Image data desensitization method and device

Abstract

The disclosed embodiment discloses a method and device for desensitizing image data, wherein the desensitizing method includes: determining a target single-channel image of a native image in a preset channel; determining an area of ​​interest in the target single-channel image based on a behavior recognition task; determining a target sensitive image area in the target single-channel image based on the area of ​​interest in the target single-channel image; determining a mask image of the target single-channel image based on the target sensitive image area; determining a mask image of the native image based on the mask image of the target single-channel image; and performing data desensitization processing on the native data of the native image based on the mask image of the native image. The disclosed embodiment can ensure the data security of the remaining image areas in the native image except for the image areas that require behavior recognition, thereby improving the security of user privacy data.

Description

Technical Field

[0001] The present disclosure relates to the field of data security technology, and in particular to a method and device for desensitizing image data. Background Art

[0002] The fragmented layout of functional areas in traditional car cockpits creates obstacles to human-vehicle interaction. With the application of intelligent control in vehicles, smart cockpits have emerged. Smart cockpits can bring a more intelligent and safe interactive experience, and are also key technologies in assisted driving and autonomous driving.

[0003] The smart cockpit used in related technologies recognizes the behavior of people in the car by taking images of them, and the vehicle itself or the server controls the vehicle based on the behavior recognition results. Although this improves the safety of vehicle control, it also causes user privacy leaks. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure is proposed. The embodiments of the present disclosure provide a method and device for desensitizing image data.

[0005] According to a first aspect of an embodiment of the present disclosure, a method for desensitizing image data is provided, comprising:

[0006] Determine the target single-channel image of the native image in the preset channel;

[0007] Determine a region of interest in the target single-channel image based on a behavior recognition task;

[0008] Determining a target sensitive image area in the target single-channel image based on a region of interest in the target single-channel image;

[0009] Based on the target sensitive image area, determining a mask image of the target single-channel image;

[0010] Determining a mask image of the native image based on the mask image of the target single-channel image;

[0011] Based on the mask image of the native image, data desensitization processing is performed on the native data of the native image.

[0012] According to a second aspect of an embodiment of the present disclosure, a device for desensitizing image data is provided, comprising:

[0013] A single-channel image determination module, used to determine a target single-channel image of a native image in a preset channel;

[0014] A region of interest determination module, used to determine the region of interest in the target single-channel image based on a behavior recognition task;

[0015] A sensitive image region determining module, configured to determine a target sensitive image region in the target single-channel image based on a region of interest in the target single-channel image;

[0016] A first mask image determination module, configured to determine a mask image of the target single-channel image based on the target sensitive image area;

[0017] a second mask image determination module, configured to determine the mask image of the native image based on the mask image of the target single-channel image;

[0018] The data desensitization module is used to perform data desensitization processing on the native data of the native image based on the mask image of the native image.

[0019] According to a third aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and the computer program is used to execute the desensitization method of image data described in the first aspect above.

[0020] According to a fourth aspect of an embodiment of the present disclosure, an electronic device is provided, the electronic device comprising:

[0021] processor;

[0022] a memory for storing instructions executable by the processor;

[0023] The processor is used to read the executable instructions from the memory and execute the instructions to implement the image data desensitization method described in the first aspect above.

[0024] Based on the method and device for desensitizing image data provided by the above-mentioned embodiments of the present disclosure, first determine the target single-channel image of the native image in the preset channel, then determine the region of interest in the target single-channel image based on the behavior recognition task, and then determine the target sensitive image region in the target single-channel image based on the region of interest in the target single-channel image, and then determine the mask image of the target single-channel image. After obtaining the mask image of the target single-channel image, generate a mask image of the native image, and finally perform data desensitization processing on the native data of the native image based on the mask image of the native image, thereby ensuring the data security of the remaining image regions in the native image except for the image regions that require behavior recognition, and improving the security of user privacy data.

[0025] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and other purposes, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail in conjunction with the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation of the present disclosure. In the drawings, the same reference numerals generally represent the same components or steps.

[0027] Figure 1 is a flowchart of a method for desensitizing image data according to an embodiment of the present disclosure;

[0028] Figure 2 is a schematic diagram of generating a mask image of a native image from a native image in an example of the present disclosure;

[0029] Figure 3 is a flow chart of step S1 in one embodiment of the present disclosure;

[0030] Figure 4 is a flow chart of step S2 in one embodiment of the present disclosure;

[0031] Figure 5 is a flow chart of step S4 in one embodiment of the present disclosure;

[0032] Figure 6 is a flow chart of step S6 in one embodiment of the present disclosure;

[0033] Figure 7 is a structural block diagram of a device for desensitizing image data in one embodiment of the present disclosure;

[0034] Figure 8 is a structural block diagram of a single-channel image determination module 100 in one embodiment of the present disclosure;

[0035] Fig. 9 is a structural block diagram of an area of ​​interest determination module 200 in one embodiment of the present disclosure;

[0036] Fig.10 is a structural block diagram of a first mask image determination module 400 in one embodiment of the present disclosure;

[0037] Fig.11 is a structural block diagram of a data desensitization module 600 in one embodiment of the present disclosure;

[0038] Fig.12 is a structural block diagram of a device for desensitizing image data in another embodiment of the present disclosure;

[0039] Fig.13 is a structural diagram of an electronic device provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0040] Below, the exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described here.

[0041] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0042] Those skilled in the art can understand that the terms "first" and "second" in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate the necessary logical order between them.

[0043] It should also be understood that in the embodiments of the present disclosure, “plurality” may refer to two or more than two, and “at least one” may refer to one, two, or more than two.

[0044] It should also be understood that any component, data or structure mentioned in the embodiments of the present disclosure can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0045] In addition, the term "and / or" in the present disclosure is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in the present disclosure generally indicates that the associated objects before and after are in an "or" relationship.

[0046] It should also be understood that the description of the various embodiments in the present disclosure focuses on the differences between the various embodiments, and the same or similar aspects thereof can be referenced to each other, and for the sake of brevity, they will not be described one by one.

[0047] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0048] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0049] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0050] The embodiments of the present disclosure can be applied to electronic devices such as terminal devices, computer systems, servers, etc., which can operate with many other general or special computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, servers, etc. include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, small computer systems, large computer systems, and distributed cloud computing technology environments including any of the above systems, etc.

[0051] Electronic devices such as terminal devices, computer systems, servers, etc. can be described in the general context of computer system executable instructions (such as program modules) executed by computer systems. Generally, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment, where tasks are performed by remote processing devices linked through a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media including storage devices.

[0052] Exemplary Methods

[0053] Figure 1 is a flow chart of the method for desensitizing image data according to an embodiment of the present disclosure. This embodiment can be applied to electronic devices, such as Figure 1 As shown, the following steps are included:

[0054] S1: Determine a target single-channel image of a native image in a preset channel.

[0055] In the embodiments of the present disclosure, the native image is collected by an image sensor. For example, the native image may be a raw image of the driver collected by an image sensor in the vehicle and not image restored during assisted driving or automatic driving, or a raw image of the passenger collected by an image sensor in the vehicle and not image restored.

[0056] Figure 2 FIG. 1 is a schematic diagram of generating a mask image of a native image from a native image in an example of the present disclosure. Figure 2As shown, after the native image is acquired, the vehicle control system or a terminal (such as a mobile phone or a server) that can be connected to control the vehicle can determine the target single-channel image of the native image in the preset channel. The target single-channel image can be one of the R channel image, the G channel image, and the B channel image obtained after the native image is channel-separated, which is determined according to user settings or according to the image quality of the R channel image, the G channel image, and the B channel image obtained after the native image is channel-separated.

[0057] It should be noted that Figure 2 The resolution of the native image shown is 8x8, and the resolution of the target single-channel image is 4x4. In fact, the resolution of the native image and the target single-channel image is determined by the resolution of the image sensor. Figure 2 The 8x8 resolution and 4x4 resolution shown are for illustrative purposes only and do not constitute a limitation to the present disclosure. The embodiments of the present disclosure do not limit the specific sizes of the native image and the mask image. For example, the native image may be an image with a resolution of 1920x1280, an image with a resolution of 1024x768, or an image of other sizes.

[0058] The following embodiment will use the vehicle control system as an example to perform image data desensitization, but those skilled in the art will appreciate that image data desensitization may also be performed using a terminal that can be connected to control the vehicle.

[0059] S2: Determine the region of interest in the target single-channel image based on the action recognition task.

[0060] Specifically, the image recognition technology is used to determine the region of interest in the target single-channel image based on the behavior recognition task. The behavior recognition task may be driver phone call recognition or driver smoking recognition, etc. For example, when the original image is a raw image of the driver in the car, the driver phone call recognition can be performed on the R channel image of the raw image, and the driver's hand area and side face area (excluding the eye area, with the corner of the eye as the boundary of the region of interest) in the R channel image are used as the region of interest.

[0061] S3: Determine a target sensitive image region in the target single-channel image based on the region of interest in the target single-channel image.

[0062] In the disclosed embodiment, all image areas except the area of ​​interest in the target single-channel image can be used as target sensitive image areas, or specified areas other than the area of ​​interest in the native image (such as the driver's eye area and the facial areas of other passengers) can be used as target sensitive image areas.

[0063] S4: Determine a mask image of the target single-channel image based on the target sensitive image area.

[0064] Specifically, a mask image having the same size as the target single-channel image is provided as the mask image of the target single-channel image. In the mask image of the target single-channel image, the image area corresponding to the target sensitive image area is marked, for example, with a special value.

[0065] Please refer again Figure 2 In this example, the special tag uses a tag with a value of 1. The special tag is used in the subsequent data desensitization processing step. If a certain image block of the mask image is read with a special tag, it is decided to perform data desensitization processing on the data of the image block corresponding to the original image; if a certain image block of the mask image is read without a special tag, the data desensitization processing is not performed on the data of the image block corresponding to the original image.

[0066] In this example, the target sensitive area in the target single-channel image is: the image area corresponding to the value 1 in the mask image of the target single-channel image, for example, the target sensitive area in the R channel image includes the image area of ​​4 image blocks in the central part of the R channel image.

[0067] S5: Determine a mask image of the native image based on the mask image of the target single-channel image.

[0068] Please continue to refer to Figure 2 In this example, since there is a corresponding relationship between the target single-channel image and the native image, the mask image of the native image can be determined from the mask image of the target single-channel image based on the corresponding relationship.

[0069] Since the target single-channel image's mask image has markers set in it, there are also markers in the native image's mask image, which can be used to determine the image area in the native image that needs to be desensitized. In this example, the image area that needs to be desensitized is the image area in the native image that has a value of 1 in the target single-channel image's mask image.

[0070] S6: Based on the mask image of the native image, perform data desensitization processing on the native data of the native image.

[0071] Specifically, based on the marks in the mask image of the native image, the image area in the native image that needs to be subjected to data desensitization processing can be determined, and then the native data in the image area is subjected to data desensitization processing.

[0072] In this embodiment, the target single-channel image of the native image in the preset channel is first determined, and then the region of interest in the target single-channel image is determined based on the behavior recognition task, and then the target sensitive image region in the target single-channel image is determined based on the region of interest in the target single-channel image, and then the mask image of the target single-channel image is determined. After obtaining the mask image of the target single-channel image, a mask image of the native image is generated, and finally, based on the mask image of the native image, the native data of the native image is desensitized to ensure the data security of the remaining image regions in the native image except for the image regions that require behavior recognition, thereby improving the security of user privacy data.

[0073] Figure 3 FIG. 1 is a flow chart of step S1 in one embodiment of the present disclosure. Figure 3 As shown, step S1 includes:

[0074] S1-1: Based on array distribution information of the image sensor, determine a plurality of single-channel images corresponding to the native image, wherein image sizes of the plurality of single-channel images are smaller than an image size of the native image.

[0075] Please refer again Figure 2 The native image in the image sensor array distribution information is shown in the native image. The array distribution information includes the setting position of each channel, that is, the setting position of the R channel, the G channel, and the B channel. Based on the array distribution information of the native image, the R channel image, the G channel image, and the B channel image of the native image can be obtained.

[0076] S1-2: Based on multiple single-channel images, determine a target single-channel image.

[0077] Specifically, based on system settings, or based on the image quality of the R channel image, the G channel image, and the B channel image, one channel image can be selected from the R channel image, the G channel image, and the B channel image as the target single channel image, for example, the R channel image is selected as the target single channel image.

[0078] In this embodiment, a target single-channel image for target recognition can be quickly determined based on the array distribution information of the image sensor.

[0079] Figure 4 FIG. 1 is a flow chart of step S2 in one embodiment of the present disclosure. Figure 4 As shown, step S2 includes:

[0080] S2-1: Based on the behavior recognition model, determine the positions of multiple target feature points in the target single-channel image.

[0081] Specifically, the multiple target feature points are determined according to the behavior recognition task. For example, when the behavior recognition task is to recognize the driver making a phone call, the multiple target feature points include the driver's hand feature points and side face feature points. The multiple target feature points are extracted by a pre-trained behavior recognition model. The behavior recognition model is trained based on the single-channel image corresponding to the raw image of the sample person making a phone call and the single-channel image corresponding to the raw image of the sample person not making a phone call.

[0082] S2-2: Determine a region of interest in the target single-channel image based on the positions of multiple target feature points.

[0083] Specifically, a behavior recognition model is used to determine a region of interest in a target single-channel image based on positions of a plurality of target feature points in the target single-channel image.

[0084] In this embodiment, the region of interest in the target single-channel image can be quickly and accurately identified through the pre-trained behavior recognition model.

[0085] Figure 5 FIG. 4 is a flow chart of step S4 in one embodiment of the present disclosure. Figure 5 As shown, step S4 includes:

[0086] S4-1: Based on the position of the target sensitive image area on the target single-channel image, determine a first image area in the mask image of the target single-channel image that corresponds to the target sensitive image area. Figure 2 In , the first image region is the image region with a value of 1 in the mask image of the target single-channel image.

[0087] S4-2: Set a first value for each unit image block in the first image area. The first value may be determined according to the range of the color component of the target single-channel image. For example, when the range of the color component of the target single-channel image is 0 to 1, the first value may be 1. When the range of the color component of the target single-channel image is 0 to 255, the first value may also be 255.

[0088] S4-3: Determine the remaining image area after removing the first image area from the mask image of the target single-channel image as the second image area. Figure 2 In , the second image region is the image region with a value of 0 in the mask image of the target single-channel image.

[0089] S4-4: Setting a second numerical value for each unit image block in the second image area. The first numerical value is different from the second numerical value. The second numerical value may be determined according to the range of the color component of the target single-channel image. Exemplarily, when the range of the color component of the target single-channel image is 0 to 1, the second numerical value may be 0. When the range of the color component of the target single-channel image is 0 to 255, the second numerical value may be 0.

[0090] In this embodiment, a mask image of a target single-channel image including a first numerical value and a second numerical value is set, and image areas that need to be subjected to data desensitization processing and image areas that do not need to be subjected to data desensitization processing can be marked by different numerical values, so that a mask image of a native image can be generated according to the mask image of the target single-channel image in a subsequent step, and then the image areas in the native image that need to be subjected to data desensitization processing can be quickly determined according to the mask image of the native image.

[0091] In one embodiment of the present disclosure, step S5 specifically includes: based on the image size of the native image and the image size of the target single-channel image, upsampling the mask image of the target single-channel image to obtain the mask image of the native image.

[0092] Please refer again Figure 2 , when the resolution of the native image is 8x8 and the resolution of the target single-channel image is 4x4, Figure 2 The 8x8 resolution and 4x4 resolution shown are only for illustrative purposes, and the native image may be an image with a resolution of 1920x1280, an image with a resolution of 1024x768, or an image of other sizes. Based on the size relationship between the image size of the native image and the image size of the target single-channel image, the mask image of the target single-channel image is upsampled, for example, the mask image with a resolution of 4x4 is upsampled to a mask image with a resolution of 8x8, i.e., the mask image of the native image.

[0093] In this embodiment, based on the size relationship between the image size of the native image and the image size of the target single-channel image, as well as the mask image of the target single-channel image, the mask image of the native image can be quickly obtained, so that the subsequent steps of performing data desensitization processing on the image data of the native image based on the mask image of the native image are facilitated.

[0094] In one embodiment of the present disclosure, the upsampling process in step S5 includes at least one of nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, and adaptive image interpolation based on image edge gradient information. Among them, the nearest neighbor interpolation is also called zero-order interpolation, which makes the gray value of the transformed pixel equal to the gray value of the input pixel closest to it. Bilinear interpolation is to perform linear interpolation in two directions respectively, for example, first perform linear interpolation in the horizontal direction, and then perform linear interpolation in the vertical direction. The pixel value of a certain position can be finally obtained through two linear interpolations. Bicubic interpolation can obtain the pixel value of the pixel point by weighted averaging the pixel values ​​of the sixteen sampling points in the 4x4 neighborhood of the pixel point to be interpolated. Two polynomial interpolation cubic functions are required, one for each direction. Adaptive image interpolation based on image edge gradient information is to consider the weight information related to the distance when performing upsampling interpolation. At the same time, it is necessary to consider the weight information related to the edge gradient according to the gradient of the image. Along the edge direction of the image, the gradient value of the image is smaller; perpendicular to the edge direction of the image, the gradient value of the image is larger, that is, the interpolation weight of the pixel points along the edge direction is larger, and the weight value of the pixel points perpendicular to the edge direction is smaller, which can ensure better interpolation performance at the edge position of the image.

[0095] Figure 6 FIG. 1 is a flow chart of step S6 in one embodiment of the present disclosure. Figure 6 As shown, step S6 includes:

[0096] S6-1: Based on the mask image of the native image, determine the target sensitive image region in the native image, wherein the target sensitive image region in the native image corresponds to the position of the target sensitive image region in the target single-channel image.

[0097] Please refer again Figure 2 , for all second image regions whose values ​​in the mask image of the original image are 1, the corresponding regions in the original image are used as target sensitive image regions.

[0098] S6-2: Setting the pixel points in the target sensitive image area in the native image to a preset pixel value, wherein the difference between the preset pixel value and the pixel boundary value is within a preset difference range, for example, the preset difference range may be [0, 5].

[0099] In an example of the present disclosure, the pixel value range in the native image is 0 to 255, and the preset difference range is [0, 5]. At this time, the target pixel value can be 0, 1, 2, 3, 4, 251, 252, 253, 254 or 255.

[0100] In this embodiment, by setting the pixel points in the target sensitive image area in the native image to the target pixel value, the real user privacy data can be prevented from being reversely restored, and the user's privacy data can be prevented from being restored after the native image is restored to form an RGB true color image, thereby effectively protecting the user's privacy.

[0101] In one embodiment of the present disclosure, before step S1, the method further includes:

[0102] S0: When the preset image desensitization start conditions are met, proceed to step S1.

[0103] Specifically, the embodiment of the present disclosure is provided with two modes, one of which is a desensitization mode, and the other is a normal image processing mode, and the two modes can be switched.

[0104] The default state is the normal image processing mode, that is, when the image sensor captures the original image, it does not perform desensitization, but directly restores the original image to obtain a three-channel image, and then stores the three-channel image, or provides the three-channel image to a preset remote terminal, such as a mobile phone bound to the user, or a designated server.

[0105] When the preset image desensitization start conditions are met, the process proceeds to steps S1-S6 to desensitize the image data.

[0106] Taking the scene of collecting images inside a car as an example, if at least one of the following preset image desensitization start conditions is met, steps S1-S6 are entered to desensitize the image data:

[0107] 1. Detect vehicle start;

[0108] 2. The pressure sensor detects that the pressure on the seat of the vehicle is greater than a preset pressure threshold;

[0109] 3. Detect that the smart device in the car is turned on, for example, detect that the display in front of a seat in the car is turned on;

[0110] 4. Detect that an external device is connected to the car's plug, for example, detect that a mobile phone is connected to the car's USB plug;

[0111] 5. Check and detect that the in-car network is connected to a terminal via Bluetooth connection, for example, the in-car network is connected to a mobile phone via Bluetooth connection;

[0112] 6. Detecting that someone is in the car, for example, detecting that someone is in the car through an infrared detection device, an in-car audio device, or an in-car radar.

[0113] In this embodiment, when the preset image desensitization start condition is met, the mode is switched to the desensitization mode, and then the native data of the native image collected by the image sensor is subjected to data desensitization processing. When desensitization is not needed, the normal image processing mode is maintained, and the native image can be converted into a three-channel image, and normal image processing is performed to achieve monitoring inside the vehicle when no one is in the vehicle. This embodiment can selectively decide whether to perform data desensitization processing on the image data, so as to meet the different needs of users.

[0114] Any image data desensitization method provided in the embodiments of the present disclosure may be executed by any appropriate device with data processing capabilities, including but not limited to: a terminal device and a server, etc. Alternatively, any image data desensitization method provided in the embodiments of the present disclosure may be executed by a processor, such as the processor executing any image data desensitization method mentioned in the embodiments of the present disclosure by calling corresponding instructions stored in a memory. This will not be described in detail below.

[0115] Exemplary Devices

[0116] Figure 7 : is a structural block diagram of an image data desensitization device in one embodiment of the present disclosure. Figure 7 As shown, the image data desensitization device includes: a single-channel image determination module 100, an area of ​​interest determination module 200, a sensitive image area determination module 300, a first mask image determination module 400, a second mask image determination module 500 and a data desensitization module 600.

[0117] Among them, the single-channel image determination module 100 is used to determine the target single-channel image of the native image in a preset channel. The region of interest determination module 200 is used to determine the region of interest in the target single-channel image based on the behavior recognition task. The sensitive image area determination module 300 is used to determine the target sensitive image area in the target single-channel image based on the region of interest in the target single-channel image. The first mask image determination module 400 is used to determine the mask image of the target single-channel image based on the target sensitive image area. The second mask image determination module 500 is used to determine the mask image of the native image based on the mask image of the target single-channel image. The data desensitization module 600 is used to perform data desensitization processing on the native data of the native image based on the mask image of the native image.

[0118] Figure 8 FIG. 1 is a structural block diagram of a single-channel image determination module 100 in one embodiment of the present disclosure. Figure 8 As shown, the single-channel image determination module 100 includes:

[0119] A first determining unit 101 is configured to determine a plurality of single-channel images corresponding to the native image based on array distribution information of the image sensor, wherein the image sizes of the plurality of single-channel images are smaller than the image size of the native image;

[0120] The second determining unit 102 is configured to determine the target single-channel image based on the multiple single-channel images.

[0121] Fig. 9 FIG. 2 is a block diagram of a region of interest determination module 200 in one embodiment of the present disclosure. Fig. 9 As shown, the region of interest determination module 200 includes:

[0122] A feature point positioning unit 201 is used to determine the positions of multiple target feature points in the target single-channel image based on a behavior recognition model;

[0123] The region of interest determining unit 202 is configured to determine the region of interest in the target single-channel image based on the positions of the multiple target feature points.

[0124] Fig.10 FIG. 4 is a structural block diagram of the first mask image determination module 400 in one embodiment of the present disclosure. Fig.10 As shown, the first mask image determination module 400 includes:

[0125] A first image region determining unit 401 is configured to determine, based on a position of the target sensitive image region on the target single-channel image, a first image region in the mask image of the target single-channel image corresponding to the target sensitive image region;

[0126] A first value setting unit 402, configured to set a first value for each unit image block in the first image area;

[0127] A second image region determining unit 403 is used to determine a remaining image region after removing the first image region from the mask image of the target single-channel image as a second image region;

[0128] The second value setting unit 404 is configured to set a second value for each unit image block in the second image area; wherein the first value is different from the second value.

[0129] In one embodiment of the present disclosure, the second mask image determination module 500 is used to upsample the mask image of the target single-channel image based on the image size of the native image and the image size of the target single-channel image to obtain the mask image of the native image.

[0130] In one embodiment of the present disclosure, the upsampling process includes at least one of a nearest neighbor interpolation process, a bilinear interpolation process, a bicubic interpolation process, and an adaptive image interpolation process based on image edge gradient information.

[0131] Fig.11 FIG. 6 is a structural block diagram of a data desensitization module 600 in one embodiment of the present disclosure. Fig.11 As shown, the data desensitization module 600 includes:

[0132] A target sensitive image region determining unit 601 is used to determine a target sensitive image region in the native image based on the mask image of the native image, wherein the target sensitive image region in the native image corresponds to a position of the target sensitive image region in the target single-channel image;

[0133] The pixel value setting unit 602 is used to set the pixel points in the target sensitive image area in the native image to a preset pixel value, wherein the difference between the preset pixel value and the pixel boundary value is within a preset difference range.

[0134] Fig.12 FIG. 1 is a structural block diagram of an image data desensitization device in another embodiment of the present disclosure. Fig.12 As shown, the image data desensitization device also includes:

[0135] The detection module 700 is used to desensitize the image data through the single-channel image determination module 100, the region of interest determination module 200, the sensitive image region determination module 300, the first mask image determination module 400, the second mask image determination module 500 and the data desensitization module 600 when it is detected that the preset image desensitization start conditions are met.

[0136] It should be noted that the specific implementation of the image data desensitization device of the embodiment of the present disclosure is similar to the specific implementation of the image data desensitization method of the embodiment of the present disclosure. Please refer to the part of the image data desensitization method for details. In order to reduce redundancy, it will not be described in detail.

[0137] Exemplary Electronic Devices

[0138] Below, reference Fig.13 To describe the electronic device according to the embodiment of the present disclosure. Fig.13 As shown, the electronic device includes one or more processors 131 and a memory 132 .

[0139] The processor 131 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0140] The memory 132 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 131 may run the program instructions to implement the desensitization method of image data of the various embodiments of the present disclosure described above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0141] In one example, the electronic device may further include: an input device 133 and an output device 134, which are interconnected via a bus system and / or other forms of connection mechanisms (not shown). The input device 133 may be, for example, a keyboard, a mouse, etc. The output device 134 may include, for example, a display, a speaker, a printer, and a communication network and a remote output device connected thereto, etc.

[0142] Of course, to simplify, Fig.13 Only some of the components related to the present disclosure in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device may further include any other appropriate components.

[0143] Exemplary computer-readable storage media

[0144] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0145] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0146] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0147] The block diagrams of the devices, apparatuses, equipment, and systems involved in this disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including," "comprising," "having," and the like are open words, referring to "including but not limited to," and can be used interchangeably therewith. The words "or" and "and" used herein refer to the words "and / or," and can be used interchangeably therewith, unless the context clearly indicates otherwise. The word "such as" used herein refers to the phrase "such as but not limited to," and can be used interchangeably therewith.

[0148] The method and apparatus of the present disclosure may be implemented in many ways. For example, the method and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.

[0149] It should also be noted that in the apparatus, device and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0150] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0151] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for desensitizing image data. include: Determine the target single-channel image of the native image in the preset channel; Determine a region of interest in the target single-channel image based on a behavior recognition task; Determining a target sensitive image area in the target single-channel image based on a region of interest in the target single-channel image; Based on the target sensitive image area, determining a mask image of the target single-channel image; Determining a mask image of the native image based on the mask image of the target single-channel image; Based on the mask image of the native image, data desensitization processing is performed on the native data of the native image.

2. The method for desensitizing image data according to claim 1, in, Determining the target single-channel image of the native image in the preset channel includes: Determining a plurality of single-channel images corresponding to the native image based on array distribution information of the image sensor, wherein image sizes of the plurality of single-channel images are smaller than image size of the native image; Based on the multiple single-channel images, the target single-channel image is determined.

3. The method for desensitizing image data according to claim 1, in, The determining of the region of interest in the target single-channel image based on the behavior recognition target includes: Based on the behavior recognition model, determining the positions of a plurality of target feature points in the target single-channel image; Based on the positions of the multiple target feature points, a region of interest in the target single-channel image is determined.

4. The method for desensitizing image data according to claim 1, in, The step of determining the mask image of the target single-channel image based on the target sensitive image area includes: Based on the position of the target sensitive image area on the target single-channel image, determining a first image area in the mask image of the target single-channel image corresponding to the target sensitive image area; Setting a first value for each unit image block in the first image area; Determine the remaining image area after removing the first image area from the mask image of the target single-channel image as the second image area; Setting a second value for each unit image block in the second image area; The first value is different from the second value.

5. The method for desensitizing image data according to claim 2, in, The determining the mask image of the native image based on the mask image of the target single-channel image includes: Based on the image size of the native image and the image size of the target single-channel image, upsampling processing is performed on the mask image of the target single-channel image to obtain the mask image of the native image.

6. The method for desensitizing image data according to claim 1, in, The step of performing data desensitization processing on native data of the native image based on the mask image of the native image includes: Determining a target sensitive image region in the native image based on the mask image of the native image, wherein the target sensitive image region in the native image corresponds to a position of the target sensitive image region in the target single-channel image; The pixel points in the target sensitive image area in the native image are set to a preset pixel value, wherein the difference between the preset pixel value and the pixel boundary value is within a preset difference range.

7. The method for desensitizing image data according to claim 1, in, Before determining the target single-channel image of the native image in the preset channel, the method further includes: When the preset image desensitization start condition is met, the step of determining the target single-channel image of the native image in the preset channel is entered.

8. A desensitizing device for image data, include: A single-channel image determination module, used to determine a target single-channel image of a native image in a preset channel; A region of interest determination module, used to determine the region of interest in the target single-channel image based on a behavior recognition task; A sensitive image region determination module, configured to determine a target sensitive image region in the target single-channel image based on a region of interest in the target single-channel image; A first mask image determination module, configured to determine a mask image of the target single-channel image based on the target sensitive image area; a second mask image determination module, configured to determine the mask image of the native image based on the mask image of the target single-channel image; The data desensitization module is used to perform data desensitization processing on the native data of the native image based on the mask image of the native image.

9. A computer-readable storage medium, wherein the storage medium stores a computer program, wherein the computer program is used to execute the image data desensitization method described in any one of claims 1 to 7.

10. An electronic device, wherein the electronic device include: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the image data desensitization method described in any one of claims 1-7.

Images