of the lip opening and closing of the vocal subject.

[0083] Secondly, obtain the original audio segment corresponding to the video segment.

[0084] After obtaining the audio-related motion feature data in the video clip in the above step, this step is used to obtain the original audio clip corresponding to the above video clip, that is, to obtain the original audio clip corresponding to the above video clip in recording time. The target object corresponding to the original audio clip may appear in whole or in part in the above video clip.

[0085] Thirdly, according to the audio-related motion feature data, the target audio segment is obtained by analyzing the original audio segment.

[0086] After obtaining the audio-related motion feature data in the video clip and the original audio clip corresponding to the video clip in the above steps, this step is used to analyze and obtain the target audio clip from the original audio clip based on the above audio-related motion feature data, that is, to perform speech separation on the original audio clip based on the information obtained above to obtain the target audio clip corresponding to the target object in the video clip.

[0087] The reason for separating the video data is that the target biometric data includes the physiological characteristic data of the target object and the behavioral characteristic data of the target object. After separating the video data into an image set and a target audio segment, it is easier and more convenient to target the image set and the target audio segment. Directly extracting the physiological characteristic data and the behavioral characteristic data of the target object from the video data will increase the dimension of data processing compared with first separating the video data and then extracting features from the image set and the target audio segment respectively. Specifically, the data processing dimension when extracting the physiological characteristic data and the behavioral characteristic data of the target object from the image set and the target audio segment respectively is one, that is, the image dimension and the audio dimension. Directly extracting the physiological characteristic data and the behavioral characteristic data of the target object from the video data respectively is two, that is, the image dimension and the audio dimension. The complexity of processing data with two dimensions is generally much higher than the complexity of processing data with one dimension respectively. In addition, even if the physiological characteristic data and the behavioral characteristic data of the target object are extracted directly from the video data, it is still necessary to first extract the physiological characteristic data and the behavioral characteristic data of the target object, and then extract the physiological characteristic data and the behavioral characteristic data of the target object from the image set and the target audio clip.

[0088] In order to protect the user privacy of the target object and other objects, in the first embodiment of the present application, after obtaining the video data, it is necessary to perform blurring processing on the video data corresponding to other objects other than the target object in the video data. In addition, after obtaining the target biometric data, it is also necessary to perform blurring processing on the video data corresponding to the target object in the video data.

[0089] In step S202, a correspondence between target biometric data and target user information is established in the target terminal.

[0090] In the first embodiment of the present application, the target terminal is an electronic device loaded with software or programs for information or data transmission, information or data information processing to implement corresponding supply, such as: personal handheld smart terminal device, smart speaker, tablet computer, personal computer, etc.

[0091] When establishing the correspondence between the target biometric data and the target user information in the target terminal, it is necessary to first determine whether there is target historical biometric data matching the target biometric data in the target terminal, and then implement the following two different correspondence construction methods according to different determination results:

[0092] The first way is that when there is target historical biometric data matching the target biometric data in the target terminal, the user information corresponding to the target historical biometric data is determined as the target user information, and a corresponding relationship is established.

[0093] The second method is that if there is no target historical biometric data matching the target biometric data in the target terminal, user information for the target biometric data is created in the target terminal as target user information, and a corresponding relationship is established.

[0094] Implementing two different corresponding relationship construction methods for different judgment results can ensure that no matter whether there is historical biometric data that can match the target biometric data in the target terminal, it can be created.

[0095] In the first embodiment of the present application, the process of determining whether there is target historical biometric data matching the target biometric data in the target terminal is as follows: Figure 3 , which is a flow chart of a method for determining biometric data matching provided in the first embodiment of the present application.

[0096] Step S301: Obtain historical biometric data in the target terminal.

[0097] In the first embodiment of the present application, the historical biometric data is the biometric data that already exists in the target terminal, generally the object feature data that has been determined to have annotated data. The historical biometric data may include not only the historical biometric data corresponding to the target object, but also the historical biometric data corresponding to the non-target object.

[0098] Step S302: Compare the target biometric data with the historical biometric data to obtain the similarity between the target biometric data and the historical biometric data.

[0099] In the first embodiment of the present application, the specific operation for obtaining the similarity between the target biometric data and the historical biometric data is as follows: the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data are compared with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object.

[0100] Step S303: The historical biometric feature data whose similarity reaches the similarity threshold is used as the target historical biometric feature data.

[0101] In the first embodiment of the present application, the historical biometric data whose similarity reaches the similarity threshold is used as the target historical biometric data, including: the historical biometric data whose first similarity reaches the first similarity threshold or the second similarity reaches the second similarity threshold is used as the target historical biometric data. That is, as long as one of the similarities between the physiological characteristic data and the behavioral characteristic data of the object in the historical biometric data and the physiological characteristic data and the behavioral characteristic data of the target object exceeds the preset similarity threshold, the historical biometric data is used as the target historical biometric data.

[0102] In step S203, according to the corresponding relationship, the target biometric data is labeled in the target terminal, the labeled data corresponding to the target biometric data is obtained, and the labeled data corresponding to the target biometric data is used as data for training the biometric data processing model. The biometric data processing model is a network model used by the server to process biometric data.

[0103] In the first embodiment of the present application, the annotation data corresponding to the biometric data is generally the annotation data generated after the data is annotated for the attributes corresponding to the biometric data, and is used to indicate the type of the biometric data, the data's ownership relationship, the data's capacity and other information. For example, the annotation data of the target biometric data includes: the target biometric data is the biometric data belonging to the object with ID 1, and the target biometric data includes the first physiological characteristic data, the second physiological characteristic data, the first behavioral characteristic data, the second behavioral characteristic data, etc. ...

[0104] In actual applications, since the labeled data corresponding to the target biometric data is obtained by performing data labeling based on the correspondence between the target biometric data and the target user information established in the target terminal, in the first embodiment of the present application, the labeled data corresponding to the target biometric data is the target biometric data corresponding to the target user information. Furthermore, when obtaining the labeled data corresponding to the target biometric data, it is necessary to first obtain the target biometric data corresponding to the target user information based on the correspondence, and then perform data labeling on the target biometric data corresponding to the target user information in the target terminal to obtain the labeled data corresponding to the target biometric data. In actual applications, there may be matching target historical biometric data in the target terminal, that is, the physiological feature numbers and behavioral feature data in the target biometric data may partially or overlap with the physiological feature numbers and behavioral feature data of the target historical biometric data. By fusing the target biometric data with the target historical biometric data, not only can the duplication of the biometric data corresponding to the target user information be avoided, but also the annotation data corresponding to the physiological feature numbers and behavioral feature data of the target historical biometric data that match the physiological feature numbers and behavioral feature data in the target biometric data can be directly obtained as the annotation data for the physiological feature numbers and behavioral feature data in the partially overlapping target biometric data, thereby reducing the workload of data annotation for the physiological feature numbers and behavioral feature data in the target biometric data.

[0105] When the target biometric data only includes the physiological characteristic data of the target object, the target biometric data is fused with the target historical biometric data in the target terminal according to the corresponding relationship, including: obtaining a third similarity between the physiological characteristic data of the target object in the target biometric data and the physiological characteristic data of the object in the target historical biometric data, and merging the physiological characteristic data of the target object whose third similarity is greater than a third similarity threshold and the physiological characteristic data of the object.

[0106] When the target biometric data only includes behavioral characteristic data of the target object, the target biometric data is fused with the target historical biometric data in the target terminal according to the corresponding relationship, including: obtaining a fourth similarity between the behavioral characteristic data of the target object in the target biometric data and the behavioral characteristic data of the object in the target historical biometric data, and merging the behavioral characteristic data of the target object whose fourth similarity is greater than a fourth similarity threshold and the behavioral characteristic data of the object.

[0107] When the target biometric data includes both the physiological characteristic data of the target object and the behavioral characteristic data of the target object, it is necessary to merge the physiological characteristic data of the target object and the physiological characteristic data of the object whose third similarity is greater than the third similarity threshold, and the behavioral characteristic data of the target object and the behavioral characteristic data of the object whose fourth similarity is greater than the fourth similarity threshold, respectively.

[0108] It should be noted that in the first embodiment of the present application, whether extracting physiological characteristic data of the target object, behavioral characteristic data of the target object, or extracting physiological characteristic data of the object and behavioral characteristic data of the object are generally implemented based on existing image feature extraction models, speech feature extraction models, etc. Among them, the image feature extraction model, speech feature extraction model and other models can be models based on convolutional neural networks, models based on deep neural networks, and models based on generative adversarial networks. The neural networks based on the image feature extraction model, speech feature extraction model and other models are not specifically limited here.

[0109] The data processing method provided in the first embodiment of the present application, after obtaining the target biometric data, first, establishes a correspondence between the target biometric data and the target user information in the target terminal, then, based on the correspondence, annotates the target biometric data in the target terminal, obtains the annotated data corresponding to the target biometric data, and uses the annotated data corresponding to the target biometric data as data for training the biometric data processing model. The data processing method provided in the first embodiment of the present application, after establishing the correspondence between the target biometric data and the target user information in the target terminal, further annotates the target biometric data in the target terminal based on the correspondence between the target biometric data and the target user information, obtains the annotated data corresponding to the target biometric data, thereby ensuring that the data annotation of the user's biometric data is realized when the target user does not upload the user data to the server or the cloud, so as to be used for training the biometric data processing model, thereby solving the privacy leakage problem caused by uploading the user's biometric data to the server or the cloud.

[0110] Second embodiment

[0111] Corresponding to the application scenario embodiment of the data processing method provided by the present application and the data processing method provided by the first embodiment, the second embodiment of the present application provides a data processing device. Since the device embodiment is basically similar to the application scenario embodiment and the first embodiment, the description is relatively simple, and the relevant parts refer to the application scenario embodiment and the partial description of the first embodiment. The device embodiment described below is only illustrative.

[0112] Please refer to Figure 4, which is a schematic diagram of a data processing device provided in the second embodiment of the present application.

[0113] The data processing device comprises:

[0114] A biometric data acquisition unit 401, used to obtain target biometric data;

[0115] A correspondence establishing unit 402, used to establish a correspondence between the target biometric data and the target user information;

[0116] The data labeling unit 403 is used to label the target biometric data according to the corresponding relationship, obtain the labeled data corresponding to the target biometric data, and use the labeled data corresponding to the target biometric data as data for training a biometric data processing model, wherein the biometric data processing model is a network model used by the server to process biometric data.

[0117] Optionally, the correspondence establishing unit 402 is specifically used to determine whether there is target historical biometric data matching the target biometric data in the target terminal; if so, determine that the user information corresponding to the target historical biometric data is the target user information, and establish the correspondence.

[0118] Optionally, the method further includes: if there is no target historical biometric data matching the target biometric data in the target terminal, creating user information for the target biometric data in the target terminal as the target user information, and establishing the corresponding relationship.

[0119] Optionally, the determining whether there is target historical biometric data matching the target biometric data in the target terminal includes:

[0120] Obtaining historical biometric data in the target terminal;

[0121] Comparing the target biometric data with the historical biometric data to obtain a similarity between the target biometric data and the historical biometric data;

[0122] The historical biometric feature data whose similarity reaches a similarity threshold is used as the target historical biometric feature data.

[0123] Optionally, comparing the target biometric data with the historical biometric data to obtain the similarity between the target biometric data and the historical biometric data includes: comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object;

[0124] The taking the historical biometric data whose similarity reaches a similarity threshold as the target historical biometric data includes: taking the historical biometric data whose first similarity reaches a first similarity threshold or the second similarity reaches a second similarity threshold as the target historical biometric data.

[0125] Optionally, the data annotation unit 403 is specifically used to fuse the target biometric data with the target historical biometric data in the target terminal according to the corresponding relationship to obtain the target biometric data corresponding to the target user information; and to perform data annotation on the target biometric data corresponding to the target user information in the target terminal to obtain annotation data corresponding to the target biometric data.

[0126] Optionally, based on the corresponding relationship, the target biometric data and the target historical biometric data are merged in the target terminal, including: obtaining a third similarity between the physiological characteristic data of the target object in the target biometric data and the physiological characteristic data of the object in the target historical biometric data, and merging the physiological characteristic data of the target object and the physiological characteristic data of the object whose third similarity is greater than a third similarity threshold.

[0127] Optionally, based on the corresponding relationship, the target biometric data and the target historical biometric data are merged in the target terminal, including: obtaining a fourth similarity between the behavior characteristic data of the target object in the target biometric data and the behavior characteristic data of the object in the target historical biometric data, and merging the behavior characteristic data of the target object and the behavior characteristic data of the object whose fourth similarity is greater than a fourth similarity threshold.

[0128] Optionally, the physiological characteristic data of the target object includes facial physiological characteristic data of the target object.

[0129] Optionally, the behavior characteristic data of the target object includes voice behavior characteristic data of the target object.

[0130] Optionally, the biometric data obtaining unit is used to obtain video data including a target object; determine the target object in the video data, and perform biometric extraction on the target object to obtain the target biometric data corresponding to the target object.

[0131] Optionally, obtaining the video data including the target object includes: collecting the video data through the target terminal.

[0132] Optionally, the data processing method provided in the second embodiment of the present application further includes: a video data processing unit, which is used to, after obtaining the video data, perform fuzzy processing on the video data corresponding to objects other than the target object in the video data; and after obtaining the target biometric data, perform fuzzy processing on the video data corresponding to the target object in the video data.

[0133] Third embodiment

[0134] Corresponding to the application scenario embodiment of the data processing method provided by the present application and the data processing method provided by the first embodiment, the third embodiment of the present application further provides an electronic device. Since the third embodiment is basically similar to the application scenario embodiment and the first embodiment, the description is relatively simple, and the relevant parts refer to the application scenario embodiment and the partial description of the first embodiment. The third embodiment described below is only illustrative.

[0135] As shown in 5, Figure 5 A schematic diagram of an electronic device provided in an embodiment of the present application.

[0136] The electronic device includes:

[0137] Processor 501; and

[0138] The memory 502 is used to store the program of the data processing method. After the device is powered on and the program of the data processing method is run by the processor, the following steps are performed:

[0139] Obtain target biometric data;

[0140] Establishing a correspondence between the target biometric data and target user information in the target terminal;

[0141] According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

[0142] It should be noted that, for the detailed description of the electronic device provided in the third embodiment of the present application, reference can be made to the application scenario embodiment of the data processing method provided in the present application and the related description of the data processing method provided in the first embodiment, which will not be repeated here.

[0143] Fourth embodiment

[0144] Corresponding to the application scenario embodiment of the data processing method provided by the present application and the data processing method provided by the first embodiment, the fourth embodiment of the present application further provides a storage medium. Since the fourth embodiment is basically similar to the application scenario embodiment and the first embodiment, the description is relatively simple, and the relevant parts refer to the application scenario embodiment and the partial description of the first embodiment. The fourth embodiment described below is only illustrative.

[0145] The storage medium provided in the fourth embodiment stores a program for the data processing method, and the program is executed by a processor to perform the following steps:

[0146] Obtain target biometric data;

[0147] Establishing a correspondence between the target biometric data and target user information in the target terminal;

[0148] According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

[0149] It should be noted that for the detailed description of the storage medium provided in the fourth embodiment of the present application, reference can be made to the application scenario embodiment of the data processing method provided in the present application and the related description of the data processing method provided in the first embodiment, which will not be repeated here.

[0150] Fifth embodiment

[0151] Corresponding to the scenario embodiment of the data processing method provided by the present application and the data processing method provided in the first embodiment, the fifth embodiment of the present application provides another data processing method. Since the data processing method of the fifth embodiment is basically similar to the application scenario embodiment and the first embodiment, the description is relatively simple, and the relevant parts refer to the application scenario embodiment and the partial description of the first embodiment. The fifth embodiment described below is only illustrative.

[0152] The steps corresponding to the data processing method provided in the fifth embodiment of the present application are as follows:

[0153] First, obtain the target biometric data.

[0154] In the fifth embodiment of the present application, the target biometric data obtained is the biometric data of the target object, the target object includes but is not limited to the target person object, the target biometric data includes the physiological characteristic data of the target object, the behavioral characteristic data of the target object, and may include only the physiological characteristic data of the target object, or only the behavioral characteristic data of the target object, or may include both the physiological characteristic data of the target object and the behavioral characteristic data of the target object. Among them, the physiological characteristic data of the target object includes: the facial biometric data of the target object, the iris biometric data of the target object, the fingerprint characteristic data of the target object, and the palm print characteristic data of the target object, etc., and the behavioral characteristic data of the target object includes: the voice characteristic data of the target object.

[0155] Secondly, a correspondence between target biometric data and target user information is established in the target terminal.

[0156] In the fifth embodiment of the present application, the target terminal is an electronic device loaded with software or programs for information or data transmission, information or data information processing to implement corresponding supply, such as: personal handheld smart terminal device, smart speaker, tablet computer, personal computer, etc.

[0157] Again, according to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain the labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training the biometric data recognition model. The biometric data recognition model is a network model used by the server to process biometric data.

[0158] In the fifth embodiment of the present application, the historical biometric data is the biometric data that already exists in the target terminal, generally the object feature data that has been determined to have annotated data. The historical biometric data may include not only the historical biometric data corresponding to the target object, but also the historical biometric data corresponding to the non-target object.

[0159] Finally, the target biometric data recognition model to be trained is obtained, and the labeled data corresponding to the target biometric data is used to perform model training on the target biometric data recognition model in the user terminal.

[0160] In the fifth embodiment of the present application, the annotation data corresponding to the biometric data is generally the annotation data generated after the data is annotated for the attributes corresponding to the biometric data, and is used to indicate the type of the biometric data, the data's ownership, the data's capacity and other information. For example, the annotation data of the target biometric data includes: the target biometric data is the biometric data belonging to the object with ID 1, and the target biometric data includes the fifth physiological characteristic data, the second physiological characteristic data, the fifth behavioral characteristic data, the second behavioral characteristic data, etc. ...

[0161] Although the present application is disclosed as above in the preferred embodiment, it is not intended to limit the present application. Any technical personnel in this field may make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

[0162] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0163] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0164] 1. Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer-readable media does not include non-transitory media such as modulated data signals and carrier waves.

[0165] 2. Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

Claims

1. A data processing method, characterized in that: include: Obtain target biometric data; Establishing a correspondence between the target biometric data and the target user information in the target terminal, including: obtaining historical biometric data in the target terminal; comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object; using the historical biometric data whose first similarity reaches a first similarity threshold or whose second similarity reaches a second similarity threshold as the target historical biometric data; determining that the user information corresponding to the target historical biometric data is the target user information, and establishing the correspondence; According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

2. The data processing method according to claim 1, characterized in that: Also includes: If the historical biometric data whose first similarity reaches a first similarity threshold or whose second similarity reaches a second similarity threshold does not exist in the target terminal, user information for the target biometric data is created in the target terminal as the target user information, and the corresponding relationship is established.

3. The data processing method according to claim 1, characterized in that: The step of labeling the target biometric data in the target terminal according to the corresponding relationship to obtain the labeling data corresponding to the target biometric data includes: According to the corresponding relationship, the target biometric data is merged with the target historical biometric data in the target terminal to obtain the target biometric data corresponding to the target user information; The target biometric data corresponding to the target user information is annotated in the target terminal to obtain annotated data corresponding to the target biometric data.

4. The data processing method according to claim 3, characterized in that: The method of fusing the target biometric data with the target historical biometric data in the target terminal according to the corresponding relationship includes: obtaining a third similarity between the physiological characteristic data of the target object in the target biometric data and the physiological characteristic data of the object in the target historical biometric data, and merging the physiological characteristic data of the target object and the physiological characteristic data of the object whose third similarity is greater than a third similarity threshold.

5. The data processing method according to claim 3, characterized in that: The target biometric data and the target historical biometric data are merged in the target terminal according to the corresponding relationship, including: obtaining a fourth similarity between the behavior feature data of the target object in the target biometric data and the behavior feature data of the object in the target historical biometric data, and merging the behavior feature data of the target object and the behavior feature data of the object whose fourth similarity is greater than a fourth similarity threshold.

6. The data processing method according to claim 1 or 4, characterized in that: The physiological characteristic data of the target object includes facial physiological characteristic data of the target object.

7. The data processing method according to claim 1 or 5, characterized in that: The behavior feature data of the target object includes voice behavior feature data of the target object.

8. The data processing method according to claim 1, characterized in that: The obtaining of target biometric data comprises: obtaining video data including a target object; A target object in the video data is determined, and biometric feature extraction is performed on the target object to obtain the target biometric feature data corresponding to the target object.

9. The data processing method according to claim 8, characterized in that: The obtaining of the video data including the target object includes: collecting the video data through the target terminal.

10. The data processing method according to claim 8 or 9, characterized in that: Also includes: After obtaining the video data, performing fuzzy processing on video data corresponding to other objects other than the target object in the video data; And after obtaining the target biometric data, the video data corresponding to the target object in the video data is fuzzy processed.

11. A data processing device, applied to a target terminal, characterized in that: include: A biometric data acquisition unit, used to acquire target biometric data; A correspondence establishing unit, used for establishing a correspondence between the target biometric data and the target user information, including: obtaining the historical biometric data in the target terminal; comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object; using the historical biometric data whose first similarity reaches a first similarity threshold or whose second similarity reaches a second similarity threshold as the target historical biometric data; determining that the user information corresponding to the target historical biometric data is the target user information, and establishing the correspondence; A data labeling unit is used to label the target biometric data according to the corresponding relationship, obtain the labeled data corresponding to the target biometric data, and use the labeled data corresponding to the target biometric data as data for training a biometric data processing model, wherein the biometric data processing model is a network model used by the server to process biometric data.

12. An electronic device, characterized in that: include: processor; as well as The memory is used to store a program for the data processing method. After the device is powered on and the program of the data processing method is run by the processor, the following steps are performed: Obtain target biometric data; Establishing a correspondence between the target biometric data and the target user information in the target terminal, including: obtaining historical biometric data in the target terminal; comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object; using the historical biometric data whose first similarity reaches a first similarity threshold or whose second similarity reaches a second similarity threshold as the target historical biometric data; determining that the user information corresponding to the target historical biometric data is the target user information, and establishing the correspondence; According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

13. A storage medium, characterized in that: A program for a data processing method is stored, and the program is executed by a processor to perform the following steps: Obtain target biometric data; Establishing a correspondence between the target biometric data and the target user information in the target terminal, including: obtaining historical biometric data in the target terminal; comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object; using the historical biometric data whose first similarity reaches a first similarity threshold or whose second similarity reaches a second similarity threshold as the target historical biometric data; determining that the user information corresponding to the target historical biometric data is the target user information, and establishing the correspondence; According to the corresponding relationship, the target biometric data is labeled in the target terminal to obtain labeled data corresponding to the target biometric data, and the labeled data corresponding to the target biometric data is used as data for training a biometric data processing model, which is a network model used by the server to process biometric data.

14. A data processing method, characterized in that: include: Obtain target biometric data; Establishing a correspondence between the target biometric data and the target user information in the target terminal, including: obtaining historical biometric data in the target terminal; comparing the physiological characteristic data of the target object and the behavioral characteristic data of the target object in the target biometric data with the physiological characteristic data of the object and the behavioral characteristic data of the object in the historical biometric data, respectively, to obtain a first similarity between the physiological characteristic data of the target object and the physiological characteristic data of the object, and a second similarity between the behavioral characteristic data of the target object and the behavioral characteristic data of the object; using the historical biometric data whose first similarity reaches a first similarity threshold or whose second similarity reaches a second similarity threshold as the target historical biometric data; determining that the user information corresponding to the target historical biometric data is the target user information, and establishing the correspondence; According to the corresponding relationship, the target biometric data is annotated in the target terminal to obtain annotated data corresponding to the target biometric data, and the annotated data corresponding to the target biometric data is used as data for training a biometric data recognition model, wherein the biometric data recognition model is a network model used by a server to process biometric data; A target biometric data recognition model to be trained is obtained, and the labeled data corresponding to the target biometric data is used to perform model training on the target biometric data recognition model in the user terminal.

Images