Sensitive information hiding processing method and system and related device

Abstract

The invention discloses a sensitive information hiding processing method and system and a related device, and relates to the field of image processing, and the method comprises the steps: carrying out the sensitive region recognition of a to-be-processed image through a deep learning algorithm, and obtaining the pixel points of a plurality of sensitive regions in the to-be-processed image; pixel point coordinates and pixel values of pixel points in each sensitive area are extracted, and the pixel values of the pixel points in each sensitive area are changed into preset values after extraction; generating a random position table of the pixels of the plurality of sensitive areas based on the number of the pixels of the to-be-processed image, the coordinates of the pixels and a preset position generation function; and according to the random image position corresponding to each pixel point in the random position table, embedding a corresponding pixel value in the random image position to obtain a steganographic image of the to-be-processed image. According to the method, the flexibility and the accuracy of sensitive content recognition are effectively improved by utilizing the deep learning algorithm, and the hidden position of the sensitive content is randomly generated, so that the sensitive content of the image is effectively processed.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, system and related apparatus for hiding sensitive information. Background Technology

[0002] With the rapid development of information technology, the security of sensitive information has received increasing attention. As one of the carriers of information dissemination, images require secure processing of sensitive content in many application scenarios, such as online information dissemination and data storage, to prevent the leakage of sensitive information.

[0003] The secure handling of sensitive content in images mainly includes the detection and concealment of sensitive content. Regarding sensitive content detection, relying on manual review or simple detection rules results in low efficiency and is prone to false positives and false negatives. As for concealing sensitive content, using a fixed allocation method to embed sensitive content into the image lacks randomness and is easily cracked. It is evident that current methods for securely handling sensitive content in images are ineffective; therefore, how to effectively process sensitive content in images is a pressing technical problem that needs to be solved. Summary of the Invention

[0004] In view of the above problems, this application provides a method, system, and related apparatus for hiding sensitive information, so as to effectively process sensitive content in images. The specific solution is as follows:

[0005] The first aspect of this application provides a method for hiding sensitive information, the method comprising:

[0006] A deep learning algorithm is used to identify sensitive regions in the image to be processed, and the pixels of multiple sensitive regions in the image to be processed are obtained.

[0007] Extract the pixel coordinates and pixel values ​​of pixels in each sensitive region, and then change the pixel values ​​of pixels in each sensitive region to preset values ​​after extraction;

[0008] Based on the number of pixels in the image to be processed, the pixel coordinates, and a preset position generation function, a random position table of pixels in multiple sensitive regions is generated.

[0009] According to the random image position corresponding to each pixel in the random position table, the corresponding pixel value is embedded at the random image position to obtain the steganalysis image of the image to be processed.

[0010] In one possible implementation, the use of deep learning algorithms to identify sensitive regions of the image to be processed includes:

[0011] The deep learning algorithm is used to identify the image content of the image to be processed, determine whether the image content is sensitive, and obtain multiple recognition boxes of the image to be processed. Each recognition box selects an image area containing sensitive content.

[0012] Each image region selected by a recognition box is identified as a sensitive region, thus obtaining multiple sensitive regions.

[0013] In one possible implementation, changing the pixel value of each pixel in the sensitive region to a preset value includes:

[0014] Set the RGB value of each pixel in the sensitive area to 0.

[0015] In one possible implementation, generating a random position table of pixels in multiple sensitive regions based on the number of pixels in the image to be processed, the pixel coordinates, and a preset position generation function includes:

[0016] The pixel coordinates of each pixel, the number of rows of pixels and the number of columns of pixels in the image to be processed are substituted into the preset position generation function to calculate the random pixel coordinates corresponding to the pixel coordinates in the image to be processed.

[0017] Establish the correspondence between the pixel coordinates of each pixel and the calculated random pixel coordinates to obtain the random position table.

[0018] In one possible implementation, the random image position is: the random pixel coordinates in the image to be processed, which are calculated according to the preset position generation function based on the pixel coordinates.

[0019] The step of embedding the corresponding pixel value at the random image position according to the random image position corresponding to each pixel in the random position table includes:

[0020] Convert the pixel value of each pixel into a binary number, and retain the high four bits of the binary number;

[0021] Convert the pixel value corresponding to the random pixel coordinates of each pixel into a binary number;

[0022] Modify the lower four bits of the binary number of each random pixel coordinate to the higher four bits of the binary number of each pixel.

[0023] Wherein, the high four bits are the first four binary bits from left to right of the binary number, and the low four bits are the first four binary bits from right to left of the binary number.

[0024] One possible implementation also includes:

[0025] Obtain the steganalytic image;

[0026] According to the random location table, the embedded pixel value of each pixel is extracted from each random image location, and the pixel value of each pixel in each sensitive area is changed to the embedded pixel value of the pixel to obtain the recovered image to be processed.

[0027] A second aspect of this application provides a system for hiding sensitive information, the system comprising:

[0028] The recognition unit is used to identify sensitive regions of the image to be processed using a deep learning algorithm, and to obtain the pixel points of multiple sensitive regions in the image to be processed.

[0029] An extraction unit is used to extract the pixel coordinates and pixel values ​​of pixels in each sensitive region, and after extraction, change the pixel values ​​of pixels in each sensitive region to preset values;

[0030] The generation unit is used to generate a random position table of pixels in multiple sensitive regions based on the number of pixels in the image to be processed, the pixel coordinates, and a preset position generation function.

[0031] An embedding unit is used to embed a corresponding pixel value at a random image position corresponding to each pixel in the random position table, thereby obtaining a steganalyte of the image to be processed.

[0032] In one possible implementation, the recognition unit utilizes a deep learning algorithm to identify sensitive regions of the image to be processed, specifically configured as follows:

[0033] The deep learning algorithm is used to identify the image content of the image to be processed, determine whether the image content is sensitive, obtain multiple recognition boxes of the image to be processed, and the image area selected by each recognition box contains sensitive content; the image area selected by each recognition box is determined as the sensitive area, and multiple sensitive areas are obtained.

[0034] A third aspect of this application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0035] The memory is used to store computer programs;

[0036] The processor is used to execute the computer program so that the electronic device can implement the sensitive information hiding processing method of the first aspect or any implementation thereof.

[0037] The fourth aspect of this application provides a computer program product including computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the sensitive information hiding processing method of the first aspect or any implementation thereof.

[0038] Based on the above technical solution, this application provides a method, system, and related apparatus for hiding sensitive information. This method uses a deep learning algorithm to identify sensitive regions in an image to be processed, obtaining pixels in multiple sensitive regions. The coordinates and pixel values ​​of each pixel in each sensitive region are extracted, and the pixel values ​​are changed to preset values ​​after extraction. Based on the image information, pixel coordinates, and a preset position generation function of the image to be processed, a random position table of pixels in multiple sensitive regions is generated. According to the random image position corresponding to each pixel in the random position table, the corresponding pixel value is embedded in the random image position to obtain a steganographic image of the image to be processed, thus hiding sensitive content. This method utilizes a deep learning algorithm to identify sensitive content in an image. Compared to manual review or simple detection rules, it effectively improves the flexibility and accuracy of sensitive content identification. Furthermore, this method randomly generates random image positions for each pixel based on image information and a preset position generation function, effectively increasing the randomness of the hidden position of sensitive content. Therefore, this method can effectively process sensitive content in images. Attached Figure Description

[0039] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0040] Figure 1 A flowchart illustrating a method for hiding sensitive information provided in an embodiment of this application;

[0041] Figure 2 A schematic diagram of a sensitive information hiding processing system provided in an embodiment of this application;

[0042] Figure 3 This application provides a hardware structure block diagram of an electronic device. Detailed Implementation

[0043] The embodiments of this application are described below with reference to the accompanying drawings. The terminology used in the implementation section of this application is for explaining specific embodiments only and is not intended to limit the scope of this application.

[0044] The embodiments of this application will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are equally applicable to similar technical problems.

[0045] The terms “comprising” and “having”, and any variations thereof, in the specification and accompanying drawings of this application are intended to cover non-exclusive inclusion, such that a process, method, system, product, or apparatus that comprises a series of elements is not necessarily limited to those elements, but may include other elements not expressly listed or inherent to such process, method, product, or apparatus.

[0046] The secure handling of sensitive content in images mainly includes the detection and hiding of sensitive content. Existing encryption and decryption algorithms for sensitive content in images encrypt and decrypt the entire image, which cannot automatically identify local sensitive content and perform local encryption and decryption, resulting in additional resource consumption. Furthermore, users without decryption permissions cannot read normal image content other than sensitive content.

[0047] To address the aforementioned problems, embodiments of this application provide a method for hiding sensitive information. The method for hiding sensitive information according to embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0048] Reference Figure 1 , Figure 1 A flowchart illustrating a method for hiding sensitive information provided in an embodiment of this application is shown below. Figure 1 As shown in the figure, the method for hiding sensitive information provided in this application embodiment may include steps S10 to S13, which are described in detail below.

[0049] S10. Use deep learning algorithms to identify sensitive regions in the image to be processed and obtain the pixels of multiple sensitive regions in the image to be processed.

[0050] S11. Extract the pixel coordinates and pixel values ​​of each pixel in each sensitive area, and change the pixel values ​​of each pixel in the sensitive area to preset values ​​after extraction.

[0051] Deep learning algorithms, in this context, refer to machine learning algorithms that automatically learn data features from massive amounts of data based on a multi-layered neural network structure to achieve complex tasks such as classification, prediction, and generation. This embodiment primarily uses PaddleOCR to identify sensitive regions in the image to be processed. PaddleOCR is an open-source optical character recognition library. This embodiment uses a self-learning deep learning algorithm to identify sensitive content, capable of handling distortions, warps, various colors, and different types of text, demonstrating good versatility.

[0052] Sensitive content in an image can refer to image content that does not conform to specific scene specifications and requires restricted dissemination and processing. Sensitive areas can refer to image regions within the image to be processed that contain sensitive content; these areas may contain only the sensitive content or some other image content. This embodiment achieves local encryption of sensitive information image regions with minimal impact through the identification and processing of sensitive areas, protecting information security without affecting the access to other publicly available information within the image.

[0053] In this embodiment, when recognizing an image to be processed using a deep learning algorithm, the algorithm identifies the image content and determines whether the content is sensitive. Each recognition box can select an image region that contains sensitive content, resulting in multiple recognition boxes for the image to be processed. This embodiment can define the image region selected by each recognition box as a sensitive region, thus obtaining multiple sensitive regions. Specifically, the recognition box can be a rectangular coordinate system, represented as follows:

[0054] ;

[0055] in, It can represent a recognition box; It can represent the coordinates of the top-left corner of the rectangle relative to the top-left corner of the image to be processed; This can represent the coordinates of the bottom right corner of the rectangle relative to the top left corner of the image to be processed; where the coordinates of the top left corner of the image to be processed can be the origin of the pixel coordinate system of the image to be processed. It can represent the horizontal coordinate of the pixel coordinate system. It can represent the vertical coordinate of the pixel coordinate system.

[0056] Therefore, multiple bounding boxes in the image to be processed can be represented as a coordinate array, which can be expressed as follows:

[0057] ;

[0058] This embodiment can traverse the coordinate array of the image to be processed, determine the position of the sensitive region in the image based on the coordinates of each recognition box in the coordinate array, and extract the pixel data (which can represent the sensitive content in the sensitive region) in the sensitive region. This pixel data can include the pixel coordinates and pixel value of each pixel in the sensitive region. The pixel value of each pixel can include the channel data of that pixel. The channel data of a pixel can refer to independent numerical components describing the pixel color and attributes. Each channel can correspond to a specific information dimension, and the combination of channels can constitute a complete representation of the pixel value. After extracting the pixel data of each sensitive region, this embodiment can change the pixel value of each pixel within the recognition box to a preset value based on the coordinates of the recognition box corresponding to the sensitive region, thereby eliminating the visual traces of the original sensitive content and improving the concealment and recoverability of the sensitive content. The preset value can refer to a fixed pixel value (a fixed value for each color channel of the pixel), such as 0 (RGB value (0, 0, 0), representing pure black) or 255 (RGB value (255, 255, 255), representing pure white). In this embodiment, after extracting pixel data from each sensitive region, the RGB value of each pixel in the sensitive region is set to 0.

[0059] S12. Based on the number of pixels, pixel coordinates, and preset position generation function of the image to be processed, generate a random position table of pixels in multiple sensitive areas;

[0060] S13. According to the random image position corresponding to each pixel in the random position table, embed the corresponding pixel value at the random image position to obtain the steganalysis image of the image to be processed.

[0061] In this embodiment, after extracting the sensitive content from the image to be processed, a random location table for the sensitive content is generated based on the information of the image to be processed (the number of pixels in the image) and the characteristics of the sensitive content (the coordinates of pixels in the sensitive area). This table records the random hidden locations (random image positions) of the sensitive content in the image to be processed, and the sensitive content is hidden in its respective hidden location. This embodiment uses a random allocation algorithm to allocate the hidden locations of sensitive content. Due to the randomness of the random allocation algorithm in allocating hidden locations, it can effectively improve the randomness and security of hiding sensitive content.

[0062] The number of pixels in the image to be processed can refer to the total number of all pixel units (pixels) constituting the image. This number can be specifically the product of the image width (the number of pixel units in the horizontal direction) and the image height (the number of pixel units in the vertical direction). The random image location can refer to the pixel coordinates of the pixels in the image to be processed used to embed sensitive content. The preset location generation function can refer to the function used to generate the hidden locations of pixels in the image, and its representation can be as follows:

[0063] ;

[0064] in, It can represent a generated random position; It can represent the x-coordinate of a pixel belonging to a sensitive area in the pixel coordinate system of the image to be processed; It can represent the ordinate value of a pixel belonging to a sensitive area in the pixel coordinate system of the image to be processed; It can represent a random seed, and can be an integer; It can represent the modulo operation, the core of which is to calculate the remainder after dividing two numbers; It can represent the number of pixel units in the vertical direction of the image to be processed; It can represent the number of pixel units in the horizontal direction of the image to be processed.

[0065] ;

[0066] ;

[0067] in, and Both can be large prime numbers; This can represent a bitwise XOR operation on two results (after conversion to binary). XOR is a logical operation whose core rule is: compare two binary bits; if the two binary bits are the same, the XOR result is 0; if the two binary bits are different, the XOR result is 1. For example, given two binary numbers A and B, A: 0011 and B: 0101, comparing them bit by bit: starting with the first binary bit from left to right, both A and B are 0, so the XOR result is 0. Starting with the second binary bit from left to right, both A and B are 0, so the XOR result is 1, and so on. The result for B is 0110. It can represent a right shift operation of 32 bits, extracting the high 32 bits (the first 32 binary bits from left to right) from the result of the operation to obtain a 32-bit value.

[0068] In this embodiment, the process of generating the random location table can be specifically described as follows:

[0069] In this embodiment, the pixel coordinates of each pixel, the number of rows of pixels and the number of columns of pixels in the image to be processed can be substituted into a preset position generation function to calculate the random pixel coordinates corresponding to the pixel coordinates in the image to be processed; the correspondence between the pixel coordinates of each pixel and the calculated random pixel coordinates can be established to obtain a random position table.

[0070] After obtaining the random location table in this embodiment, the extracted hidden content can be embedded into the corresponding random hidden locations according to the random location table. Specifically, this embodiment can convert the pixel data of each extracted sensitive region into a corresponding matrix S, the size of which can be h×w×n, where h is the width of the sensitive region (the number of pixel units in the horizontal direction of the sensitive region), w is the height of the sensitive region (the number of pixel units in the vertical direction of the sensitive region), and n can be the number of channels in the sensitive region. For example, if the image to be processed is an RGB image, and its sensitive region includes the R channel, G channel, and B channel, then n=3.

[0071] For each sensitive region matrix S, expand matrix S row-wise to obtain a corresponding one-dimensional array. Convert each pixel value in the one-dimensional array into a binary number (a numerical value represented in binary format). Since pixel values ​​are integers within the range of 0 to 255, the maximum number of bits in the corresponding binary number is eight. Furthermore, after converting each pixel value in the one-dimensional array into a binary number, if the number of bits is no more than four, retain all the bit values; if the number of bits is more than four, retain only the significant portion (a portion of the bit values), thus obtaining the binary sequence corresponding to the one-dimensional array. ,and length for: The sequence It can include partial binary bits of multiple pixels. Furthermore, this embodiment can simultaneously convert the pixel value corresponding to the random pixel coordinates of each pixel into a binary number.

[0072] This embodiment obtains the binary sequence of each sensitive region. Then, the sequence can be traversed. Get sequence The binary information of each pixel is processed, and the corresponding random pixel coordinates of each pixel are determined according to the random position table. If the binary number of the random pixel coordinates is greater than four bits, some of the binary bits of the random pixel coordinates are modified to retain the binary bits of each pixel. If the binary number of the random pixel coordinates is less than four bits, the binary bits of the pixel can be directly added after the binary number of the random pixel coordinates. After modification, the number of binary bits of the random pixel coordinates is no more than eight bits, thereby realizing the embedding of sensitive content.

[0073] Specifically, in this embodiment, after converting the pixel value of each pixel into a binary number, the high four bits of the binary number are retained. The low four bits of the binary number of each random pixel coordinate are modified to the high four bits of the binary number of each pixel, thereby achieving the embedding of sensitive content. The high four bits are the first four binary bits from left to right, and the low four bits are the first four binary bits from right to left. Since the high four bits of each pixel determine the main information of the pixel that can be observed by the human eye, including pixel brightness, color, and other key information, and are the core of pixel information, only the high four bits of the pixel are retained when embedding the pixel value into the image.

[0074] For example, the binary number corresponding to a pixel containing sensitive content can be represented as: b1, b2, b3, b4, b5, b6, b7, b8. Keeping the high four bits, we get: b1, b2, b3, b4. The binary number corresponding to its random pixel coordinates can be represented as: r1, r2, r3, r4, r5, r6, r7, r8. Each b or r can represent a binary bit. By modifying the low four bits of r1, r2, r3, r4, r5, r6, r7, r8 and embedding the high four bits of the pixel, the modified binary number corresponding to the random pixel coordinates becomes: r1, r2, r3, r4, b1, b2, b3, b4.

[0075] After embedding the pixel values ​​of each pixel in the sensitive region in this embodiment, a steganographic image to be processed (the image to be processed after processing the sensitive content) can be obtained. When this embodiment needs to recover the hidden sensitive content, the steganographic image of the image to be processed can be obtained, and the embedded pixel values ​​of each pixel can be extracted from each random image position according to the random position table. The pixel values ​​of each pixel in the sensitive region are changed to the corresponding embedded pixel values, and each random image position is restored to its initial state before the pixel values ​​were embedded, thus obtaining the recovered image to be processed. Here, the embedded pixel value can refer to the pixel value of the pixel belonging to the sensitive region embedded in the random image position. Since only part of the information (part of the binary bits) of the pixel is retained when embedding the pixel values, after recovery, each pixel belonging to the sensitive region only contains part of the information. And since the retained part of the information is the main information of the pixel that can be observed by the human eye, the recovered image is unlikely to have problems such as image blurring or distortion.

[0076] For example, the binary number corresponding to a random pixel coordinate in a steganographic image is: r1, r2, r3, r4, b1, b2, b3, b4. The embedded pixel values ​​of the pixels belonging to the sensitive area are extracted as: b1, b2, b3, b4. The random pixel coordinates are restored as: r1, r2, r3, r4, r5, r6, r7, r8. The restored pixels belonging to the sensitive area are: b1, b2, b3, b4, 0, 0, 0, 0.

[0077] This application provides a method for hiding sensitive information. This method uses a deep learning algorithm to identify sensitive regions in an image to be processed, obtaining pixels in multiple sensitive regions. The coordinates and values ​​of each pixel in each sensitive region are extracted, and the pixel values ​​are changed to preset values ​​after extraction. Based on the image information, pixel coordinates, and a preset position generation function of the image to be processed, a random position table of pixels in multiple sensitive regions is generated. According to the random image position corresponding to each pixel in the random position table, the corresponding pixel value is embedded in the random image position to obtain a steganographic image of the image to be processed, thus hiding the sensitive content. This method utilizes a deep learning algorithm to identify sensitive content in an image. Compared to manual review or simple detection rules, it effectively improves the flexibility and accuracy of sensitive content identification. Furthermore, this method randomly generates random image positions for each pixel based on image information and a preset position generation function, effectively increasing the randomness of the hidden position of sensitive content. Therefore, this method can effectively process sensitive content in images.

[0078] The above describes a method for hiding sensitive information provided by embodiments of this application. The following will describe a system that applies the above-described method for hiding sensitive information.

[0079] Please see Figure 2 , Figure 2 This is a schematic diagram of a sensitive information hiding processing system provided in an embodiment of this application. Figure 2 As shown, the sensitive information hiding and processing system may include:

[0080] The recognition unit 100 is used to use a deep learning algorithm to identify sensitive regions of the image to be processed and obtain the pixels of multiple sensitive regions in the image to be processed.

[0081] The extraction unit 110 is used to extract the pixel coordinates and pixel values ​​of pixels in each sensitive region, and to change the pixel values ​​of pixels in each sensitive region to preset values ​​after extraction.

[0082] The generation unit 120 is used to generate a random position table of pixels in multiple sensitive regions based on the number of pixels, pixel coordinates and a preset position generation function of the image to be processed.

[0083] The embedding unit 130 is used to embed the corresponding pixel value at the random image position according to the random image position corresponding to each pixel point in the random position table, so as to obtain the steganalysis image of the image to be processed.

[0084] In one possible implementation, the recognition unit 100 uses a deep learning algorithm to identify sensitive regions of the image to be processed, which can be specifically configured as follows:

[0085] The deep learning algorithm is used to identify the image content of the image to be processed, determine whether the image content is sensitive, obtain multiple recognition boxes of the image to be processed, and the image area selected by each recognition box contains sensitive content; the image area selected by each recognition box is determined as a sensitive area, and multiple sensitive areas are obtained.

[0086] In one possible implementation, the extraction unit 110 changes the pixel value of each pixel in the sensitive region to a preset value, which can be specifically configured as follows: the RGB value of each pixel in the sensitive region is set to 0.

[0087] In one possible implementation, the generation unit 120 can be specifically configured as follows:

[0088] The pixel coordinates of each pixel, the number of rows of pixels and the number of columns of pixels in the image to be processed are substituted into the preset position generation function to calculate the random pixel coordinates corresponding to the pixel coordinates in the image to be processed; the correspondence between the pixel coordinates of each pixel and the calculated random pixel coordinates is established to obtain the random position table.

[0089] In one possible implementation, the random image position is: the random pixel coordinates in the image to be processed, which are calculated based on the pixel coordinates of the preset position generation function.

[0090] In the embedding unit 130, the corresponding pixel value is embedded at the random image position according to the random image position corresponding to each pixel point in the random position table. This can be specifically configured as follows:

[0091] Convert the pixel value of each pixel to a binary number and retain the high four bits of the binary number; convert the pixel value of the random pixel coordinate corresponding to each pixel to a binary number; modify the low four bits of the binary number of each random pixel coordinate to the high four bits of the binary number of each pixel; where the high four bits are the first four binary bits from left to right of the binary number, and the low four bits are the first four binary bits from right to left of the binary number.

[0092] In one possible implementation, the sensitive information hiding processing system may further include a recovery unit, which may be specifically configured as follows:

[0093] Obtain the steganalysis image; extract the embedded pixel value of each pixel from each random image position according to the random position table, change the pixel value of each pixel in the sensitive area to the corresponding embedded pixel value, and obtain the recovered image to be processed.

[0094] This application also provides an electronic device in its embodiments. (See reference...) Figure 3 The diagram illustrates a structural schematic suitable for implementing the electronic device in the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 3 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0095] like Figure 3 As shown, the electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. When the electronic device is powered on, the RAM 303 also stores various programs and data required for the operation of the electronic device. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output interface (I / O interface) 305 is also connected to the bus 304.

[0096] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, memory cards, hard drives, etc.; and communication devices 309. Communication device 309 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or have alternatively.

[0097] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to implement any of the sensitive information hiding processing methods provided in this application.

[0098] This application also provides a computer-readable storage medium that carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement any of the sensitive information hiding processing methods provided in this application.

[0099] It should also be noted that the system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the system embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.

[0100] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0101] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0102] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).

[0103] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0104] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0105] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for hiding sensitive information, characterized in that, The method for hiding the sensitive information includes: A deep learning algorithm is used to identify sensitive regions in the image to be processed, and the pixels of multiple sensitive regions in the image to be processed are obtained. Extract the pixel coordinates and pixel values ​​of pixels in each sensitive region, and then change the pixel values ​​of pixels in each sensitive region to preset values ​​after extraction; Based on the number of pixels in the image to be processed, the pixel coordinates, and a preset position generation function, a random position table of pixels in multiple sensitive regions is generated. According to the random image position corresponding to each pixel in the random position table, the corresponding pixel value is embedded at the random image position to obtain the steganalysis image of the image to be processed.

2. The method for hiding sensitive information according to claim 1, characterized in that, The process of using deep learning algorithms to identify sensitive regions in the image to be processed includes: The deep learning algorithm is used to identify the image content of the image to be processed, determine whether the image content is sensitive, and obtain multiple recognition boxes of the image to be processed. Each recognition box selects an image area containing sensitive content. Each image region selected by a recognition box is identified as a sensitive region, thus obtaining multiple sensitive regions.

3. The method for hiding sensitive information according to claim 1, characterized in that, Change the pixel value of each pixel in the sensitive region to a preset value, including: Set the RGB value of each pixel in the sensitive area to 0.

4. The method for hiding sensitive information according to claim 1, characterized in that, The step of generating a random position table of pixels in multiple sensitive regions based on the number of pixels in the image to be processed, the pixel coordinates, and a preset position generation function includes: The pixel coordinates of each pixel, the number of rows of pixels and the number of columns of pixels in the image to be processed are substituted into the preset position generation function to calculate the random pixel coordinates corresponding to the pixel coordinates in the image to be processed. Establish the correspondence between the pixel coordinates of each pixel and the calculated random pixel coordinates to obtain the random position table.

5. The method for hiding sensitive information according to claim 1, characterized in that, The random image position is: the random pixel coordinates in the image to be processed, calculated according to the preset position generation function; The step of embedding the corresponding pixel value at the random image position according to the random image position corresponding to each pixel in the random position table includes: Convert the pixel value of each pixel into a binary number, and retain the high four bits of the binary number; Convert the pixel value corresponding to the random pixel coordinates of each pixel into a binary number; Modify the lower four bits of the binary number of each random pixel coordinate to the higher four bits of the binary number of each pixel. Wherein, the high four bits are the first four binary bits from left to right of the binary number, and the low four bits are the first four binary bits from right to left of the binary number.

6. The method for hiding sensitive information according to claim 1, characterized in that, Also includes: Obtain the steganalytic image; According to the random location table, the embedded pixel value of each pixel is extracted from each random image location, and the pixel value of each pixel in each sensitive area is changed to the embedded pixel value of the pixel to obtain the recovered image to be processed.

7. A system for hiding and processing sensitive information, characterized in that, The sensitive information hiding and processing system includes: The recognition unit is used to identify sensitive regions of the image to be processed using a deep learning algorithm, and to obtain the pixel points of multiple sensitive regions in the image to be processed. An extraction unit is used to extract the pixel coordinates and pixel values ​​of pixels in each sensitive region, and after extraction, change the pixel values ​​of pixels in each sensitive region to preset values; The generation unit is used to generate a random position table of pixels in multiple sensitive regions based on the number of pixels in the image to be processed, the pixel coordinates, and a preset position generation function. An embedding unit is used to embed a corresponding pixel value at a random image position corresponding to each pixel in the random position table, thereby obtaining a steganalyte of the image to be processed.

8. The sensitive information hiding and processing system according to claim 7, characterized in that, The recognition unit utilizes a deep learning algorithm to identify sensitive regions in the image to be processed. The specific configuration is as follows: The deep learning algorithm is used to identify the image content of the image to be processed, determine whether the image content is sensitive, and obtain multiple recognition boxes of the image to be processed. Each recognition box selects an image area containing sensitive content. Each image region selected by a recognition box is identified as a sensitive region, thus obtaining multiple sensitive regions.

9. An electronic device, characterized in that, It includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program to enable the electronic device to implement the sensitive information hiding processing method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, It includes computer-readable instructions that, when executed on an electronic device, cause the electronic device to implement the sensitive information hiding processing method as described in any one of claims 1 to 6.

Images