CN114238910A

CN114238910A - Data processing method, device and equipment

Info

Publication number: CN114238910A
Application number: CN202111574447.6A
Authority: CN
Inventors: 张长浩; 傅欣艺; 王维强
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2021-12-21
Filing date: 2021-12-21
Publication date: 2022-03-25

Abstract

An embodiment of the specification provides a data processing method, a data processing device and data processing equipment, wherein the method comprises the following steps: receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user; determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, wherein the image desensitization model is obtained by training a model constructed by a preset depth learning algorithm based on a first loss function, a second loss function and a historical first image, the first loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service; sending the target desensitization image to a server to cause the server to execute the target service based on the target desensitization image.

Description

Data processing method, device and equipment

Technical Field

The present disclosure relates to the field of data processing technologies, and in particular, to a data processing method, device and apparatus.

Background

With the rapid development of computer technology, the real identity of a user can be verified through private data such as face images, fingerprints, irises and the like, and corresponding services are provided for the user after the verification is passed.

In order to protect the security of the private data of the user, after the terminal device obtains the private data of the user, the private data of the user can be encrypted through the secret key and sent to the server, and the server carries out decryption through the secret key and carries out identity authentication on the user according to the decrypted private data.

However, the encrypted private data may be stolen by a malicious third party during data transmission, and the security of encrypting the private data of the user by using the key is poor, so that the private data of the user is easily stolen by malicious intent. Therefore, a solution is needed to improve the security of user private data.

Disclosure of Invention

An object of the embodiments of the present specification is to provide a data processing method, apparatus and device, so as to provide a solution capable of improving security of user private data.

In order to implement the above technical solution, the embodiments of the present specification are implemented as follows:

in a first aspect, an embodiment of the present specification provides a data processing method, including: receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user, wherein the first image comprises biological characteristic data of the target user; determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, wherein the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, the first loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service; sending the target desensitization image to a server to cause the server to execute the target service based on the target desensitization image.

In a second aspect, an embodiment of the present specification provides a data processing method, including: acquiring a training data set for a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user; training an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service; and sending the trained image desensitization model to a client so that the client processes a target image containing biological characteristic data of a target user based on the trained image desensitization model.

In a third aspect, an embodiment of the present specification provides a data processing apparatus, including: the instruction receiving module is used for receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user, wherein the first image comprises biological characteristic data of the target user; the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service; and the image sending module is used for sending the target desensitization image to a server so that the server executes the target service based on the target desensitization image.

In a fourth aspect, an embodiment of the present specification provides a data processing apparatus, including: the data acquisition module is used for acquiring a training data set aiming at a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user; the model training module is used for training an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service; and the data sending module is used for sending the trained image desensitization model to a client so that the client processes a target image containing biological characteristic data of a target user based on the trained image desensitization model.

In a fifth aspect, an embodiment of the present specification provides a data processing apparatus, including: a processor; and a memory arranged to store computer executable instructions that, when executed, cause the processor to: receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user, wherein the first image comprises biological characteristic data of the target user; determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, wherein the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, the first loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service; sending the target desensitization image to a server to cause the server to execute the target service based on the target desensitization image.

In a sixth aspect, an embodiment of the present specification provides a data processing apparatus, including: a processor; and a memory arranged to store computer executable instructions that, when executed, cause the processor to: acquiring a training data set for a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user; training an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service; and sending the trained image desensitization model to a client so that the client processes a target image containing biological characteristic data of a target user based on the trained image desensitization model.

In a seventh aspect, embodiments of the present specification provide a storage medium for storing computer-executable instructions, where the executable instructions, when executed, implement the following processes: receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user, wherein the first image comprises biological characteristic data of the target user; determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, wherein the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, the first loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service; sending the target desensitization image to a server to cause the server to execute the target service based on the target desensitization image.

In an eighth aspect, embodiments of the present specification provide a storage medium for storing computer-executable instructions, which when executed by a processor implement the following process: acquiring a training data set for a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user; training an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service; and sending the trained image desensitization model to a client so that the client processes a target image containing biological characteristic data of a target user based on the trained image desensitization model.

Drawings

In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and for those skilled in the art, other drawings can be obtained according to the drawings without any creative effort.

FIG. 1A is a flow chart of one embodiment of a data processing method of the present disclosure;

FIG. 1B is a schematic diagram of a data processing method according to an embodiment of the present disclosure;

FIG. 2 is a schematic processing diagram of another embodiment of a data processing method;

FIG. 3 is a schematic diagram of data processing according to the present description;

FIG. 4 is a schematic diagram of yet another data processing system according to the present disclosure;

FIG. 5A is a flow chart of one embodiment of a data processing method of the present description;

FIG. 5B is a schematic processing diagram illustrating another embodiment of a data processing method;

FIG. 6 is a schematic processing diagram of another embodiment of a data processing method;

FIG. 7 is a schematic processing diagram of another embodiment of a data processing method;

FIG. 8 is a schematic diagram of data processing according to the present description;

FIG. 9 is a block diagram of an embodiment of a data processing apparatus according to the present disclosure;

FIG. 10 is a block diagram of another embodiment of a data processing apparatus according to the present disclosure;

fig. 11 is a schematic structural diagram of a data processing apparatus according to the present specification.

Detailed Description

The embodiment of the specification provides a data processing method, a data processing device and data processing equipment.

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present specification, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without any creative effort shall fall within the protection scope of the present specification.

Example one

As shown in fig. 1A and fig. 1B, an execution subject of the method may be a terminal device, where the terminal device may be a mobile terminal device such as a mobile phone and a tablet computer, a terminal device such as a personal computer, or a smart wearable device such as a smart watch. The method may specifically comprise the steps of:

in S102, a trigger execution instruction of the target user for the target service is received, and in response to the trigger execution instruction, a first image corresponding to the target user is acquired.

The target service may be any service that needs to authenticate the target user based on the biometric data of the target user, for example, the target service may be a resource transfer service, a privacy information change service, or the like, and the first image may include the biometric data of the target user, for example, the first image may include a face image, fingerprint information, iris information, or the like of the target user.

In implementation, with the rapid development of computer technology, the true identity of a user can be verified through private data such as a face image, a fingerprint, an iris and the like, and a corresponding service is provided for the user after the verification is passed. In order to protect the security of the private data of the user, after the terminal device obtains the private data of the user, the private data of the user can be encrypted through the secret key and sent to the server, and the server carries out decryption through the secret key and carries out identity authentication on the user according to the decrypted private data. However, the encrypted private data may be stolen by a malicious third party during data transmission, and the security of encrypting the private data of the user by using the key is poor, so that the private data of the user is easily stolen by malicious intent. Therefore, a solution is needed to improve the security of user private data. Therefore, the embodiments of the present disclosure provide a technical solution that can solve the above problems, and refer to the following specifically.

Taking the target service as the resource transfer service as an example, the target user may trigger and start the resource management application in the terminal device, and trigger and execute the resource transfer service in the resource management application. The terminal device can start the camera when detecting a triggering execution instruction of a target user for the resource transfer service, and acquire a first image including a face image of the target user through the camera. Or, the terminal device may further generate the first image based on fingerprint information of the target user when detecting that the target user triggers the execution instruction for the resource transfer service.

The first image obtaining method is an optional and realizable obtaining method, and in an actual application scenario, there may be a plurality of different obtaining methods, which may be different according to different actual application scenarios, and this is not specifically limited in this embodiment of the present specification.

In S104, a target desensitization image corresponding to the first image is determined based on the pre-trained image desensitization model.

The image desensitization model can be obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, the first loss function can be used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet preset image use requirements of target services.

In implementation, the terminal device may receive a pre-trained image desensitization model sent by the server, that is, the server may train the model constructed by the preset deep learning algorithm based on the first loss function, the second loss function, and the historical first image, to obtain the trained image desensitization model.

Or the terminal device may also train the model constructed by the preset deep learning algorithm based on the first loss function, the second loss function and the historical first image to obtain the trained image desensitization model.

For example, the terminal device may send a service identifier of the target service to the server, and receive a historical first image determined by the server based on the service identifier of the target service, where the historical first image may be an image that includes user biometric data and is acquired by the server and corresponds to the target service in a preset model training period. The terminal device may train the model constructed by the neural network algorithm based on the acquired historical first image, the first loss function, and the second loss function to obtain a trained image desensitization model.

For example, the first loss function may be used to determine whether the desensitized image output by the image desensitization model (i.e., the historical first desensitized image) meets the preset image desensitization requirement according to the image similarity between the historical first image and the historical first desensitized image (i.e., the desensitized image corresponding to the historical first image is determined based on the historical first image by the image desensitization model). Specifically, for example, if the image similarity between the historical first image and the historical first desensitization image is smaller than the preset similarity threshold, it may be determined that the desensitization image output by the image desensitization model meets the preset image desensitization requirement.

The second loss function may be used to make the desensitized image output by the image desensitization model meet the preset image usage requirement of the target service, for example, the second loss function may be used to determine whether the desensitized image output by the image desensitization model meets the preset image usage requirement of the target service according to the recognizable rate of the historical first desensitized image. Specifically, if the recognizable rate of the historical first desensitization image is not less than the preset recognizable rate threshold, it may be determined that the desensitization image output by the image desensitization model meets the preset image use requirement of the target service. The determination method of the recognizable rate of the historical first desensitized image may be various, and may be different according to different actual application scenarios, which is not specifically limited in this specification.

The terminal equipment can input the first image into a pre-trained image desensitization model to obtain a target desensitization image corresponding to the first image, so that the obtained target desensitization image meets the preset image desensitization requirement and the preset image use requirement of the target service.

In addition, different first loss functions, different second loss functions and different deep learning algorithms for constructing the image desensitization model can be set for different target services, so that the image use requirements and the image desensitization requirements of the different target services can be met.

For example, the image desensitization model can meet different image desensitization requirements by setting different similarity thresholds, and the image desensitization model can meet different image use requirements by setting different identifiable rate thresholds.

In S106, the target desensitization image is sent to the server to cause the server to perform a target transaction based on the target desensitization image.

In implementation, the terminal device may send the target desensitization image and the service identifier of the target service to the server, and the server may perform authentication processing on the target user based on the target desensitization image, and execute the target service after the authentication is passed.

The embodiment of the specification provides a data processing method, which receives a trigger execution instruction of a target user for a target service, and in response to the trigger execution instruction, acquiring a first image corresponding to the target user, the first image containing biometric data of the target user, determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, the image desensitization model is based on the first loss function, the second loss function and the historical first image, the method comprises the steps that a model constructed by a preset deep learning algorithm is trained, a first loss function is used for enabling a desensitization image output by an image desensitization model to meet preset image desensitization requirements, a second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of a target service, and the target desensitization image is sent to a server so that the server can execute the target service based on the target desensitization image. Because the first loss function can be used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function can be used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service, the image desensitization model obtained based on the training of the first loss function and the second loss function can enable the target desensitization model to meet the image desensitization requirement and the image use requirement of the target service.

Example two

As shown in fig. 2, an execution subject of the method may be a terminal device, where the terminal device may be a mobile terminal device such as a mobile phone and a tablet computer, a terminal device such as a personal computer, or an intelligent wearable device such as a smart watch. The method may specifically comprise the steps of:

in S202, a trigger execution instruction of the target user for the target service is received, and in response to the trigger execution instruction, a first image corresponding to the target user is acquired.

Wherein the first image may contain biometric data of the target user.

For the specific processing procedure of S202, reference may be made to relevant contents of S102 in the first embodiment, which is not described herein again.

In S204, a history first image is acquired.

In implementation, the historical first image may be an image determined by the server based on the service identifier of the target service, or the historical first image may also be one or more images in a training data set stored in the terminal device in advance, and the images in the training data set may be images sent to the terminal device by the server within a preset model training period.

In S206, the historical first image is input to the image desensitization model, resulting in a historical first desensitization image.

In practice, the processing manner of S206 may be varied in practical applications, and an alternative implementation manner is provided below, which may specifically refer to the following steps one to two:

step one, based on a preset image convolution algorithm, performing feature extraction processing on a historical first image to obtain a second image.

And secondly, based on a preset image recombination algorithm, carrying out image recombination processing on the second image to obtain a historical first desensitized image.

In implementation, as shown in fig. 3, taking the historical first desensitization image as an example containing a face image of a user, the terminal device may extract effective features related to face recognition in the historical first image based on an image convolution algorithm (Conv), that is, may obtain a second image, and perform inverse processing on the second image through an image reconstruction algorithm (e.g., DeConv) to obtain the historical first desensitization image.

In addition, in practical applications, the processing manner for generating the second image in the first step may be various, and an alternative implementation manner is provided as follows, which may specifically refer to the following processing:

and compressing the historical first image based on a preset normalization algorithm and a preset signal convolution algorithm to obtain a second image. For example, as shown in fig. 4, the second image may be obtained by compressing the historical first image through a signal convolution algorithm (e.g., SignalConv2D) and a normalization algorithm (e.g., GDN), where the signal convolution algorithm may ensure strict alignment of pixel positions, and the GDN may ensure the quality of the second image with less introduced noise. Correspondingly, the second image may be subjected to image reconstruction processing by using a pixel shuffle image reconstruction algorithm to obtain a historical first desensitized image, and specifically, the 28 × 48 second image may be restored to the historical 112 × 3 first desensitized image by using the pixel shuffle image reconstruction algorithm.

In addition, uniform noise is introduced through the GDN to carry out disturbance processing on the historical first image, quantization loss in image storage can be simulated, and data processing accuracy is improved.

In S208, it is determined whether to retrain the image desensitization model according to the historical first image, the historical first desensitization image, the first loss function, and the second loss function, to obtain a trained image desensitization model.

The first loss function can determine whether the image desensitization model meets preset image desensitization requirements or not through the historical first images and the historical first desensitization images, and the second loss function can determine whether the image desensitization model meets the preset image use requirements of the target service or not through the historical desensitization images.

In an implementation, for example, the first loss function may determine whether the image desensitization model satisfies the preset image desensitization requirement by one or more of a first loss score, a second loss score, and a third loss score, the first loss score may be used to determine desensitization effects of the historical first images and the historical first desensitization images at a pixel level, the second loss score may be used to determine desensitization effects of the historical first images and the historical first desensitization images at a feature vector level, and the third loss score may be used to determine desensitization effects of the historical first images and the historical first desensitization images at an identified effect level. Wherein the desensitization effect may represent a distance between the historical first image and the historical first desensitization image.

In practical applications, the processing manner of S208 may be various, and an alternative implementation manner is provided below, which may specifically refer to the following steps one to five:

the method comprises the steps of firstly, obtaining a first vector corresponding to a historical first image and a second vector corresponding to a historical first desensitized image based on a preset image dimension reduction algorithm, and determining a first loss score based on the first vector and the second vector.

The preset image dimension reduction algorithm may be any algorithm that can be used for performing dimension reduction processing on an image, for example, the image dimension reduction algorithm may be a Discrete Cosine Transform (DCT) algorithm, a Principal Component Analysis (PCA) algorithm, or the like.

In implementation, the historical first image and the historical first desensitization image may be processed based on a preset image dimension reduction algorithm, respectively, to obtain a first vector corresponding to the historical first image and a second vector corresponding to the historical first desensitization image. The mean square error of the first vector and the second vector may be used as an error value for the historical first desensitized image, and the first loss score may be determined based on the error value for the historical first desensitized image.

For example, assuming that there are 3 historical first images corresponding to 3 historical first desensitized images, a first vector corresponding to each historical first image and a second vector corresponding to each historical first desensitized image may be obtained based on a preset image dimension reduction algorithm. A mean square error 1 between the acquired historical first image 1 and the historical first desensitized image 1, a mean square error 2 between the historical first image 2 and the historical first desensitized image 2, and a mean square error 3 between the historical first image 3 and the historical first desensitized image 3, and an average of the mean square error 1, the mean square error 2, and the mean square error 3 is determined as a first loss score. Thus, a smaller first loss score indicates a larger difference between the historical first image and the historical first desensitized image at the pixel level, i.e., a better desensitization.

And secondly, acquiring a third vector corresponding to the historical first image and a fourth vector corresponding to the historical first desensitized image based on the pre-trained feature extraction model, and determining a second loss score based on the third vector and the fourth vector.

The feature extraction model may be obtained by training a model constructed by a preset feature extraction algorithm based on a historical second image, and the historical second image may be an image containing user biological feature data.

In implementation, assuming that there are 3 historical first images corresponding to the 3 historical first desensitized images, the cosine similarity between each historical first image and the corresponding historical first desensitized image may be obtained based on the third vector and the fourth vector, and the second loss score may be determined according to the determined cosine similarities, specifically, the cosine similarity 1 corresponding to the historical first image 1 and the historical first desensitized image 1, the cosine similarity 2 corresponding to the historical first image 2 and the historical first desensitized image 2, and the mean value between the cosine similarity 3 corresponding to the historical first image 3 and the historical first desensitized image 3 may be determined as the second loss score. Thus, a smaller second loss score indicates a larger difference between the historical first image and the historical first desensitized image at the feature vector level, i.e., a better desensitization.

And thirdly, acquiring a first recognition rate corresponding to the historical first images and a second recognition rate corresponding to the historical first desensitized images based on a pre-trained first recognition model, and determining a third loss score based on the first recognition rate and the second recognition rate.

The first recognition model can be obtained by training a model constructed by a preset image recognition algorithm based on a historical second image.

In an implementation, the recognition rate of the historical first images and each historical first image in the training data set may be obtained based on a first recognition model trained in advance, and the historical first images in the training data set may be sorted based on the recognition rate. And selecting the corresponding historical first images based on the sorted historical first images and the preset selection number, and determining a first identification rate based on the identification rate corresponding to the selected historical first images.

For example, assuming that 10 historical first images are included in the training data set, the recognition rate between the historical first image 1 and the other 9 historical first images may be obtained based on the pre-trained first recognition model, and the 9 historical first images may be sorted according to the recognition rate. The historical first images with the recognition rates of the top 5 can be selected, and the average value of the recognition rates of the selected 5 historical first images is determined as the first recognition rate corresponding to the historical first images.

Based on the determination method of the first identification rate, a second identification rate corresponding to the historical first desensitized image can be determined, and the ratio of the second identification rate to the first identification rate can be determined as a third loss score, so that the smaller the third loss score is, the less recognizable the image after desensitization is, namely, the better the desensitization effect is.

And step four, acquiring a third recognition rate corresponding to the historical first desensitized image based on a pre-trained second recognition model, and determining a fourth loss score based on the third recognition rate.

The second identification model can be obtained by training a model constructed by a preset second identification algorithm based on a historical second desensitization image, and the historical second desensitization image is obtained by desensitizing the historical second image.

In implementation, the algorithms for constructing the first recognition model, the second recognition model and the feature extraction model may be the same or different machine learning algorithms, and the machine learning algorithms for the algorithms for constructing the first recognition model, the second recognition model and the feature extraction model may be different according to different practical application scenarios, which is not specifically limited in this specification.

Based on the pre-trained second recognition model, a third recognition rate corresponding to the historical first desensitized image can be obtained, and the third recognition rate is determined as a fourth loss score, so that the higher the fourth loss score is, the better image distinguishability is still achieved after image desensitization, and the loss of information amount can be ensured to be smaller through the fourth loss score.

The determination method of the first loss score, the second loss score, the third loss score and the fourth loss score is an optional and realizable determination method, and in an actual application scenario, there may be a plurality of different determination methods, which may be different according to different actual application scenarios, and this is not specifically limited in the embodiment of the present specification.

And step five, determining whether to retrain the image desensitization model based on one or more of the first loss score, the second loss score and the third loss score, and the fourth loss score.

In practice, as shown in fig. 4, in an actual application scenario, the second image may be generated by a signal convolution algorithm and a normalization algorithm, and correspondingly, the fifth step may be processed by the following steps a1 to A3:

in a1, the second image is decompressed to obtain a third image.

In implementation, the second image may be decompressed by a signal convolution algorithm (e.g., SignalConv2D) and an inverse normalization algorithm (e.g., IGDN) to obtain a third image.

In a2, a fifth loss value is determined based on the third image and the first image.

Wherein the fifth loss value may be used to determine a degree of difference between the first image and the third image.

In an implementation, a mean square error between the first image and the third image may be determined as a fifth loss value to ensure that the historical first desensitized image can contain the primary information in the historical first image by the fifth loss value.

At a3, a determination is made whether to retrain the image desensitization model based on one or more of the first loss score, the second loss score, and the third loss score, and the fourth loss score and the fifth loss score.

In S210, a target desensitization image corresponding to the first image is determined based on the pre-trained image desensitization model.

In S212, the target desensitization image is sent to the server to cause the server to perform a target transaction based on the target desensitization image.

For the specific processing procedure of S210 to S212, reference may be made to the relevant contents of S104 to S106 in the first embodiment, which are not described herein again.

EXAMPLE III

As shown in fig. 5A and 5B, an execution subject of the method may be a server, and the server may be an independent server, or a server cluster composed of a plurality of servers. The method may specifically comprise the steps of:

in S502, a training data set for a target service is acquired.

Wherein the training data set may comprise a plurality of first images, which may comprise biometric data of the user.

In implementation, the server may obtain a first image sent by the client (i.e., the terminal device) based on a preset model training period, where the server may store the first image in a corresponding training data set according to a service identifier of a target service sent by the client.

In S504, the image desensitization model constructed by the preset deep learning algorithm is trained based on the first image, the first loss function, and the second loss function, so as to obtain a trained image desensitization model.

The first loss function can be used for enabling the desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function can be used for enabling the desensitization image output by the image desensitization model to meet preset image use requirements of target services.

In implementation, the training process of the image desensitization model can be referred to in relation to S208 in embodiment two, and is not described herein again.

In S506, the trained image desensitization model is sent to the client, so that the client processes the target image containing the biometric data of the target user based on the trained image desensitization model.

In implementation, the server may train the image desensitization model again based on a preset model training period, and send the trained image desensitization model to the client.

The embodiment of the specification provides a data processing method, which comprises the steps of obtaining a training data set aiming at a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user; training an image desensitization model constructed by a preset deep learning algorithm based on a first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling desensitization images output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling desensitization images output by the image desensitization model to meet preset image use requirements of target services; and sending the trained image desensitization model to the client so that the client processes the target image containing the biological characteristic data of the target user based on the trained image desensitization model. Because the first loss function can be used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function can be used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service, the image desensitization model obtained based on the training of the first loss function and the second loss function can enable the target desensitization model to meet the image desensitization requirement and the image use requirement of the target service.

Example four

As shown in fig. 6, an execution subject of the method may be a server, and the server may be an independent server, or a server cluster composed of multiple servers. The method may specifically comprise the steps of:

in S602, a training data set for a target service is acquired.

Wherein the training data set comprises a plurality of first images comprising biometric data of the user.

In S604, the image desensitization model constructed by the preset deep learning algorithm is trained based on the first image, the first loss function, and the second loss function, so as to obtain a trained image desensitization model.

In S606, the trained image desensitization model is sent to the client.

For the specific processing procedures of S602 to S604, reference may be made to the relevant contents of S502 to S504 in the first embodiment, which are not described herein again.

In S608, the target desensitized image sent by the client is received.

The target desensitization image can be a desensitization image corresponding to the target image determined by the client based on the trained image desensitization model.

In S610, the target service is processed based on the target desensitization image.

In implementation, the server may perform authentication processing on the target user based on the target desensitization image, and process the target service after the authentication passes.

In addition, after the identity authentication is passed, the server can also store the target desensitization image as a first image in a training data set of the target service so as to train the image desensitization model again in the model training period

EXAMPLE five

As shown in fig. 7, an execution subject of the method may be a terminal device (i.e., a client) or a server, where the terminal device may be a mobile terminal device such as a mobile phone and a tablet computer, a terminal device such as a personal computer, or an intelligent wearable device such as a smart watch, and the server may be an independent server or a server cluster composed of multiple servers. The method may specifically comprise the steps of:

in S702, the server obtains a training data set for the target service.

Wherein the training data set comprises a plurality of historical first images, the historical first images comprising biometric data of the user.

In implementation, as shown in fig. 8, the server may obtain historical first data that includes biometric data of the user and is sent by the client during a preset model training period.

In S704, the server trains the image desensitization model constructed by the preset deep learning algorithm based on the historical first image, the first loss function, and the second loss function, to obtain the trained image desensitization model.

In S706, the server sends the trained image desensitization model to the client.

In S708, the client receives a trigger execution instruction of the target user for the target service, and acquires a first image corresponding to the target user in response to the trigger execution instruction.

In S710, the client determines a target desensitization image corresponding to the first image based on the pre-trained image desensitization model.

In S712, the client sends the target desensitized image to the server.

In S714, the server processes the target service based on the target desensitization image.

The embodiment of the present specification provides a data processing method, because a first loss function can be used to make a desensitization image output by an image desensitization model meet a preset image desensitization requirement, a second loss function can be used to make a desensitization image output by an image desensitization model meet a preset image usage requirement of a target service, therefore, the image desensitization model obtained by training based on the first loss function and the second loss function can make the target desensitization model take into account the image desensitization requirement and the image use requirement of the target service, so that, on one hand, the problem of information leakage caused by stealing of the target desensitization image by a malicious third party in the data transmission process can be avoided, on the other hand, the server can execute the target service based on the target desensitization image, namely, under the condition of ensuring the normal execution of the target service, the security of the privacy data of the target user is improved.

EXAMPLE six

Based on the same idea, the data processing method provided in the embodiment of the present specification further provides a data processing apparatus, as shown in fig. 9.

The data processing apparatus includes: an instruction receiving module 901, an image determining module 902 and an image sending model 903, wherein:

an instruction receiving module 901, configured to receive a trigger execution instruction of a target user for a target service, and obtain a first image corresponding to the target user in response to the trigger execution instruction, where the first image includes biometric data of the target user;

an image determining module 902, configured to determine a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, where the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function, and a historical first image, the first loss function is used to make a desensitization image output by the image desensitization model meet a preset image desensitization requirement, and the second loss function is used to make the desensitization image output by the image desensitization model meet a preset image usage requirement of the target service;

an image sending module 903, configured to send the target desensitized image to a server, so that the server executes the target service based on the target desensitized image.

In an embodiment of this specification, the apparatus further includes:

the first acquisition module is used for acquiring the historical first image;

the second acquisition module is used for inputting the historical first image into the image desensitization model to obtain a historical first desensitization image;

and the model training module is used for determining whether to retrain the image desensitization model according to the historical first image, the historical first desensitization image, the first loss function and the second loss function so as to obtain the trained image desensitization model.

In this embodiment of the present specification, the first loss function determines whether the image desensitization model satisfies a preset image desensitization requirement through the historical first image and the historical first desensitization image, and the second loss function determines whether the image desensitization model satisfies the preset image usage requirement of the target service through the historical desensitization image.

In this specification, the first loss function determines whether the image desensitization model satisfies a preset image desensitization requirement through one or more of a first loss score, a second loss score and a third loss score, wherein the first loss score is used for determining desensitization effects of the historical first image and the historical first desensitization image at a pixel level, the second loss score is used for determining desensitization effects of the historical first image and the historical first desensitization image at a feature vector level, and the third loss score is used for determining desensitization effects of the historical first image and the historical first desensitization image at an identified effect level.

In an embodiment of this specification, the model training module is configured to:

acquiring a first vector corresponding to the historical first image and a second vector corresponding to the historical first desensitized image based on a preset image dimension reduction algorithm, and determining the first loss score based on the first vector and the second vector;

acquiring a third vector corresponding to the historical first image and a fourth vector corresponding to the historical first desensitized image based on a pre-trained feature extraction model, and determining a second loss score based on the third vector and the fourth vector, wherein the feature extraction model is obtained by training a model constructed by a preset feature extraction algorithm based on a historical second image;

acquiring a first recognition rate corresponding to the historical first image and a second recognition rate corresponding to the historical first desensitized image based on a pre-trained first recognition model, and determining a third loss score based on the first recognition rate and the second recognition rate, wherein the first recognition model is obtained by training a model constructed by a preset image recognition algorithm based on the historical second image;

acquiring a third recognition rate corresponding to the historical first desensitized image based on a pre-trained second recognition model, and determining a fourth loss score based on the third recognition rate, wherein the second recognition model is obtained by training a model constructed by a preset second recognition algorithm based on the historical second desensitized image, and the historical second desensitized image is obtained by desensitizing the historical second image;

determining whether to retrain the image desensitization model based on one or more of the first loss score, the second loss score, and the third loss score, and the fourth loss score.

In an embodiment of the present specification, the second image acquisition is configured to:

based on a preset image convolution algorithm, performing feature extraction processing on the historical first image to obtain a second image;

and based on a preset image recombination algorithm, carrying out image recombination processing on the second image to obtain the historical first desensitized image.

In an embodiment of this specification, the second obtaining is configured to:

compressing the historical first image based on a preset normalization algorithm and a preset signal convolution algorithm to obtain a second image;

the model training module is configured to:

decompressing the second image to obtain a third image;

determining a fifth loss value based on the third image and the first image, the fifth loss value being used to determine a degree of difference between the first image and the third image;

determining whether to retrain the image desensitization model based on one or more of the first loss score, the second loss score, and the third loss score, and the fourth loss score and the fifth loss score.

The embodiment of the specification provides a data processing device, which receives a trigger execution instruction of a target user for a target service, and in response to the trigger execution instruction, acquiring a first image corresponding to the target user, the first image containing biometric data of the target user, determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, the image desensitization model is based on the first loss function, the second loss function and the historical first image, the method comprises the steps that a model constructed by a preset deep learning algorithm is trained, a first loss function is used for enabling a desensitization image output by an image desensitization model to meet preset image desensitization requirements, a second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of a target service, and the target desensitization image is sent to a server so that the server can execute the target service based on the target desensitization image. Because the first loss function can be used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function can be used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service, the image desensitization model obtained based on the training of the first loss function and the second loss function can enable the target desensitization model to meet the image desensitization requirement and the image use requirement of the target service.

EXAMPLE seven

Based on the same idea, the data processing method provided in the embodiment of the present specification further provides a data processing apparatus, as shown in fig. 10.

The data processing apparatus includes: a data acquisition module 1001, a model training module 1002, and a data transmission module 1003, wherein:

a data obtaining module 1001, configured to obtain a training data set for a target service, where the training data set includes a plurality of first images, and the first images include biometric data of a user;

a model training module 1002, configured to train an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, where the first loss function is used to enable a desensitization image output by the image desensitization model to meet a preset image desensitization requirement, and the second loss function is used to enable the desensitization image output by the image desensitization model to meet a preset image usage requirement of the target service;

a data sending module 1003, configured to send the trained image desensitization model to a client, so that the client processes a target image containing biometric data of a target user based on the trained image desensitization model.

In an embodiment of this specification, the apparatus further includes:

a data receiving module, configured to receive a target desensitization image sent by the client, where the target desensitization image is a desensitization image corresponding to the target image and determined by the client based on the trained image desensitization model;

and the service processing module is used for processing the target service based on the target desensitization image.

Example eight

Based on the same idea, embodiments of the present specification further provide a data processing apparatus, as shown in fig. 11.

The data processing apparatus, which may vary considerably in configuration or performance, may include one or more processors 1101 and a memory 1102, where the memory 1102 may store one or more stored applications or data. Wherein memory 1102 may be transient or persistent. The application programs stored in memory 1102 may include one or more modules (not shown), each of which may include a series of computer-executable instructions for the data processing device. Still further, the processor 1101 may be arranged in communication with the memory 1102 for executing a series of computer executable instructions in the memory 1102 on the data processing device. The data processing apparatus may also include one or more power supplies 1103, one or more wired or wireless network interfaces 1104, one or more input-output interfaces 1105, one or more keyboards 1106.

In particular, in this embodiment, the data processing apparatus includes a memory, and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the data processing apparatus, and the one or more programs configured to be executed by the one or more processors include computer-executable instructions for:

receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user, wherein the first image comprises biological characteristic data of the target user;

determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model, wherein the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, the first loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image desensitization requirement, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirement of the target service;

sending the target desensitization image to a server to cause the server to execute the target service based on the target desensitization image.

Optionally, before the determining, based on the pre-trained image desensitization model, a target desensitization image corresponding to the first image, the method further includes:

acquiring the historical first image;

inputting the historical first image into the image desensitization model to obtain a historical first desensitization image;

and determining whether to retrain the image desensitization model according to the historical first image, the historical first desensitization image, the first loss function and the second loss function so as to obtain the trained image desensitization model.

Optionally, the first loss function determines whether the image desensitization model meets a preset image desensitization requirement through the historical first image and the historical first desensitization image, and the second loss function determines whether the image desensitization model meets the preset image usage requirement of the target service through the historical desensitization image.

Optionally, the first loss function determines whether the image desensitization model satisfies a preset image desensitization requirement by one or more of a first loss score for determining desensitization effects of the historical first images and the historical first desensitization images at a pixel level, a second loss score for determining desensitization effects of the historical first images and the historical first desensitization images at a feature vector level, and a third loss score for determining desensitization effects of the historical first images and the historical first desensitization images at an identified effect level.

Optionally, the determining whether to retrain the image desensitization model according to the historical first image, the historical first desensitization image, the first loss function, and the second loss function includes:

Optionally, the inputting the historical first image into the image desensitization model to obtain a historical first desensitization image includes:

Optionally, the performing, based on a preset image convolution algorithm, feature extraction processing on the historical first image to obtain a second image includes:

determining whether to retrain the image desensitization model based on the one or more of the first loss score, the second loss score, and the third loss score, and the fourth loss score, comprising:

decompressing the second image to obtain a third image;

In addition, in this embodiment, the data processing apparatus includes a memory, and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the data processing apparatus, and the one or more programs configured to be executed by the one or more processors include computer-executable instructions for:

acquiring a training data set for a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user;

training an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service;

and sending the trained image desensitization model to a client so that the client processes a target image containing biological characteristic data of a target user based on the trained image desensitization model.

Optionally, the method further comprises:

receiving a target desensitization image sent by the client, wherein the target desensitization image is a desensitization image which is determined by the client based on the trained image desensitization model and corresponds to the target image;

and processing the target service based on the target desensitization image.

The embodiment of the specification provides a data processing device, since a first loss function can be used to make a desensitization image output by an image desensitization model meet a preset image desensitization requirement, a second loss function can be used to make the desensitization image output by the image desensitization model meet the preset image use requirement of a target service, therefore, the image desensitization model obtained by training based on the first loss function and the second loss function can make the target desensitization model take into account the image desensitization requirement and the image use requirement of the target service, so that, on one hand, the problem of information leakage caused by stealing of the target desensitization image by a malicious third party in the data transmission process can be avoided, on the other hand, the server can execute the target service based on the target desensitization image, namely, under the condition of ensuring the normal execution of the target service, the security of the privacy data of the target user is improved.

Example nine

The embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the processes of the data processing method embodiments, and can achieve the same technical effects, and in order to avoid repetition, the details are not repeated here. The computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk.

The embodiment of the specification provides a computer-readable storage medium, since a first loss function can be used to make the desensitization image output by the image desensitization model meet the preset image desensitization requirement, a second loss function can be used to make the desensitization image output by the image desensitization model meet the preset image usage requirement of the target service, therefore, the image desensitization model obtained by training based on the first loss function and the second loss function can make the target desensitization model take into account the image desensitization requirement and the image use requirement of the target service, so that, on one hand, the problem of information leakage caused by stealing of the target desensitization image by a malicious third party in the data transmission process can be avoided, on the other hand, the server can execute the target service based on the target desensitization image, namely, under the condition of ensuring the normal execution of the target service, the security of the privacy data of the target user is improved.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

In the 90 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually making an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development and writing, but the original code before compiling is also written by a specific Programming Language, which is called Hardware Description Language (HDL), and HDL is not only one but many, such as abel (advanced Boolean Expression Language), ahdl (alternate Hardware Description Language), traffic, pl (core universal Programming Language), HDCal (jhdware Description Language), lang, Lola, HDL, laspam, hardward Description Language (vhr Description Language), vhal (Hardware Description Language), and vhigh-Language, which are currently used in most common. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller as pure computer readable program code, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be considered a hardware component, and the means included therein for performing the various functions may also be considered as a structure within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the various elements may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.

As will be appreciated by one skilled in the art, embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, one or more embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of the present description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Embodiments of the present description are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

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

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, 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 disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

One or more embodiments of the present description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. One or more embodiments of the specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only an example of the present specification, and is not intended to limit the present specification. Various modifications and alterations to this description will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present specification should be included in the scope of the claims of the present specification.

Claims

1. A method of data processing, comprising:

2. The method of claim 1, further comprising, prior to the determining a target desensitization image corresponding to the first image based on a pre-trained image desensitization model:

acquiring the historical first image;

3. The method of claim 2, the first loss function determining whether the image desensitization model satisfies preset image desensitization requirements from the historical first images and the historical first desensitization images, the second loss function determining whether the image desensitization model satisfies preset image usage requirements of the target service from the historical desensitization images.

4. The method of claim 3, the first loss function determining whether the image desensitization model satisfies a preset image desensitization requirement by one or more of a first loss score for determining desensitization effects of the historical first images and the historical first desensitization images at a pixel level, a second loss score for determining desensitization effects of the historical first images and the historical first desensitization images at a feature vector level, and a third loss score for determining desensitization effects of the historical first images and the historical first desensitization images at an identified effect level.

5. The method of claim 4, the determining whether to retrain the image desensitization model according to the historical first image, the historical first desensitization image, the first loss function, and the second loss function, comprising:

6. The method of claim 5, the inputting the historical first image into the image desensitization model resulting in a historical first desensitization image, comprising:

7. The method according to claim 6, wherein the performing feature extraction processing on the historical first image based on a preset image convolution algorithm to obtain a second image comprises:

decompressing the second image to obtain a third image;

8. A method of data processing, comprising:

9. The method of claim 8, further comprising:

and processing the target service based on the target desensitization image.

10. A data processing apparatus comprising:

the instruction receiving module is used for receiving a trigger execution instruction of a target user for a target service, and responding to the trigger execution instruction to obtain a first image corresponding to the target user, wherein the first image comprises biological characteristic data of the target user;

the image desensitization model is obtained by training a model constructed by a preset deep learning algorithm based on a first loss function, a second loss function and a historical first image, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service;

and the image sending module is used for sending the target desensitization image to a server so that the server executes the target service based on the target desensitization image.

11. A data processing apparatus comprising:

the data acquisition module is used for acquiring a training data set aiming at a target service, wherein the training data set comprises a plurality of first images, and the first images comprise biological characteristic data of a user;

the model training module is used for training an image desensitization model constructed by a preset deep learning algorithm based on the first image, a first loss function and a second loss function to obtain the trained image desensitization model, wherein the first loss function is used for enabling a desensitization image output by the image desensitization model to meet preset image desensitization requirements, and the second loss function is used for enabling the desensitization image output by the image desensitization model to meet the preset image use requirements of the target service;

and the data sending module is used for sending the trained image desensitization model to a client so that the client processes a target image containing biological characteristic data of a target user based on the trained image desensitization model.

12. A data processing apparatus, the data processing apparatus comprising:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

13. A data processing apparatus, the data processing apparatus comprising:

a processor; and

14. A storage medium for storing computer-executable instructions, which when executed by a processor implement the following:

15. A storage medium for storing computer-executable instructions, which when executed by a processor implement the following: