Image encryption method and device, computer equipment and storage medium

Abstract

The invention belongs to the technical field of artificial intelligence, and relates to an image encryption method, which comprises the following steps: acquiring a to-be-encrypted original image; preprocessing the original image based on a preprocessing module to obtain a state matrix corresponding to the original image; performing dynamic key generation processing on the state matrix based on the target zero neural network to obtain an original key; performing random enhancement processing on the original key based on a random security enhancement module to obtain a target key; performing encryption processing on the original image based on the target key to obtain an encrypted image corresponding to the original image; and carrying out output processing on the encrypted image. The invention further provides an image encryption device, computer equipment and a storage medium. In addition, the invention also relates to a block chain technology, and the error positioning information can be stored in a block chain. The image encryption method and device can be applied to business scenes of image encryption in the financial field and the medical field, and dynamic adaptability and safety of the image encryption process are effectively ensured by combining multiple function modules.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology and can be applied to fields such as financial technology and digital medicine, and in particular to image encryption methods, devices, computer equipment and storage media. Background Art

[0002] In traditional image encryption technology, static keys or predefined key expansion mechanisms (such as the Advanced Encryption Standard (AES) and chaotic map encryption) have long dominated. These methods achieve pixel value permutation and diffusion through fixed keys or iteratively generated pseudo-random sequences based on initial values. However, the static nature of the keys poses significant security risks to the encryption system, resulting in low image encryption security. Specifically, static keys are easily stolen or reverse-engineered during transmission or storage, and the encryption process lacks the ability to dynamically respond to plaintext image features, making the encrypted results susceptible to key patterns or pixel distribution characteristics under known plaintext attacks or statistical analysis.

[0003] For example, in remote electronic credential encryption scenarios in the financial sector, traditional encryption methods may use fixed keys to encrypt high-value images such as bank transfer receipts and electronic contracts. If an attacker can obtain key patterns from a portion of a plaintext image (such as a transfer receipt in a known format), they can use statistical analysis to decrypt other encrypted images, leading to the leakage of sensitive data such as customer account information and transaction records. Such security vulnerabilities not only threaten user privacy but can also lead to financial fraud or financial losses, seriously affecting the compliance operations and market reputation of financial institutions.

[0004] Therefore, it is urgent to propose an image encryption method that can effectively resist modern cryptanalysis attacks to improve the security of image encryption. Summary of the Invention

[0005] The purpose of the embodiments of the present application is to propose an image encryption method, apparatus, computer equipment and storage medium to solve the technical problem of low security in existing image encryption technology.

[0006] In a first aspect, an image encryption method is provided, comprising:

[0007] Obtain the original image to be encrypted;

[0008] Preprocessing the original image based on a preset preprocessing module to obtain a state matrix corresponding to the original image;

[0009] Performing dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain a corresponding original key;

[0010] Performing random enhancement processing on the original key based on a preset random security enhancement module to obtain a corresponding target key;

[0011] encrypting the original image based on the target key to obtain an encrypted image corresponding to the original image;

[0012] Output processing is performed on the encrypted image.

[0013] In a second aspect, an image encryption device is provided, comprising:

[0014] A first acquisition module, configured to acquire an original image to be encrypted;

[0015] A preprocessing module, configured to preprocess the original image based on a preset preprocessing module to obtain a state matrix corresponding to the original image;

[0016] A generation module, configured to perform dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain a corresponding original key;

[0017] An enhancement module, configured to perform random enhancement processing on the original key based on a preset random security enhancement module to obtain a corresponding target key;

[0018] an encryption module, configured to encrypt the original image based on the target key to obtain an encrypted image corresponding to the original image;

[0019] An output module is used to perform output processing on the encrypted image.

[0020] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned image encryption method when executing the computer program.

[0021] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned image encryption method are implemented.

[0022] In the scheme implemented by the above-mentioned image encryption method, device, computer equipment and storage medium, the original image to be encrypted is first obtained; then the original image is preprocessed based on a preset preprocessing module to obtain a state matrix corresponding to the original image; then the state matrix is ​​dynamically key generated based on a preset target annihilation neural network to obtain the corresponding original key; subsequently, the original key is randomly enhanced based on a preset random security enhancement module to obtain the corresponding target key; further, the original image is encrypted based on the target key to obtain an encrypted image corresponding to the original image; finally, the encrypted image is output. After obtaining the original image to be encrypted, the present application preprocesses the original image using a preprocessing module to obtain a state matrix corresponding to the original image, then dynamically key generates the state matrix based on a target annihilation neural network to obtain the original key, then randomly enhances the original key based on a random security enhancement module to obtain the target key, and then encrypts the original image based on the target key to obtain an encrypted image and outputs it. Through the above image encryption processing method, this application realizes end-to-end conversion from image to key and then to encrypted image by combining the preprocessing module, target zeroing neural network and random security enhancement module. The image encryption mechanism follows the core process chain of feature extraction-state evolution-key generation-security enhancement-image encryption. Each link is closely related to the image content, effectively ensuring the dynamic adaptability and security of the image encryption process. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the solutions in this application, a brief introduction will be given below to the drawings required for use in the description of the embodiments of this application. Obviously, the drawings described below are some embodiments of this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0024] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0025] Figure 2 is a flow chart of an embodiment of an image encryption method according to the present application;

[0026] Figure 3 is a structural diagram of an embodiment of an image encryption device according to the present application;

[0027] Figure 4 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of the application are for the purpose of describing specific embodiments only and are not intended to limit this application. The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0029] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0030] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0031] like Figure 1 As shown, system architecture 100 may include a terminal device 101, a network 102, and a server 103. Terminal device 101 may be a laptop computer 1011, a tablet computer 1012, or a mobile phone 1013. Network 102 is a medium for providing a communication link between terminal device 101 and server 103. Network 102 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0032] The user can use the terminal device 101 to interact with the server 103 via the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0033] The terminal device 101 can be various electronic devices with a display screen and supporting web browsing. In addition to the laptop computer 1011, tablet computer 1012 or mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer and a desktop computer, etc.

[0034] The server 103 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal device 101 .

[0035] It should be noted that the image encryption method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the image encryption device is generally set in the server / terminal device.

[0036] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0037] Continue to refer Figure 2 , shows a flowchart of an embodiment of the image encryption method according to the present application. According to different needs, the order of the steps in the flowchart can be changed, and some steps can be omitted. The image encryption method provided by the embodiment of the present application can be applied to any scenario where image encryption is required, and the image encryption method can be applied to products in these scenarios. For example, it can be applied to business scenarios of image encryption in the financial and medical fields. The image encryption method includes the following steps:

[0038] Step S201: obtaining the original image to be encrypted.

[0039] In this embodiment, the electronic device (eg Figure 1The server / terminal device shown in the figure can obtain the original image to be encrypted via a wired or wireless connection. It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra wideband) connections, and other wireless connection methods currently known or developed in the future. The execution subject of this application may specifically be an image encryption system, which can be simply referred to as the system. The image encryption system includes three main functional units: a preprocessing module, a target zeroing neural network, and a random security enhancement module. The encryption process begins with the input of the original image and ultimately outputs an encrypted image after a series of processing. The modules are interconnected via a data bus to form a closed-loop information processing pipeline, achieving end-to-end conversion from image to key and then to encrypted image. The system's operating mechanism follows the core process chain of "feature extraction - state evolution - key generation - security enhancement - image encryption". Each link is closely related to the image content to ensure the dynamic adaptability and security of the encryption process.

[0040] When a user inputs an image to be encrypted (such as a scanned copy of a paper insurance policy or a medical image), the system first standardizes the image and converts it into a high-dimensional feature representation through a preprocessing module. This step not only unifies the formats of images from different sources, but more importantly, it extracts the key features of the image, laying the foundation for subsequent dynamic key generation. The preprocessed feature tensor is input into the target zeroing neural network (or ZNN core layer), where the feature tensor is reorganized into the initial state matrix S(0) and the ZNN dynamic system iteration is started. The ZNN core layer calculates the state evolution according to the preset dynamic system equation and captures the high-dimensional nonlinear relationship between state vectors through the correlation matrix. This correlation information is used to dynamically generate encryption keys, so that the keys are closely bound to the content characteristics of the input image. The generated original key then enters the random security enhancement module, where controllable randomness is introduced based on the chaotic system to further enhance the security and unpredictability of the key. Finally, the system uses the enhanced key to perform encryption operations on the original image, implements pixel-level hybrid transformations through a custom diffusion operator, and outputs the final encrypted image.

[0041] This application can be applied to image encryption business processing scenarios in the financial and medical fields. For example, in the financial field, the above-mentioned original images may include images such as scanned copies of paper insurance policies, photos of accident scenes, invoices, and anti-counterfeiting patterns on checks. In the medical field, the above-mentioned original images may include medical images (such as CT slice images, MRI slice images), full-field digital slices, dermatoscope photos, and other images.

[0042] Step S202 : pre-processing the original image based on a preset pre-processing module to obtain a state matrix corresponding to the original image.

[0043] In this embodiment, the preprocessing module, also referred to as an input preprocessing module, is responsible for converting raw images of various formats and sizes into standardized feature representations suitable for neural network processing. This preprocessing module implements the conversion from raw pixel data to high-dimensional feature tensors through a series of image processing operations. The specific implementation process of preprocessing the raw image based on the preset preprocessing module to obtain a state matrix corresponding to the raw image will be further described in detail in subsequent specific embodiments of this application and will not be elaborated on here.

[0044] Step S203 : performing dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain a corresponding original key.

[0045] In this embodiment, the above-mentioned target zeroing neural network can also be referred to as the ZNN core layer. The target zeroing neural network is the core computing unit of this application, which is responsible for the dynamic evolution of the state matrix and key generation. This layer adopts differential-algebraic collaborative design, and drives the key generation process through the inherent dynamics of the neural network state. The ZNN core layer includes three key subcomponents: a state updater, an association matrix calculator, and a key extractor, which together constitute a complete dynamic key generation system. Among them, the above-mentioned specific implementation process of performing dynamic key generation processing on the state matrix based on the preset target zeroing neural network to obtain the corresponding original key will be further described in detail in the subsequent specific embodiments of this application, and will not be elaborated on here.

[0046] Step S204: performing random enhancement processing on the original key based on a preset random security enhancement module to obtain a corresponding target key.

[0047] In this embodiment, the random security enhancement module is a key component of the application's security assurance, responsible for introducing the necessary randomness while maintaining key reproducibility. This module, composed of a chaotic pseudo-random number generator and a key mixing unit, jointly performs random enhancement processing on the original key. The specific implementation process of performing random enhancement processing on the original key based on the preset random security enhancement module to obtain the corresponding target key will be further described in detail in subsequent specific embodiments of this application and will not be elaborated on here.

[0048] Step S205 : encrypting the original image based on the target key to obtain an encrypted image corresponding to the original image.

[0049] In this embodiment, the specific implementation process of encrypting the original image based on the target key to obtain the encrypted image corresponding to the original image will be further described in detail in subsequent specific embodiments of this application and will not be elaborated on here.

[0050] Step S206: output the encrypted image.

[0051] In this embodiment, the generated encrypted image may be sent to relevant personnel so that the relevant personnel can use the encrypted image for subsequent business processing, thereby completing the output processing of the encrypted image.

[0052] The present application first obtains the original image to be encrypted; then pre-processes the original image based on a preset pre-processing module to obtain a state matrix corresponding to the original image; then performs dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain the corresponding original key; subsequently performs random enhancement processing on the original key based on a preset random security enhancement module to obtain the corresponding target key; further encrypts the original image based on the target key to obtain an encrypted image corresponding to the original image; and finally outputs the encrypted image. After obtaining the original image to be encrypted, the present application pre-processes the original image using a pre-processing module to obtain a state matrix corresponding to the original image, then performs dynamic key generation processing on the state matrix based on a target annihilation neural network to obtain the original key, then performs random enhancement processing on the original key based on a random security enhancement module to obtain the target key, and then encrypts the original image based on the target key to obtain an encrypted image and outputs it. Through the above image encryption processing method, this application realizes end-to-end conversion from image to key and then to encrypted image by combining the preprocessing module, target zeroing neural network and random security enhancement module. The image encryption mechanism follows the core process chain of feature extraction-state evolution-key generation-security enhancement-image encryption. Each link is closely related to the image content, effectively ensuring the dynamic adaptability and security of the image encryption process.

[0053] In some optional implementations, step S202 includes the following steps:

[0054] The original image is processed into a unified format based on the preprocessing module to obtain a corresponding first image.

[0055] In this embodiment, the format unification process includes: raw Perform format unification processing and convert various formats into standard RGB or grayscale matrix Where H represents the image height, W represents the image width, and C represents the number of channels (3 for RGB and 1 for grayscale).

[0056] Perform pixel value normalization processing on the first image to obtain a corresponding second image.

[0057] In this embodiment, the pixel value normalization process includes: mapping the pixel intensity to the interval [0, 1] by linear transformation:

[0058]

[0059] Among them, x∈[1,H], y∈[1,W], c∈[1,C], where min(I) represents the minimum pixel value in image I and max(I) represents the maximum pixel value in image I.

[0060] Perform spatial feature extraction on the second image to obtain a corresponding feature volume.

[0061] In this embodiment, the specific implementation process of performing spatial feature extraction on the second image to obtain the corresponding feature volume will be described in further detail in subsequent specific embodiments of this application and will not be elaborated on here.

[0062] Downsampling is performed on the feature volume to obtain a corresponding multi-dimensional feature vector.

[0063] In this embodiment, the downsampling process includes downsampling by a mean pooling operation, which is defined as follows:

[0064]

[0065] Where s is the downsampling factor, k∈[1,K], floor denotes rounding down, and · denotes matrix multiplication. Downsampling by using mean pooling can effectively improve computational efficiency.

[0066] The multi-dimensional feature vector is reorganized to obtain the state matrix.

[0067] In this embodiment, the multi-dimensional feature tensor after downsampling is (where H' = floor(H / s), ) is reorganized into a state matrix As the initial condition for the subsequent target zeroing neural network:

[0068] Where n = H ′ ×W ′, m=K, Reshape represents the operation of converting a three-dimensional tensor into a two-dimensional matrix.

[0069] The present application performs format unification processing on the original image based on the preprocessing module to obtain the corresponding first image; then performs pixel value normalization processing on the first image to obtain the corresponding second image; then performs spatial feature extraction on the second image to obtain the corresponding feature volume; subsequently performs downsampling processing on the feature volume to obtain the corresponding multi-dimensional feature vector; finally, reorganizes the multi-dimensional feature vector to obtain the state matrix. The present application performs format unification processing, pixel value normalization processing, spatial feature extraction, downsampling processing and reorganization processing on the original image based on the use of the preprocessing module to achieve efficient and accurate preprocessing of the original image, which not only unifies the formats of images from different sources, but more importantly, extracts the key features of the image, ensures the accuracy of the obtained state matrix, and is conducive to laying a solid foundation for the subsequent dynamic key generation.

[0070] In some optional implementations of this embodiment, performing spatial feature extraction on the second image to obtain a corresponding feature volume includes the following steps:

[0071] Get the preset spatial feature extraction algorithm.

[0072] In this embodiment, there is no specific limitation on the selection of the above-mentioned spatial feature extraction algorithm, and any general algorithm with a spatial feature extraction function can be randomly selected according to actual business needs.

[0073] Call the preset number of convolution kernels.

[0074] In this embodiment, there is no specific limitation on the selection of the above preset number, which can be set according to actual business needs. For example, K specially designed convolution kernels {ω k}, k = 1 to K.

[0075] Based on the convolution kernel, the spatial feature extraction algorithm is used to extract spatial features from the second image to obtain corresponding local structural features and global statistical features.

[0076] In this embodiment, the local structural features and global statistical characteristics of the second image can be extracted using the selected spatial feature extraction algorithm based on the above convolution kernel. The convolution operation is defined as:

[0077]

[0078] Among them, k∈[1,K], r represents the radius of the convolution kernel (for a 3×3 convolution kernel, r=1). These features include key visual elements such as edges, textures, and color distribution, which together constitute the "feature fingerprint" of the image and form a feature volume

[0079] The local structural features and the global statistical features are integrated to obtain the feature volume.

[0080] In this embodiment, the local structural features and the global statistical features may be integrated to obtain corresponding integrated features, and the integrated features may be used as the feature volume.

[0081] The present application obtains a preset spatial feature extraction algorithm; then calls a preset number of convolution kernels; then, based on the convolution kernels, uses the spatial feature extraction algorithm to extract spatial features from the second image to obtain corresponding local structural features and global statistical features; and subsequently integrates the local structural features and the global statistical features to obtain the feature volume. The present application obtains a spatial feature extraction algorithm and calls a preset number of convolution kernels, then, based on the combination of the convolution kernels and the spatial feature extraction algorithm, performs spatial feature extraction on the second image to obtain local structural features and global statistical features, and then integrates the obtained local structural features and global statistical features, thereby achieving efficient and accurate generation of the required feature volume, ensuring the accuracy and standardization of the obtained feature volume.

[0082] In some optional implementations, the target annulment neural network includes a state updater, an association matrix calculator, and a key extractor; step S203 includes the following steps:

[0083] The state updater in the target annihilation neural network performs state update processing corresponding to the state matrix to obtain a corresponding dynamic state matrix.

[0084] In this embodiment, the state updater is responsible for performing the core dynamic calculations of the ZNN (target zeroing neural network). Based on the theory of zeroing neural networks, the state update process follows the differential equation:

[0085]

[0086] Where S(t)∈R n×m is the state matrix at time t, n and m represent the dimensions of the state space (usually corresponding to the dimensions of the preprocessed image features). The coefficient γ controls the speed of state update, and E(t) is the dynamic error function, defined as:

[0087] E(t)=S(t)-S target(t)+λ·C(t)

[0088] In this error function, S target(t) is the target state determined by the input image features, λ is a coefficient regulating the strength of the association, and C(t) is the correlation matrix describing the correlation between states. By iteratively solving this differential equation, the system evolves the state matrix S(t) in a direction that minimizes the error function, thereby generating a dynamic state sequence related to the input image content, namely the dynamic state matrix mentioned above.

[0089] An incidence matrix corresponding to the dynamic state matrix is ​​calculated based on the incidence matrix calculator.

[0090] In this embodiment, the correlation matrix calculator is responsible for capturing the nonlinear correlation between the dimensions in the state matrix. The correlation matrix C(t) is calculated by the formula C(t) = S(t) T W S(t) is calculated, where m(t) T is the transposed matrix of S(t), W∈R m×m is a trainable weight matrix. This design enables the system to automatically capture higher-order correlations in the state space, rather than being limited to first-order linear relationships in traditional methods. The weight matrix W is optimized through offline training, enabling it to extract key-related features of the image state in different application scenarios. The correlation matrix C(t) not only participates in the state update process but also serves as the core information source for key generation. It carries the inherent structural information of the state matrix and provides the "genetic genes" related to the image content for the dynamic key.

[0091] Based on the key extractor, a preset sliding window strategy is used to extract sub-blocks corresponding to the target function from the incidence matrix.

[0092] In this embodiment, the key extractor extracts the encryption key from the correlation matrix. It employs a sliding window strategy, extracting the original key from the correlation matrix via the function K(t) = ExtractBlock(C(t), i, j, k, l). ExtractBlock(C(t), i, j, k, l) extracts a k×l submatrix with (i, j) as the upper-left corner coordinates from the matrix C(t). This subblock is then extracted from the correlation matrix as the original key. The parameters i and j represent the starting position of the window. They are not fixed but dynamically determined based on the entropy of the state matrix. This means that even for the same state matrix, if its information distribution (entropy) changes, the position of the extracted key will also change, increasing the unpredictability of key generation. The parameters k and l determine the size of the extracted key, typically set to a number of bits that meets the requirements of the specific encryption algorithm, such as 128 bits or 256 bits. The extraction operation also employs loop boundary processing to ensure that even if the window exceeds the matrix boundaries, the complete key data can be retrieved through loops. This enables the system to process input images of any size.

[0093] The sub-block is used as the original key.

[0094] In this embodiment, existing image encryption methods using zeroing neural networks (ZNNs) typically rely on fixed key generation rules and fail to fully utilize the inherent dynamic properties of neural network states. This approach suffers from the following drawbacks: 1) Existing ZNN encryption models lack the ability to exploit high-order correlations in the state matrix during key expansion, resulting in a highly predictable key evolution path. Traditional ZNN encryption models employ fixed parameter update rules, making it impossible for keys to adaptively adjust based on image content. Existing methods only exploit the first-order properties of the ZNN state matrix and fail to exploit nonlinear correlations between multidimensional states. 3) Existing random mechanisms lack coordinated optimization with neural network states, potentially rendering keys unrecoverable. 4) Most schemes fail to establish a feedback link between the key and image features, making them vulnerable to chosen-plaintext attacks. In summary, while existing technologies have made some progress in the field of image encryption, they still suffer from key staticity, insufficient utilization of correlations, deficient randomness control, and a lack of dynamic adaptation. These issues render encryption systems insecure against modern cryptanalysis techniques, making it difficult to meet the growing demands of information security.

[0095] Based on the above-mentioned defects, this application proposes a dynamic key generation image encryption method based on the zeroing neural network (ZNN). The core innovation of this method lies in the adaptive key expansion mechanism driven by the correlation of the neural network state, which enables the key to adapt to the statistical characteristics of the input image, thereby enhancing the ability to resist cryptanalysis attacks. In addition, another innovation of the ZNN core layer (target zeroing neural network) is the adaptive step size mechanism. The time step of the key update Δt = η·‖E(t)‖ F The system dynamically adjusts the error function based on the Frobenius norm (the square root of the sum of the squares of all matrix elements), where η is the baseline step size coefficient. This design allows the system to quickly approach the target state using a larger step size when the error is large, while automatically reducing the step size for finer adjustments as convergence approaches, achieving a balance between computational efficiency and stability.

[0096] The present application performs a state update process corresponding to the state matrix based on the state updater in the target annihilation neural network to obtain a corresponding dynamic state matrix; then calculates the correlation matrix corresponding to the dynamic state matrix based on the correlation matrix calculator; then extracts the sub-block corresponding to the target function from the correlation matrix based on the key extractor using a preset sliding window strategy; and subsequently uses the sub-block as the original key. The present application calculates the state evolution according to the preset dynamic system equation by using the target annihilation neural network, and captures the high-dimensional nonlinear relationship between the state vectors through the correlation matrix. This correlation information is used to dynamically generate encryption keys, so that the keys are closely bound to the content characteristics of the input image, thereby achieving a deep coupling between the key generation process and the state evolution of the target annihilation neural network, so that each encryption operation generates a unique key sequence related to the image content, effectively resisting replay attacks and key cracking attempts, thereby effectively reducing the computational complexity of key generation and improving processing efficiency while ensuring encryption strength.

[0097] In some optional implementations, the random security enhancement module includes a chaotic pseudo-random number generator and a key mixing unit; step S204 includes the following steps:

[0098] The chaotic pseudo-random number generator in the random security enhancement module performs information coupling on the state matrix to generate a corresponding pseudo-random sequence.

[0099] In this embodiment, the chaotic pseudo-random number generator is based on an improved Lorenz chaotic system, which is dynamically associated with the image content by coupling state matrix information. The system equations include:

[0100]

[0101] Where σ, ρ, and β are parameters of the dynamic characteristics of the control system, which are usually set in the chaotic region to ensure the complexity and unpredictability of the output sequence. The parameter α controls the coupling strength between the state matrix and the chaotic system, and the function H(S(t)) calculates the entropy of the state matrix and is defined as H(S(t)) = -∑ i,j p i,j log(p i,j ), where p i,j is the normalized probability distribution of the elements in the matrix S(t), serving as an additional input to the chaotic system. This coupling design allows the chaotic system's trajectory to be associated with the image content while maintaining the inherent sensitivity and unpredictability of chaotic systems. The system numerically solves this set of differential equations to generate a series of pseudorandom values ​​P(t), resulting in a pseudorandom sequence used for subsequent key enhancement.

[0102] Get the preset combination method.

[0103] In this embodiment, the above combination method corresponds to the formula way, in which This combination preserves the image-dependent nature of the original key while introducing random perturbations of the chaotic system, significantly enhancing the key's security.

[0104] Based on the key mixing unit, the original key and the pseudo-random sequence are combined using the combination method to obtain corresponding combined data.

[0105] In this embodiment, the above-mentioned key mixing unit is used to combine the original key K(t) extracted from the target annihilation neural network with the pseudo-random sequence P(t) generated by the chaotic pseudo-random number generator using the above-mentioned combination method, and the obtained combined data is used as the final encryption key, that is, the above-mentioned target key.

[0106] The combined data is used as the target key.

[0107] In this embodiment, the target key K final (t) It is closely related to the image content and has sufficient randomness and unpredictability, and can effectively resist various cryptanalysis attacks.

[0108] The present application generates a corresponding pseudo-random sequence by information coupling the state matrix based on the chaotic pseudo-random number generator in the random security enhancement module; then obtains a preset combination method; then, based on the key mixing unit, uses the combination method to combine the original key and the pseudo-random sequence to obtain corresponding combination data; and subsequently uses the combination data as the target key. The present application generates a pseudo-random sequence by information coupling the state matrix based on the chaotic pseudo-random number generator in the random security enhancement module; then, based on the key mixing unit, uses the obtained combination method to combine the original key and the pseudo-random sequence to generate the final target key. The above combination method retains the image dependence characteristics of the original key, while introducing random perturbations of the chaotic system, significantly enhancing the security strength of the key. In addition, the generated target key is closely related to the image content of the original image, and has sufficient randomness and unpredictability, and can effectively resist various cryptanalysis attacks.

[0109] In some optional implementations of this embodiment, step S205 includes the following steps:

[0110] Get a custom diffusion operator.

[0111] In this embodiment, the selection of the above-mentioned diffusion operator can be determined according to actual business needs.

[0112] Call the encryption formula with the diffusion operator.

[0113] In this embodiment, the encryption formula includes: The diffusion operator involves three steps: row circular shift, column XOR, and pixel permutation. These operations combine to achieve a thorough mixing and diffusion of image pixels, ensuring that the statistical characteristics of the encrypted image are evenly distributed, making it difficult to restore the original image information through statistical analysis. The mod256 operation (where mod stands for modulo, ensuring the result value is between 0 and 255) ensures that the encrypted pixel values ​​remain within the valid 8-bit image range, facilitating storage and transmission.

[0114] The original image and the target key are encrypted based on the encryption formula to obtain an encrypted target image.

[0115] In this embodiment, an encryption operation may be performed on the original image and the target key based on the use of the confidential formula to obtain an encrypted target image as the final encrypted image.

[0116] The target image is used as the encrypted image.

[0117] This application obtains a custom diffusion operator; then calls an encryption formula associated with the diffusion operator; then encrypts the original image and the target key based on the encryption formula to obtain an encrypted target image; and subsequently uses the target image as the encrypted image. This application performs encryption operations on the original image using the enhanced target key and implements pixel-level hybrid transformations using a custom diffusion operator to output the final encrypted image, thereby achieving efficient and secure encryption of the original image and effectively improving the security and adaptability of image encryption.

[0118] In some optional implementations of this embodiment, after step S205, the electronic device may further perform the following steps:

[0119] Get the preset image storage policy.

[0120] In this embodiment, the above-mentioned image storage strategy may specifically use strategies such as cyclic overwriting or scheduled backup to ensure that the image data to be stored will not be lost due to insufficient storage space.

[0121] Recall the preset specified storage medium.

[0122] In this embodiment, there is no specific limitation on the selection of the above-mentioned designated storage medium, and it may include, for example, a local database, a disk, a cloud server, a blockchain, etc.

[0123] Based on the image storage policy, the encrypted image is stored in the designated storage medium.

[0124] In this embodiment, the encrypted image may be stored in the designated storage medium in a storage manner according to the image storage policy.

[0125] The present application obtains a preset image storage strategy, then calls a preset designated storage medium, and subsequently stores the encrypted image in the designated storage medium based on the image storage strategy. After encrypting the original image based on the use of the target key and obtaining an encrypted image corresponding to the original image, the present application also intelligently obtains a preset image storage strategy and calls a designated storage medium, and then stores the encrypted image based on the obtained image storage strategy and the designated storage medium. This effectively ensures the data security of the encrypted image and ensures that the target image will not be lost due to insufficient storage space, effectively improving the storage intelligence of the target image.

[0126] In some optional implementations, the user information obtained is obtained with the user's consent and complies with relevant laws and policies.

[0127] In addition, any software tools or components not provided by our company that appear in the embodiments of this application are merely examples and do not represent actual use.

[0128] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0129] It should be emphasized that in order to further ensure the privacy and security of the above-mentioned encrypted image, the above-mentioned encrypted image can also be stored in a node of a blockchain.

[0130] The blockchain referred to in this application is a new application model for computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Blockchain is essentially a decentralized database, a series of data blocks generated using cryptographic methods. Each data block contains information about a batch of network transactions, which is used to verify the validity of this information (to prevent counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, the platform product service layer, and the application service layer.

[0131] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0132] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0133] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware via computer-readable instructions. The computer-readable instructions can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes in the above-described method embodiments. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0134] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0135] Further references Figure 3 , as a response to the above Figure 2 The present application provides an embodiment of an image encryption device, which is similar to the embodiment of the image encryption device. Figure 2 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0136] like Figure 3 As shown, the image encryption device 300 of this embodiment includes: a first acquisition module 301, a pre-processing module 302, a generation module 303, an enhancement module 304, an encryption module 305 and an output module 306.

[0137] A first acquisition module 301 is used to acquire the original image to be encrypted;

[0138] A preprocessing module 302 is configured to preprocess the original image based on a preset preprocessing module to obtain a state matrix corresponding to the original image;

[0139] A generation module 303 is configured to perform dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain a corresponding original key;

[0140] An enhancement module 304 is configured to perform random enhancement processing on the original key based on a preset random security enhancement module to obtain a corresponding target key;

[0141] An encryption module 305 is configured to encrypt the original image based on the target key to obtain an encrypted image corresponding to the original image;

[0142] The output module 306 is used to output the encrypted image.

[0143] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0144] In some optional implementations of this embodiment, the preprocessing module 302 includes:

[0145] A first processing submodule, configured to perform format unification processing on the original image based on the preprocessing module to obtain a corresponding first image;

[0146] a second processing submodule, configured to perform pixel value normalization processing on the first image to obtain a corresponding second image;

[0147] A first extraction submodule is used to extract spatial features from the second image to obtain a corresponding feature volume;

[0148] A downsampling submodule, configured to perform downsampling processing on the feature volume to obtain a corresponding multidimensional feature vector;

[0149] The recombining submodule is used to recombinantly process the multi-dimensional feature vector to obtain the state matrix.

[0150] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0151] In some optional implementations of this embodiment, the first extraction submodule includes:

[0152] An acquisition unit, used for acquiring a preset spatial feature extraction algorithm;

[0153] A calling unit, used to call a preset number of convolution kernels;

[0154] an extraction unit, configured to perform spatial feature extraction on the second image based on the convolution kernel and using the spatial feature extraction algorithm to obtain corresponding local structural features and global statistical features;

[0155] An integration unit is used to perform feature integration on the local structural features and the global statistical features to obtain the feature volume.

[0156] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0157] In some optional implementations of this embodiment, the target annulment neural network includes a state updater, an association matrix calculator, and a key extractor; the generation module 303 includes:

[0158] An updating submodule, configured to perform a state update process corresponding to the state matrix based on the state updater in the target annihilation neural network to obtain a corresponding dynamic state matrix;

[0159] a calculation submodule, configured to calculate an incidence matrix corresponding to the dynamic state matrix based on the incidence matrix calculator;

[0160] A second extraction submodule, configured to extract a sub-block corresponding to a target function from the incidence matrix using a preset sliding window strategy based on the key extractor;

[0161] The first determining submodule is configured to use the sub-block as the original key.

[0162] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0163] In some optional implementations of this embodiment, the random security enhancement module includes a chaotic pseudo-random number generator and a key mixing unit; the enhancement module 304 includes:

[0164] A generating submodule, configured to perform information coupling on the state matrix based on the chaotic pseudo-random number generator in the random security enhancement module to generate a corresponding pseudo-random sequence;

[0165] The first acquisition submodule is used to acquire a preset combination mode;

[0166] a combining submodule, configured to combine the original key and the pseudo-random sequence using the combining method based on the key mixing unit to obtain corresponding combined data;

[0167] The second determining submodule is configured to use the combined data as the target key.

[0168] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0169] In some optional implementations of this embodiment, the encryption module 305 includes:

[0170] The second acquisition submodule is used to obtain a custom diffusion operator;

[0171] A calling submodule, used for calling the encryption formula of the diffusion operator;

[0172] an encryption submodule, configured to encrypt the original image and the target key based on the encryption formula to obtain an encrypted target image;

[0173] The third determining submodule is configured to use the target image as the encrypted image.

[0174] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0175] In some optional implementations of this embodiment, the image encryption apparatus further includes:

[0176] A second acquisition module is used to acquire a preset image storage strategy;

[0177] A calling module, used to call a preset designated storage medium;

[0178] A storage module is configured to store the encrypted image in the designated storage medium based on the image storage policy.

[0179] In this embodiment, the operations performed by the above modules or units correspond one-to-one to the steps of the image encryption method in the aforementioned embodiment, and are not described again here.

[0180] To solve the above technical problems, the present application also provides a computer device. Figure 4 , Figure 4 This is a basic structural block diagram of the computer device in this embodiment.

[0181] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 4 with components 41-43, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0182] The computer device may be a desktop computer, notebook computer, PDA, cloud server, etc. The computer device may interact with the user via a keyboard, mouse, remote control, touchpad, or voice control device.

[0183] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as the hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk equipped on the computer device 4, a smart memory card (SMC), a secure digital (SD) card, a flash memory card, etc. Of course, the memory 41 can also include both the internal storage unit of the computer device 4 and its external storage device. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer-readable instructions for the image encryption method. In addition, the memory 41 can also be used to temporarily store various types of data that have been output or are to be output.

[0184] In some embodiments, the processor 42 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to execute computer-readable instructions stored in the memory 41 or process data, such as computer-readable instructions for executing the image encryption method.

[0185] The network interface 43 may include a wireless network interface or a wired network interface. The network interface 43 is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0186] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0187] In an embodiment of the present application, first, the original image to be encrypted is obtained; then, the original image is preprocessed based on a preset preprocessing module to obtain a state matrix corresponding to the original image; then, the state matrix is ​​dynamically key generated based on a preset target annihilation neural network to obtain the corresponding original key; subsequently, the original key is randomly enhanced based on a preset random security enhancement module to obtain the corresponding target key; further, the original image is encrypted based on the target key to obtain an encrypted image corresponding to the original image; finally, the encrypted image is output. After obtaining the original image to be encrypted, the present application preprocesses the original image using a preprocessing module to obtain a state matrix corresponding to the original image, then dynamically key generates the state matrix based on a target annihilation neural network to obtain the original key, then randomly enhances the original key based on a random security enhancement module to obtain the target key, and then encrypts the original image based on the target key to obtain an encrypted image and outputs it. Through the above image encryption processing method, this application realizes end-to-end conversion from image to key and then to encrypted image by combining the preprocessing module, target zeroing neural network and random security enhancement module. The image encryption mechanism follows the core process chain of feature extraction-state evolution-key generation-security enhancement-image encryption. Each link is closely related to the image content, effectively ensuring the dynamic adaptability and security of the image encryption process.

[0188] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions. The computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the image encryption method as described above.

[0189] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0190] In an embodiment of the present application, first, the original image to be encrypted is obtained; then, the original image is preprocessed based on a preset preprocessing module to obtain a state matrix corresponding to the original image; then, the state matrix is ​​dynamically key generated based on a preset target annihilation neural network to obtain the corresponding original key; subsequently, the original key is randomly enhanced based on a preset random security enhancement module to obtain the corresponding target key; further, the original image is encrypted based on the target key to obtain an encrypted image corresponding to the original image; finally, the encrypted image is output. After obtaining the original image to be encrypted, the present application preprocesses the original image using a preprocessing module to obtain a state matrix corresponding to the original image, then dynamically key generates the state matrix based on a target annihilation neural network to obtain the original key, then randomly enhances the original key based on a random security enhancement module to obtain the target key, and then encrypts the original image based on the target key to obtain an encrypted image and outputs it. Through the above image encryption processing method, this application realizes end-to-end conversion from image to key and then to encrypted image by combining the preprocessing module, target zeroing neural network and random security enhancement module. The image encryption mechanism follows the core process chain of feature extraction-state evolution-key generation-security enhancement-image encryption. Each link is closely related to the image content, effectively ensuring the dynamic adaptability and security of the image encryption process.

[0191] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0192] Obviously, the embodiments described above are only some of the embodiments of the present application, rather than all of the embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present application specification and the accompanying drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present application.

Claims

1. An image encryption method, characterized in that: The steps include: Obtain the original image to be encrypted; Preprocessing the original image based on a preset preprocessing module to obtain a state matrix corresponding to the original image; Performing dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain a corresponding original key; Performing random enhancement processing on the original key based on a preset random security enhancement module to obtain a corresponding target key; encrypting the original image based on the target key to obtain an encrypted image corresponding to the original image; Output processing is performed on the encrypted image.

2. The image encryption method according to claim 1, wherein: The step of preprocessing the original image based on a preset preprocessing module to obtain a state matrix corresponding to the original image specifically includes: Performing format unification processing on the original image based on the preprocessing module to obtain a corresponding first image; performing pixel value normalization processing on the first image to obtain a corresponding second image; performing spatial feature extraction on the second image to obtain a corresponding feature volume; Downsampling the feature volume to obtain a corresponding multi-dimensional feature vector; The multi-dimensional feature vector is reorganized to obtain the state matrix.

3. The image encryption method according to claim 2, wherein: The step of extracting spatial features from the second image to obtain a corresponding feature volume specifically includes: Get the preset spatial feature extraction algorithm; Call the preset number of convolution kernels; Based on the convolution kernel, using the spatial feature extraction algorithm to extract spatial features from the second image to obtain corresponding local structural features and global statistical features; The local structural features and the global statistical features are integrated to obtain the feature volume.

4. The image encryption method according to claim 1, wherein: The target annihilation neural network includes a state updater, an association matrix calculator, and a key extractor; the step of performing dynamic key generation processing on the state matrix based on the preset target annihilation neural network to obtain the corresponding original key specifically includes: Executing a state update process corresponding to the state matrix based on the state updater in the target annihilation neural network to obtain a corresponding dynamic state matrix; Calculating an incidence matrix corresponding to the dynamic state matrix based on the incidence matrix calculator; Based on the key extractor, a preset sliding window strategy is used to extract a sub-block corresponding to the objective function from the incidence matrix; The sub-block is used as the original key.

5. The image encryption method according to claim 1, wherein: The random security enhancement module includes a chaotic pseudo-random number generator and a key mixing unit; the step of performing random enhancement processing on the original key based on the preset random security enhancement module to obtain the corresponding target key specifically includes: Performing information coupling on the state matrix based on the chaotic pseudo-random number generator in the random security enhancement module to generate a corresponding pseudo-random sequence; Get the preset combination method; Based on the key mixing unit, the original key and the pseudo-random sequence are combined using the combination method to obtain corresponding combined data; The combined data is used as the target key.

6. The image encryption method according to claim 1, wherein: The step of encrypting the original image based on the target key to obtain an encrypted image corresponding to the original image specifically includes: Get a custom diffusion operator; Invoking an encryption formula with the diffusion operator; encrypting the original image and the target key based on the encryption formula to obtain an encrypted target image; The target image is used as the encrypted image.

7. The image encryption method according to claim 1, wherein: After the step of encrypting the original image based on the target key to obtain an encrypted image corresponding to the original image, the method further includes: Get the preset image storage strategy; Call the preset specified storage medium; Based on the image storage policy, the encrypted image is stored in the designated storage medium.

8. An image encryption device, characterized in that: include: A first acquisition module, configured to acquire an original image to be encrypted; A preprocessing module, configured to preprocess the original image based on a preset preprocessing module to obtain a state matrix corresponding to the original image; A generation module, configured to perform dynamic key generation processing on the state matrix based on a preset target annihilation neural network to obtain a corresponding original key; An enhancement module, configured to perform random enhancement processing on the original key based on a preset random security enhancement module to obtain a corresponding target key; an encryption module, configured to encrypt the original image based on the target key to obtain an encrypted image corresponding to the original image; An output module is used to perform output processing on the encrypted image.

9. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores computer-readable instructions, and the processor implements the steps of the image encryption method according to any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the image encryption method according to any one of claims 1 to 7.

Images