Image data masking method and apparatus

Abstract

The embodiment of the present disclosure discloses an image data masking method and device, in which the masking method includes the steps of: determining a target image channel in an original image that needs to be subjected to target recognition; performing image restoration on original data corresponding to the target image channel to obtain a single-channel image corresponding to a color component of the target image channel; recognizing a target object from the single-channel image; determining a sensitive region in the original image according to the target object; and performing data masking processing on the original data in the sensitive region. The embodiment of the present disclosure can reliably prevent private image data from being stored and leaked in the subsequent digital image signal processing flow, protect private data at the source of image data generation, and improve data security.

Description

[Technical Field]

[0001] This disclosure claims priority to a Chinese patent application filed on September 30, 2021, bearing application number 202111161187.X and entitled "Method and apparatus for masking image data," the entire contents of which are incorporated herein by reference.

[0002] The present disclosure relates to computer vision technology, and more particularly to a method and apparatus for masking image data. [Background technology]

[0003] With the development of intelligent recognition technology, data privacy issues have attracted more and more attention. How to ensure the security of private data in collected images is an urgent issue that needs to be resolved. Summary of the Invention [Problem to be solved by the invention]

[0004] To solve the above technical problems, the present disclosure proposes: The embodiments of the present disclosure provide a method and apparatus for masking image data. [Means for solving the problem]

[0005] A method for masking image data according to a first aspect of an embodiment of the present disclosure includes: determining a target image channel in which target recognition in the original image needs to be performed; performing image restoration on the original data corresponding to the target image channel to obtain a single-channel image corresponding to the color components of the target image channel; Recognizing a target object from the single-channel image; determining a sensitive region in the original image based on the target object; and performing a data masking process on the original data in the sensitive area.

[0006] A masking device for image data according to a second aspect of an embodiment of the present disclosure includes: an image channel determination module for determining a target image channel in which target recognition in the original image needs to be performed; a single-channel image acquisition module for performing image restoration on original data corresponding to the target image channel to obtain a single-channel image corresponding to the color components of the target image channel; an image recognition module for recognizing a target object from the single-channel image; a sensitive region determination module for determining a sensitive region in the original image based on the target object; and a data masking module for performing data masking processing on the original data in the sensitive region.

[0007] A third aspect of the present disclosure provides a computer-readable storage medium storing a computer program for executing the image data masking method according to the first aspect.

[0008] An electronic device according to a fourth aspect of the present disclosure includes: a processor; a memory for storing instructions executable by the processor; The processor is adapted to read the executable instructions from the memory and execute the instructions to implement the method for masking image data according to the first aspect above. [Effects of the Invention]

[0009] Based on the image data masking method and apparatus according to the above embodiments of the present disclosure, first determine the target image channel that needs to be subjected to target recognition in the original image, then generate a single-channel image for the original data of the target image channel, then recognize the target object in the single-channel image, and further determine the sensitive area in the original image based on the recognized target object, and finally perform data masking processing on the original data in the sensitive area, so as to reliably prevent the storage and leakage of private image data in the subsequent digital image signal processing flow, protect the private data at the source of the image data, and improve data security.

[0010] The technical solutions of the present disclosure are further described in detail below with reference to figures and examples. [Brief explanation of the drawings]

[0011] The above and other objects, features, and advantages of the present disclosure will become more apparent when the embodiments of the present disclosure are described in more detail with reference to the drawings. The drawings are used to further understand the embodiments of the present disclosure, are a part of the specification, and are used to interpret the present disclosure together with the embodiments of the present disclosure, but do not limit the present disclosure. In the drawings, the same reference numerals generally indicate the same components or steps. [Figure 1] 1 is a flowchart of a method for masking image data according to an embodiment of the present disclosure. [Figure 2] FIG. 1 is a schematic diagram of determining a sensitive region in an original image from an original image in an example of the present disclosure; [Figure 3] 10 is a flowchart of step S1 in one embodiment of the present disclosure. [Figure 4] 10 is a flowchart of step S1 in another embodiment of the present disclosure. [Figure 5] FIG. 10 is a schematic diagram illustrating an image interpolation process performed on an original image to obtain a single-channel image in another example of the present disclosure. [Figure 6]FIG. 6 is a schematic diagram of determining sensitive regions in an original image locally based on a single-channel image in an example corresponding to FIG. 5; [Figure 7] FIG. 1 is a block diagram of the structure of an image data masking device according to an embodiment of the present disclosure. [Figure 8] FIG. 1 is a block diagram of the structure of an image channel determination module 100 in one embodiment of the present disclosure. [Figure 9] FIG. 10 is a block diagram of the structure of an image channel determination module 100 in another embodiment of the present disclosure. [Figure 10] 1 is a structural diagram of an electronic device according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, exemplary embodiments according to the present disclosure will be described in detail with reference to the drawings. Obviously, the described embodiments are only some embodiments of the present disclosure, but not all embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described herein.

[0013] It should be noted that the relative arrangement of components and steps, numerical expressions and values ​​set forth in these examples do not limit the scope of the present disclosure unless specifically stated otherwise.

[0014] As will be understood by those skilled in the art, the terms "first", "second", etc. in the embodiments of the present disclosure are used to distinguish different steps, devices, modules, etc., and do not represent any particular technical meaning or a necessary logical order between them.

[0015] It should also be understood that in the embodiments of the present disclosure, "plurality" can refer to two or more than two, and "at least one" can refer to one, two, or more than two.

[0016] Embodiments of the present disclosure may be applied to electronic devices such as terminal devices, computer systems, servers, and the like, and may operate with many other general-purpose or special-purpose computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations suitable for use in conjunction with electronic devices such as terminal devices, computer systems, servers, and the like include, but are not limited to, personal computer systems, server computer systems, thin clients, fat clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, networked personal computers, small computer systems, large computer systems, and distributed cloud computing technology environments that include any of the above systems.

[0017] Exemplary Methods 1 is a flowchart of an image data masking method according to an embodiment of the present disclosure. This embodiment can be applied to electronic devices, and as shown in FIG. 1, it includes the following steps:

[0018] In S1, the target image channel in which target recognition in the original image needs to be performed is determined.

[0019] In an embodiment of the present disclosure, the original image is collected by an image sensor. For example, the original image may be a raw image of a driver that has not been restored and collected by an image sensor inside the vehicle during driver assistance or autonomous driving of the vehicle, or a raw image of a passenger that has not been restored and collected by an image sensor inside the vehicle.

[0020] After acquiring the original image, the target image channel that needs to perform target recognition in the original image is determined by the vehicle control system or a terminal (e.g., a mobile phone or a server) that can connect and control the vehicle.

[0021] 2 is a schematic diagram of determining a sensitive region in an original image from an original image in an example of the present disclosure. As shown in FIG. 2, in this example, the resolution of the original image is 8×8, and the resolution of the single-channel image obtained after channel separation can be 4×4 (for example, the resolution of the R channel image is 4×4). In practice, the resolution of the original image is determined by the resolution of the image sensor. The 8×8 resolution and 4×4 resolution shown in FIG. 2 are only for general explanation and cannot form a limitation on the present disclosure. The embodiments of the present disclosure do not limit the specific sizes of the original image and the mask image. For example, the original image can adopt an image with a resolution of 1920×1280, an image with a resolution of 1024×768, or an image with other sizes.

[0022] The target image channel may be any one of the R channel, G channel, and B channel, and is specifically determined based on a user setting, or the target image channel is determined based on the image quality of the R channel image, G channel image, and B channel image.

[0023] In the following examples, image data masking is performed using a vehicle control system as an example, but as will be understood by those skilled in the art, image data masking can also be performed using a terminal that can be connected to and controlled by the vehicle.

[0024] In S2, image restoration is performed on the original data corresponding to the target image channel to obtain a single-channel image corresponding to the color component of the target image channel. For example, if the target image channel is the R channel, image restoration is performed on the original data corresponding to the R channel to obtain a single-channel image whose color component is R, i.e., an R channel image.

[0025] In S3, the target object is recognized from the single-channel image.

[0026] Specifically, a target object in a single-channel image is determined by image recognition technology, where the target object is determined based on an action recognition task. For example, if the original image is a raw image of a driver in a car and the action recognition task is recognizing the driver by making a phone call, the driver's profile face region and hand region in the R-channel image can be determined as the target object.

[0027] At S4, sensitive regions in the original image are determined based on the target object.

[0028] Specifically, first, the position of the target object in the single-channel image is determined. For example, in the example shown in Figure 2, the position of the target object in the single-channel image includes the image area of ​​the central 4-grid of the single-channel image. Then, based on the correspondence between the single-channel image and the original image, the sensitive area in the original image can be determined. For example, in the example shown in Figure 2, the sensitive area in the original image is the image area of ​​the central 16-grid of the original image.

[0029] In S5, the original data in the sensitive region is subjected to a data masking process, where the data masking process method can include adjusting the pixel values ​​of the pixels in the sensitive region, or performing an image blurring process on the sensitive region, etc.

[0030] In this embodiment, first, a target image channel that needs to be subjected to target recognition in the original image is determined, then a single-channel image is generated from the original data of the target image channel, then the target object is recognized in the single-channel image, and then a sensitive area in the original image is determined based on the recognized target object, and finally a data masking process is performed on the original data in the sensitive area, so that the storage and leakage of private image data can be reliably prevented in the subsequent digital image signal processing flow, private data is protected at the source of image data generation, and data security is improved.

[0031] 3 is a flowchart of step S1 in one embodiment of the present disclosure. As shown in FIG. 3, step S1 includes:

[0032] In step S1-A-1, a plurality of single-channel images corresponding to the original image are determined based on the array distribution information of the image sensor. The array distribution information includes the setting positions of each channel, i.e., the setting positions of the R channel, the G channel, and the B channel. Based on the array distribution information of the original image, the R channel image, the G channel image, and the B channel image corresponding to the original image can be obtained.

[0033] Referring back to the original image in FIG. 2, the original image shows the array distribution information of the image sensor. Based on the array distribution information, an R channel image, a G channel image, and a B channel image corresponding to the original image can be obtained. For example, the resolution of the R channel image is 4×4, and the resolution of the original image is 8×8. In this example, the image size of the R channel image is smaller than the image size of the original image.

[0034] In S1-A-2, a target image channel is determined based on the plurality of single-channel images.

[0035] Specifically, one image channel can be selected as the target image channel based on system settings or based on the image quality of the R channel image, G channel image, and B channel image, for example, the R channel is selected as the target image channel.

[0036] In this embodiment, the advantage of performing target recognition directly based on the single-channel image data of the original image is that there is no need to interpolate to obtain a full-resolution image in an interpolation step, which greatly reduces the implementation complexity of the algorithm.

[0037] 4 is a flowchart of step S1 in another embodiment of the present disclosure. As shown in FIG. 4, step S1 includes:

[0038] In S1-B-1, an image interpolation process is performed on the original image to determine a plurality of single-channel images corresponding to the original image, where the image sizes of the plurality of single-channel images are all the same as the image size of the original image.

[0039] 5 is a schematic diagram illustrating an example of performing image interpolation processing on an original image to obtain a single-channel image in another example of the present disclosure. As shown in FIG. 5, in this example, the resolution of the original image is 8×8, and the resolution of the multiple single-channel images after the image interpolation processing is also 8×8. For example, in this example, the resolution of the R channel image is 8×8, the resolution of the G channel image is 8×8, the resolution of the B channel image is 8×8, and the resolution of the original image is also 8×8. In this example, the image sizes of the R channel image, the G channel image, and the B channel image are all equal to the image size of the original image.

[0040] Note that the 8x8 resolution shown in Figure 5 is only a general description, and the actual original image and the single-channel image after image interpolation processing can be an image with a resolution of 1920x1280, an image with a resolution of 1024x768, or an image with other sizes.

[0041] In the embodiments of the present disclosure, image interpolation methods include nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, and adaptive image interpolation based on image edge gradient information. Nearest neighbor interpolation, also known as zero-order interpolation, equalizes the grayscale value of a converted pixel to the grayscale value of the nearest input pixel. Bilinear interpolation performs linear interpolation in two directions, for example, first performing one linear interpolation in the horizontal direction and then another linear interpolation in the vertical direction, thereby obtaining the final pixel value at a certain position through two linear interpolations. Bicubic interpolation is obtained by performing a weighted average of 16 sampling points within a 4x4 neighborhood of the interpolated pixel, and requires the use of two polynomials to interpolate a cubic function, one for each direction. The adaptive image interpolation method based on image edge gradient information, in addition to considering distance-related weight information, also needs to consider edge gradient-related weight information based on the image gradient when performing upsampling interpolation. The image gradient value is relatively small along the image edge direction, and relatively large along the image edge direction, i.e., the pixel interpolation weight along the edge direction is larger and the pixel interpolation weight value perpendicular to the edge direction is smaller, which can ensure a better interpolation representation of the image edge position.

[0042] In S1-B-2, a target image channel is determined based on the plurality of single-channel images.

[0043] Specifically, the target image channel may be any of the R channel, G channel, and B channel, and is specifically determined based on a user setting or based on the image quality of the R channel image, G channel image, and B channel image.

[0044] Fig. 6 is a schematic diagram of determining a sensitive region in an original image based on a single-channel image in an example corresponding to Fig. 5. As shown in Fig. 6, when the image size of the single-channel image and the image size of the original image are the same, the position of the target object in the single-channel image includes the image area of ​​the central 16 grids of the single-channel image, and the sensitive region in the original image is also the image area of ​​the central 16 grids of the original image, and the positions of the two image areas correspond to each other.

[0045] In this embodiment, the original image is interpolated to obtain the three RGB channels. A full-resolution image is obtained. For example, for an original image with a resolution of 8x8, the resolution of the single-channel image obtained by the image interpolation process shown in Figure 5 is 8x8, while the resolution of the single-channel image obtained by the channel separation process shown in Figure 2 is 4x4. Therefore, the single-channel image obtained by the image interpolation process has a higher resolution and contains more comprehensive information than the single-channel image obtained by performing channel separation on the original image, which is more advantageous for the subsequent steps to recognize sensitive information.

[0046] In one embodiment of the present disclosure, step S3 specifically includes recognizing the target object from the single-channel image by a pre-trained recognition model.

[0047] In this embodiment, the recognition model is trained in the following manner.

[0048] Based on the initial model, downsampling is performed from the sample single-channel image using a full-layer convolutional network.

[0049] Then, after downsampling, upsampling is performed based on the initial model to bring the image size back to the same as the sample single-channel image.

[0050] Next, based on the initial model, a pixel-by-pixel prediction is performed on the sample single-channel image, and backpropagation is performed based on the difference between the prediction result and the classification label of the sample single-channel image to update the parameters of the initial model. After the iteration stopping condition is met, a final recognition model is obtained.

[0051] A single-channel image that needs to undergo target recognition is taken as the input of a recognition model, which can recognize target objects in the single-channel image.

[0052] In this embodiment, the pre-trained recognition model can quickly and accurately recognize the target object in the single-channel image, which helps in the subsequent steps of determining the sensitive areas in the original image and performing data masking processing on the original data in the sensitive areas.

[0053] In one embodiment of the present disclosure, step S5 includes setting a pixel in the sensitive region as a target pixel value, where a difference value between the target pixel value and the pixel boundary value is within a preset difference value range, for example, the preset difference value range can be [0, 5].

[0054] In one example of the present disclosure, if the pixel values ​​in the original image range from 0 to 255, the preset difference value range is [0, 5], and the target pixel value at this time may be 0, 1, 2, 3, 4, 251, 252, 253, 254, or 255.

[0055] In this embodiment, by setting the pixel in the target sensitive image area in the original image as the target pixel value, it is possible to prevent the real user's private data from being reverse restored, and after the original image is restored to form an RGB true color image, the user's private data is also prevented from being restored, thereby effectively protecting the user's privacy.

[0056] In another embodiment of the present disclosure, step S5 includes performing image blurring on the sensitive region, for example, employing Gaussian blurring.

[0057] In this embodiment, the image blurring process is performed on the sensitive areas of the original image, thereby effectively protecting the user's privacy.

[0058] Furthermore, a preset convolution kernel is used to blur the sensitive region of the original image. The size of the preset convolution kernel and the weight distribution of the preset convolution kernel can be determined based on the confidentiality level type of the target corresponding to the sensitive region. If the confidentiality level type indicator of the target indicates that the target (e.g., the driver's eye region in the original image) requires high confidentiality, the resolution of the convolution kernel can be set to be large, and the weight of the convolution kernel needs to be set to have a higher blur strength. For example, the size of the preset convolution kernel can be 21×21. Therefore, it is possible to avoid the object in the sensitive area from being restored later. If the object's confidentiality level type indicator indicates that the object density (for example, the area of ​​the driver's forehead in the original image) is low, the resolution of the convolution kernel can be set small, and the weight of the convolution kernel needs to be set to have a relatively weak blurring strength. For example, the preset convolution kernel size can be 5×5. By blurring the image using a small convolution kernel size, the complexity of the image blurring process can be reduced and the efficiency of the image blurring process can be greatly improved.

[0059] In this embodiment, the image blurring process is performed on the sensitive area of ​​the original image using a preset convolution kernel, thereby effectively protecting the user's privacy.

[0060] Any image data masking method according to the embodiments of the present disclosure may be executed by any suitable device having data processing capabilities, including, but not limited to, a terminal device, a server, etc. Any image data masking method according to the embodiments of the present disclosure may also be executed by a processor, for example, the processor executes any image data masking method according to the embodiments of the present disclosure by calling corresponding instructions stored in a memory. Further description will be omitted below.

[0061] Exemplary Apparatus 7 is a block diagram of the structure of an image data masking device according to an embodiment of the present disclosure. As shown in FIG. 7, the image data masking device according to an embodiment of the present disclosure includes an image channel determination module 100, a single-channel image acquisition module 200, an image recognition module 300, a sensitive region determination module 400, and a data masking module 500.

[0062] Here, the image channel determination module 100 is used to determine a target image channel that needs to be subjected to target recognition in the original image, the single channel image acquisition module 200 is used to perform image restoration on original data corresponding to the target image channel to obtain a single channel image corresponding to the color components of the target image channel, the image recognition module 300 is used to recognize a target object from the single channel image, the sensitive region determination module 400 is used to determine a sensitive region in the original image based on the target object, and the data masking module 500 is used to perform data masking processing on the original data in the sensitive region.

[0063] 8 is a block diagram of the structure of the image channel determination module 100 in one embodiment of the present disclosure. As shown in FIG. 8, the image channel determination module 100 includes: a first determination unit 101 for determining a plurality of single-channel images corresponding to the original image according to array distribution information of an image sensor, wherein the image sizes of the plurality of single-channel images are all smaller than the image size of the original image; a second determining unit 102 for determining the target image channel based on the plurality of single-channel images.

[0064] 9 is a block diagram of the structure of an image channel determination module 100 in another embodiment of the present disclosure. As shown in FIG. 9, the image channel determination module 100 includes: a third determination unit 103 for performing image interpolation processing on the original image to determine a plurality of single-channel images corresponding to the original image, wherein the image sizes of the plurality of single-channel images are all the same as the image size of the original image; a fourth determining unit 104 for determining the target image channel based on the plurality of single-channel images.

[0065] In one embodiment of the present disclosure, the image recognition module 300 recognizes target objects from the single-channel image through a pre-trained recognition model.

[0066] In one embodiment of the present disclosure, the data masking module 500 is used to set pixels in the sensitive region as a target pixel value, where a difference value between the target pixel value and a pixel boundary value is within a preset difference value range.

[0067] In one embodiment of the present disclosure, the data masking module 500 is used to perform image blurring on the sensitive areas.

[0068] In one embodiment of the present disclosure, the data masking module 500 is used to perform image blurring on the sensitive region of the original image with a preset convolution kernel.

[0069] It should be noted that the specific embodiment of the image data masking device in the embodiment of the present disclosure is similar to the specific embodiment of the image data masking method in the embodiment of the present disclosure, and specifically, reference can be made to the image data masking method section, and the description will be omitted to reduce redundancy.

[0070] Exemplary Electronic Devices Hereinafter, an electronic device according to an embodiment of the present disclosure will be described with reference to Fig. 10. As shown in Fig. 10, the electronic device includes one or more processors 110 and a memory 120.

[0071] The processor 110 may be a central processing unit (CPU) or other form of processing unit having data processing and / or instruction execution capabilities, and may also control other components in the electronic device to perform desired functions.

[0072] The memory 120 may include one or more computer program products, which may include various types of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored in the computer-readable storage media, and the processor 110 may execute the program instructions to implement the image data masking method of each embodiment of the present disclosure and / or other desired functions. Various types of content, such as input signals, signal components, and noise components, may also be stored in the computer-readable storage media.

[0073] In one example, the electronic device may further include input device(s) 130 and output device(s) 140, which are connected to each other by a bus system and / or other form of connection mechanism (not shown). The input device(s) 130 may be, for example, a keyboard or a mouse. The output device(s) 140 may include, for example, a display, speakers, a printer, a communication network and remote output devices connected thereto, etc.

[0074] 10, for simplicity, only some of the components of the electronic device that are relevant to the present disclosure are shown, and components such as a bus, an input / output interface, etc. The electronic device may further include any other appropriate components depending on a specific application.

[0075] Exemplary Computer Program Products and Computer-Readable Storage Media In addition to the above methods and apparatus, an embodiment of the present disclosure may also be a computer program product including computer program instructions that, when executed by a processor, cause the processor to perform the steps in the image data masking method of each embodiment of the present disclosure described in the "Exemplary Methods" section above of this specification.

[0076] The computer program product may have program code for carrying out operations of embodiments of the present disclosure written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may execute entirely on a user's computing device, partially on a user's device, as separate software packages, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0077] An embodiment of the present disclosure may also be a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, cause the processor to perform steps in the image data masking method of each embodiment described in the "Exemplary Method" section above of the present disclosure.

[0078] The computer-readable storage medium may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include an electrical connection having one or more conductors, a mobile hard drive, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0079] Although the basic principles of the present disclosure have been described above with reference to specific embodiments, the benefits, advantages, effects, etc. mentioned in the present disclosure are not limited but merely illustrative, and these benefits, advantages, effects, etc. do not necessarily exist in each embodiment of the present disclosure. Furthermore, the specific details disclosed above are not limited but merely serve to serve as examples and to facilitate understanding, and the above details do not necessarily limit the present disclosure to be realized by the above specific details.

[0080] Block diagrams of devices, apparatus, instruments, and systems according to the present disclosure are merely illustrative examples and are not intended to require or imply that they be connected, arranged, or configured in the manner shown in the block diagrams. Those skilled in the art will recognize that these devices, apparatus, instruments, and systems can be connected, arranged, or configured in any manner. For example, words such as "comprise," "contain," and "have" are open-ended terms and can be used interchangeably to mean "including but not limited to." As used herein, "or" and "and" mean and can be used interchangeably with "and / or" unless the context clearly dictates otherwise. As used herein, "for example," means and can be used interchangeably with "such as, but not limited to."

[0081] It should also be noted that in the devices, apparatuses, and methods of the present disclosure, each component or step may be separated and / or recombined, and such separation and / or recombination should be considered as equivalent means of the present disclosure.

[0082] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the disclosure. Various modifications to these aspects will be 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 disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0083] The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present disclosure to the form disclosed herein. While several example forms and embodiments have been discussed above, those skilled in the art may recognize certain variations, modifications, variations, additions, and subcombinations thereof.

Claims

1. determining a target image channel in which target recognition in the original image needs to be performed; performing image restoration on the original data corresponding to the target image channel to obtain a single-channel image corresponding to the color components of the target image channel; recognizing a target object from the single-channel image using a pre-trained recognition model; determining a sensitive region in the original image based on the target object; performing a data masking process on the original data corresponding to the sensitive region; The step of determining the target image channel in which target recognition in the original image needs to be performed includes: determining a plurality of single-channel images corresponding to the original image based on array distribution information of image sensors, wherein the image sizes of the plurality of single-channel images are all smaller than the image size of the original image; determining the target image channel from the plurality of single-channel images based on system settings or image quality of the plurality of single-channel images; determining the sensitive region in the original image based on the target object, determining a position of the target object in the single-channel image; determining the sensitive region in the original image based on a correspondence between the single-channel image and the original image; The sensitive area is an area of ​​the original image that is related to a user's privacy.

2. The step of determining a target image channel in which target recognition in the original image needs to be performed includes: performing an image interpolation process on the original image to determine a plurality of single-channel images corresponding to the original image, wherein the image sizes of the plurality of single-channel images are all the same as the image size of the original image; 2. The method of claim 1, further comprising: determining the target image channel from the plurality of single-channel images based on a system setting or an image quality of the plurality of single-channel images.

3. The step of performing a data masking process on the original data in the sensitive region includes:

2. The image data masking method according to claim 1, further comprising a step of setting a pixel value of a pixel in the sensitive region as a target pixel value, wherein a difference value between the target pixel value and a boundary value of a pixel value range is within a predetermined difference value range.

4. The step of performing a data masking process on the original data in the sensitive region includes: The method of claim 1 , further comprising the step of performing an image blurring process on the sensitive area.

5. The step of performing image blurring on the sensitive region includes:

5. The method for masking image data according to claim 4, further comprising the step of performing an image blurring process on the sensitive region of the original image with a preset convolution kernel.

6. an image channel determination module for determining a target image channel in which target recognition in the original image needs to be performed; a single-channel image acquisition module for performing image restoration on original data corresponding to the target image channel to obtain a single-channel image corresponding to the color components of the target image channel; an image recognition module for recognizing a target object from the single-channel image using a pre-trained recognition model; a sensitive region determination module for determining a sensitive region in the original image based on the target object; a data masking module for performing a data masking process on the original data corresponding to the sensitive region; The image channel determination module determines the target image channel in which target recognition in the original image is required, determining a plurality of single-channel images corresponding to the original image based on array distribution information of image sensors, wherein the image sizes of the plurality of single-channel images are all smaller than the image size of the original image; determining the target image channel from the plurality of single-channel images based on a system setting or an image quality of the plurality of single-channel images; The sensitive region determining module determines the sensitive region in the original image based on the target object, determining a position of the target object in the single-channel image; determining the sensitive region in the original image based on a correspondence between the single-channel image and the original image; The sensitive area is an area in the original image that is related to a user's privacy.

7. A computer-readable storage medium storing a computer program for executing the image data masking method according to any one of claims 1 to 5.

8. a processor; a memory for storing instructions executable by the processor; The electronic device is adapted for implementing the method for masking image data according to any one of claims 1 to 5, wherein the processor reads the executable instructions from the memory and executes the instructions.

Images