A multi-dimensional image encryption method and system based on chaotic encoding and global diffusion

Abstract

The application belongs to the technical field of image encryption, and discloses a multi-dimensional image encryption method and system based on chaotic coding and global diffusion, which comprises the following steps: obtaining target image data to be encrypted, and performing structured representation on the target image data to organize the image data into multi-dimensional image data containing spatial dimensions, channel dimensions and hierarchical dimensions; selecting a Logistic mapping chaotic system, initializing the initial state of the chaotic system based on a preset key parameter, and iteratively generating a chaotic state value sequence matched with the multi-dimensional image data; in the encryption process of the multi-dimensional image data, performing nonlinear rearrangement and diffusion processing on the position of each pixel in a global range, dynamically determining the jump scale of pixel diffusion according to the chaotic state value sequence, and obtaining diffused multi-dimensional image data; under the premise of ensuring controllable and reversible encryption and decryption processes, the security and complexity of image encryption are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of image encryption technology, and more specifically, to a multidimensional image encryption method and system based on chaotic coding and global diffusion. Background Technology

[0002] Existing multidimensional image encryption methods and systems mainly suffer from the following problems:

[0003] With the widespread application of image data in information transmission, data storage, and secure communication, image encryption technology has gradually become an important means of ensuring image data security. Especially with the increasing number of multi-channel, multi-level images and high-dimensional image data, how to perform high-security encryption on multi-dimensional image data containing spatial, channel, and hierarchical dimensions has become a key focus in related technical fields.

[0004] In existing image encryption methods, chaotic systems, due to their initial value sensitivity and pseudo-random characteristics, are often used to generate the scrambling sequences or diffusion control parameters required for image encryption. However, in existing chaotic image encryption technologies, chaotic sequences are usually only used to generate fixed or quasi-fixed pixel scrambling orders or diffusion rules. The diffusion process is mostly limited to local propagation of adjacent pixels or fixed step sizes, making it difficult to achieve rapid diffusion of pixel information globally. This limits its resistance to differential attacks and statistical analysis attacks.

[0005] Furthermore, when encrypting multidimensional image data, existing technologies often simplify multidimensional images into two-dimensional planar structures for processing, failing to fully utilize the overall structural characteristics of multidimensional images in spatial, channel, and hierarchical dimensions. This results in a single diffusion path, a fixed diffusion scale, and relatively obvious correlations between pixels. Attackers may be able to reverse-engineer by analyzing pixel correlations, thereby reducing the overall security of the encryption system.

[0006] Meanwhile, existing diffusion mechanisms typically use preset or statically set relationships between pixels, lacking a mechanism to adapt to the dynamic evolution of chaotic systems. This results in insufficient randomness and unpredictability in the diffusion process. When partial information about the key or initial parameters is leaked, the diffusion rules can be easily deduced, leading to a decrease in overall encryption strength.

[0007] In image encryption processing where information loss is a concern, existing methods often employ fixed quantization or compression techniques, lacking dynamic control mechanisms related to keys or chaotic sequences. This allows encrypted images to retain certain statistical regularities, increasing the risk of analysis and cracking. Furthermore, existing technologies struggle to accurately compensate for errors introduced by quantization or rounding operations during decryption, easily leading to uncontrollable distortion in the decrypted image and affecting image restoration quality and application effectiveness.

[0008] In view of this, the present invention proposes a multidimensional image encryption method based on chaotic coding and global diffusion to solve the above problems. Summary of the Invention

[0009] To overcome the aforementioned shortcomings of existing technologies and to achieve the above objectives, this invention provides the following technical solution: a multidimensional image encryption method based on chaotic coding and global diffusion, comprising:

[0010] S1. Obtain the target image data to be encrypted, and perform a structured representation of the target image data, organizing the image data into multi-dimensional image data containing spatial dimension, channel dimension and hierarchical dimension;

[0011] S2. Select a Logistic mapping chaotic system and initialize the initial state of the chaotic system based on preset key parameters, and iteratively generate a sequence of chaotic state values ​​that match the multidimensional image data.

[0012] S3. In the process of encrypting multidimensional image data, the position of each pixel is subjected to global nonlinear rearrangement and diffusion processing. The jump scale of pixel diffusion is dynamically determined according to the chaotic state value sequence to obtain the diffused multidimensional image data.

[0013] S4. Information loss processing is performed on the diffused multidimensional image data. By performing controlled numerical compression or quantization operations on the pixel values, information loss is formed on the surface during the encryption stage to generate encrypted image data.

[0014] S5. Output and store the encrypted image data, and in the decryption stage, call the corresponding chaotic coding parameters and compensation rules to reverse process the encrypted image data to recover the multidimensional image data.

[0015] Preferably, the method for obtaining the target image data to be encrypted includes:

[0016] The target image data is acquired from image acquisition devices or image files, including digital cameras, scanners and medical imaging devices, and image files including BMP, PNG, JPEG and TIFF digital image formats.

[0017] Preferably, the method for acquiring the multidimensional image data includes:

[0018] The target image data is parsed through an image reading interface and converted into a digital pixel matrix that can be processed by a computer; the grayscale value of each pixel in the target image is digitally represented and organized into a two-dimensional or three-dimensional array according to the row and column order of the target image.

[0019] The digital pixel matrix is ​​represented in a structured way, and the row and column coordinates of the target image are organized as two-dimensional spatial dimensions. A unique spatial index is assigned to each pixel to describe the positional relationship of the pixels in two-dimensional space.

[0020] The color channels of the target image are organized as channel dimensions, and a unique index is assigned to each pixel in the channel dimension. Then, the different frames or levels of the target image are organized as level dimensions, and a unique index is assigned to each pixel in the level dimension.

[0021] Finally, a unique index is generated for each pixel in the target image data, and an index mapping relationship is established for spatial dimension, channel dimension and hierarchical dimension to obtain multidimensional image data containing spatial dimension, channel dimension and hierarchical dimension.

[0022] Preferably, the method for initializing the initial state of the chaotic system includes:

[0023] Logistic mapping is chosen as the basic model of chaotic system. The initial state of chaotic system is initialized based on preset key parameters, which include initial state values ​​and system control parameters. Initial state seeds and control parameter seeds are extracted from the preset key parameters.

[0024] The initial state seed is mapped to the effective initial value range of the Logistic map chaotic system to obtain the initial state value of the chaotic system; the control parameter seed is mapped to the effective range of the control parameters of the chaotic system to obtain the system control parameters; the initial state value and the system control parameters are mapped to the effective numerical range of the chaotic system by the key.

[0025] Preferably, the method for obtaining the chaotic state value sequence includes:

[0026] The length of the chaotic state value sequence is determined based on the total number of pixels or the matrix size of the multidimensional image data. By iterating the Logistic mapping formula, a chaotic state value sequence consistent with the number of pixels in the multidimensional image data is generated iteratively, with each chaotic state value corresponding to a pixel in the multidimensional image data.

[0027] Preferably, the method for acquiring the diffused multidimensional image data includes:

[0028] Multidimensional image data is linearized and mapped according to a preset dimensional expansion rule, and a unique linear index number is assigned to each pixel in the multidimensional image data to uniformly describe the positional relationship of each pixel in the global diffusion process; based on the chaotic state value generated by the Logistic mapping chaotic system in the corresponding iteration step, the corresponding diffusion jump scale is dynamically determined for each pixel.

[0029] The diffusion jump scale is determined by the chaotic state value and the preset maximum diffusion jump range, so that different pixels have different diffusion spans during the diffusion process, and the diffusion span covers the global range of the multidimensional image data; after determining the diffusion jump scale of the pixel, a global diffusion mapping relationship between pixels is established based on the diffusion jump scale.

[0030] By using the diffusion jump scale and global diffusion mapping relationship dynamically controlled by the Logistic mapping chaotic system, the numerical change of any pixel in the multidimensional image data can be propagated to pixel positions at different distances during the diffusion process, thus obtaining the diffused multidimensional image data.

[0031] Preferably, the method for generating encrypted image data includes:

[0032] For each pixel value in the diffused multidimensional image arranged in a predetermined scanning order, a chaotic perturbation sequence corresponding one-to-one with the pixel index is generated based on a key-driven chaotic system, and the chaotic perturbation sequence is introduced into the pixel transformation process as a quantization offset factor.

[0033] During the pixel transformation process, information loss transformation is performed on the pixel values ​​in the diffused multidimensional image data. A proportional modulation related to the compression or quantization intensity is applied to the current pixel value, and a controlled numerical compression or quantization operation is performed after superimposing chaotic perturbations. This causes limited information loss in the pixel values ​​during the encryption stage, thereby obtaining encrypted image data.

[0034] Preferably, the method for outputting and storing encrypted image data includes:

[0035] The encrypted image data is organized and encapsulated according to a preset data format. The encapsulation can be implemented using existing image file formats (such as PNG, BMP, TIFF, etc.) while maintaining the pixel arrangement order.

[0036] Encrypted image data can be stored on local storage devices, servers, or remote cloud storage systems, supporting the saving, cross-platform access, or retrieval of encrypted image data. During output and storage, encrypted image data can be transmitted through standardized interfaces, while integrity verification ensures that the encrypted image data has not been tampered with.

[0037] Preferably, the method for recovering multidimensional image data includes:

[0038] During the decryption phase, the same chaotic system parameters as those in the encryption phase are invoked to perform synchronous initialization and iterative calculations on the chaotic system, and a chaotic perturbation sequence corresponding one-to-one with the pixel index in the encryption phase is regenerated to compensate for the quantization or rounding errors introduced by the information loss-type pixel transformation in the encryption phase.

[0039] For each pixel value in the encrypted image data, a pixel recovery operation based on chaos compensation is performed using a chaotic perturbation sequence and compression or quantization parameters consistent with the encryption stage. This reverse recovery process is then applied to the encrypted pixel values ​​to recover the multidimensional image data.

[0040] A multidimensional image encryption system based on chaotic coding and global diffusion includes:

[0041] The image acquisition and representation module is used to acquire the target image data to be encrypted and to perform a structured representation of the target image data, organizing the image data into multi-dimensional image data containing spatial dimension, channel dimension and hierarchical dimension;

[0042] The parameter initialization module is used to select a Logistic map chaotic system, initialize the initial state of the chaotic system based on a preset key parameter, and iteratively generate a sequence of chaotic state values ​​that match the multidimensional image data.

[0043] The global diffusion jump module is used to perform global nonlinear rearrangement and diffusion processing on the position of each pixel during the encryption process of multidimensional image data. It dynamically determines the jump scale of pixel diffusion based on the chaotic state value sequence to obtain the diffused multidimensional image data.

[0044] The information loss encryption module is used to process the information loss of the diffused multidimensional image data. By performing controlled numerical compression or quantization on the pixel values, information loss is formed on the surface during the encryption stage to generate encrypted image data.

[0045] The encrypted output storage module is used to output and store encrypted image data, and in the decryption stage, it calls the corresponding chaotic coding parameters and compensation rules to reverse process the encrypted image data to recover the multidimensional image data.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] This invention linearizes multidimensional image data and introduces a diffusion jump scale dynamically controlled by a Logistic mapping chaotic system. This allows each pixel to correspond to a different diffusion span during the diffusion process, thus breaking through the traditional local diffusion mode based on adjacent pixels or fixed step sizes. It establishes diffusion relationships between distant pixels within the global scope of the multidimensional image data. The diffusion jump scale is jointly determined by the chaotic state value and the maximum diffusion jump range, causing the diffusion path to dynamically change with the iterative state of the chaotic system. The diffusion relationships between different pixels are no longer fixed, thereby improving the nonlinear characteristics and unpredictability of the diffusion process and effectively enhancing the encryption image's resistance to differential attacks. It enhances resistance to statistical analysis attacks; by establishing a global diffusion mapping relationship based on the diffusion jump scale, the pixel value of the current pixel is not only related to itself during the diffusion process, but also coupled with the diffusion results of other pixels selected by the chaotic system. This allows minute numerical changes in any pixel to propagate to pixel positions at different distances during the diffusion process, achieving multi-scale global diffusion of multi-dimensional image data. Since the diffusion process takes place within the global scope of the multi-dimensional image data, it can effectively break the original correlation of multi-dimensional images in spatial, channel, and hierarchical dimensions, structurally reducing the identifiability of statistical features of image data and improving the overall security and robustness of encrypted images.

[0048] By dynamically controlling the pixel jump scale using chaotic sequences, the diffusion operation covers the entire global range of the multidimensional image, achieving cross-distance, nonlinear, and multi-scale pixel coupling. Even if there are minor changes in the local image content, they can be rapidly propagated globally, enhancing resistance to differential attacks and the security of statistical analysis. In the encryption stage, a key-driven chaotic perturbation sequence is introduced as a quantization offset factor, making the information-loss transformation not only controlled by compression / quantization parameters but also key-dependent. The chaotic perturbation sequence and compression operation work together to enhance the unpredictability of the ciphertext, making it impossible to accurately deduce the original pixel distribution under unknown key conditions. In the decryption stage, the encrypted pixel values ​​are reversed using the same chaotic perturbation term, compression parameters, and compensation correction term as in the encryption stage. This compensates for information loss and rounding errors; under known key conditions, the recovered pixel values ​​are ideally identical to the diffused pixels, ensuring the reversibility of image decryption. The combination of chaotic-controlled global diffusion and information-loss pixel transformation makes the encrypted image both statistically unpredictable and accurately recoverable under legitimate key conditions. It effectively solves the problems of local diffusion, insufficient key sensitivity, and irreversible information loss in traditional multidimensional image encryption. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the process of a multidimensional image encryption method based on chaotic coding and global diffusion according to the present invention;

[0050] Figure 2 This is a schematic diagram of a multidimensional image encryption system based on chaotic coding and global diffusion according to the present invention. Detailed Implementation

[0051] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Example 1

[0053] Please see Figure 1 As shown, this embodiment provides a multidimensional image encryption method based on chaotic coding and global diffusion, specifically including the following steps:

[0054] S1. Obtain the target image data to be encrypted, and perform a structured representation of the target image data, organizing the image data into multi-dimensional image data containing spatial dimension, channel dimension and hierarchical dimension;

[0055] S2. Select a Logistic mapping chaotic system and initialize the initial state of the chaotic system based on preset key parameters, and iteratively generate a sequence of chaotic state values ​​that match the multidimensional image data.

[0056] S3. In the process of encrypting multidimensional image data, the position of each pixel is subjected to global nonlinear rearrangement and diffusion processing. The jump scale of pixel diffusion is dynamically determined according to the chaotic state value sequence to obtain the diffused multidimensional image data.

[0057] S4. Information loss processing is performed on the diffused multidimensional image data. By performing controlled numerical compression or quantization operations on pixel values, information loss is formed on the surface during the encryption stage to generate encrypted image data.

[0058] S5. Output and store the encrypted image data, and in the decryption stage, call the corresponding chaotic coding parameters and compensation rules to reverse process the encrypted image data to recover the multidimensional image data.

[0059] Methods for obtaining target image data to be encrypted include:

[0060] The target image data is acquired from image acquisition devices or image files, including digital cameras, scanners and medical imaging devices, and image files including BMP, PNG, JPEG and TIFF digital image formats.

[0061] Methods for acquiring multidimensional image data include:

[0062] The target image data is parsed through the image reading interface and converted into a digital pixel matrix that can be processed by a computer; the grayscale value of each pixel in the target image is digitally represented and organized into a two-dimensional or three-dimensional array according to the row and column order of the target image.

[0063] It should be noted that after acquiring the target image data, the image data is parsed through the image reading interface on the computer. For image files, the interface first reads the file header and metadata to obtain the image size, number of channels, color mode, and compression method, and then decodes the compressed image to restore the encoded pixel information to the original pixel values. For digital signals output by the image acquisition device, the interface reads the acquisition frame or scan line and maps the digital signal output by the device to pixel intensity values. After parsing, grayscale values ​​or color channel values ​​are extracted for each pixel of the image. Each pixel in a grayscale image corresponds to a single grayscale value, while each pixel in a color image corresponds to multiple channel values ​​(such as red, green, and blue). The extraction process traverses pixel by pixel in the row and column order of the image to ensure that the spatial position relationship of the pixels matches the original image. Figure 1 The extracted pixel values ​​are organized into a digital pixel matrix in row and column order. For grayscale images, this results in a two-dimensional matrix, where each element represents the grayscale value of the corresponding pixel; for color images, it results in a three-dimensional matrix, where the third dimension represents the color channels, and each element is the value for that channel. For multi-frame image sequences or multi-level images, this can be expanded into a four-dimensional or higher-dimensional array to maintain temporal frame or hierarchical index relationships. The resulting digital pixel matrix is ​​a standard data structure; the value corresponding to each pixel can be directly processed by the computer, preserving the spatial location and channel information of the image.

[0064] The digital pixel matrix is ​​represented in a structured way, and the row and column coordinates of the target image are organized as two-dimensional spatial dimensions. A unique spatial index is assigned to each pixel to describe the positional relationship of the pixels in two-dimensional space.

[0065] The color channels of the target image are organized as channel dimensions, and a unique index is assigned to each pixel in the channel dimension to achieve independent management and unified processing of multi-channel information; then, different frames or levels of the target image are organized as level dimensions, and a unique index is assigned to each pixel in the level dimension to achieve independent control or joint processing of pixels at different levels.

[0066] Finally, a unique index is generated for each pixel in the target image data, and an index mapping relationship is established for spatial dimension, channel dimension and hierarchical dimension to obtain multidimensional image data containing spatial dimension, channel dimension and hierarchical dimension.

[0067] Methods for initializing the initial state of a chaotic system include:

[0068] Logistic mapping is chosen as the basic model of chaotic system. The initial state of chaotic system is initialized based on preset key parameters, which include initial state values ​​and system control parameters. Initial state seeds and control parameter seeds are extracted from the preset key parameters.

[0069] The initial state seed is mapped to the effective initial value range of the Logistic map chaotic system to obtain the initial state value of the chaotic system; the control parameter seed is mapped to the effective range of the control parameters of the chaotic system to obtain the system control parameters; the initial state value and system control parameters are mapped to the effective numerical range of the chaotic system by the key to ensure the key sensitivity and unpredictability of the chaotic sequence.

[0070] Methods for obtaining chaotic state value sequences include:

[0071] The length of the chaotic state value sequence is determined based on the total number of pixels or the matrix size of the multidimensional image data. By iterating the Logistic mapping formula, a chaotic state value sequence with the same number of pixels as the multidimensional image data is generated iteratively, with each chaotic state value corresponding to a pixel in the multidimensional image data.

[0072] The iterative Logistic mapping formula is: ;in, Indicates the chaotic system at the th The chaotic state value obtained in the next iteration; These represent system control parameters to ensure the system remains in a chaotic state, determining the nonlinear characteristics and degree of chaos of the iterative sequence; Indicates the chaotic system at the th The chaotic state value obtained in the next iteration; The nonlinear modulation term representing the state value is used to introduce the feedback characteristics of the system, making the output sequence sensitive to the initial value and exhibiting chaotic behavior. An index representing the number of iterations.

[0073] Methods for obtaining diffused multidimensional image data include:

[0074] Multidimensional image data is linearized and mapped according to a preset dimensional expansion rule, and a unique linear index number is assigned to each pixel in the multidimensional image data to uniformly describe the positional relationship of each pixel in the global diffusion process; based on the chaotic state value generated by the Logistic mapping chaotic system in the corresponding iteration step, the corresponding diffusion jump scale is dynamically determined for each pixel.

[0075] The diffusion jump scale is: ;in, Indicates the first The diffusion jump scale (jump step size) corresponding to each pixel during the diffusion process. This represents the chaotic state value output by the iterative output of the Logistic Mapping chaotic system; This represents the maximum diffusion jump scale, used to limit the maximum range of the diffusion jump. Its value is related to the total number of image pixels or the current unfolded dimension length. This indicates the floor function; Represents the linear index number of a pixel;

[0076] The diffusion jump scale is determined by the chaotic state value and the preset maximum diffusion jump range, so that different pixels have different diffusion spans during the diffusion process, and the diffusion span covers the global range of the multidimensional image data. After determining the diffusion jump scale of a pixel, a global diffusion mapping relationship between pixels is established based on the diffusion jump scale. This makes the pixel value of the current pixel not only related to the pixel itself during the diffusion process, but also related to the diffusion results of other pixels selected in the global range according to the diffusion jump scale, thereby forming a nonlinear coupling across distance in the multidimensional image data.

[0077] The global diffusion mapping relationship is as follows: ;in, Indicates the number of diffusions after the first diffusion. The pixel value of each pixel; Indicates the number of times before diffusion The pixel value of each pixel; This indicates a bitwise XOR operation; Indicates the index of the current pixel. Reference pixel indices that have a skip relationship; Modulo operation is used to implement wraparound during index calculation. When the index is outside the range, the index is remapped to a valid pixel index range using a modulo operation; Indicates the total number of pixels in a multidimensional image;

[0078] By using the diffusion jump scale and global diffusion mapping relationship dynamically controlled by the Logistic mapping chaotic system, the numerical change of any pixel in the multidimensional image data can be propagated to pixel positions at different distances during the diffusion process, thus obtaining the diffused multidimensional image data.

[0079] The following technical problems of existing technologies are solved: In existing chaotic image encryption methods, chaotic sequences are usually only used to generate fixed or quasi-fixed scrambling orders or diffusion rules. The diffusion process is mostly limited to local propagation of adjacent pixels or fixed step sizes, making it difficult to achieve rapid diffusion of pixel information on a global scale, resulting in limited resistance to differential attacks and statistical analysis attacks; In the encryption process of multidimensional image data (such as images containing spatial, channel, and hierarchical dimensions), existing technologies often simplify multidimensional images to a two-dimensional plane for processing, failing to fully utilize the complexity of the overall structure of multidimensional images, resulting in a single diffusion path and a fixed diffusion scale, making it easy for attackers to reverse infer by analyzing pixel relationships; The relationships between pixels in existing diffusion mechanisms are usually preset or static and do not change with the dynamic evolution of the chaotic system, resulting in insufficient randomness and unpredictability of the diffusion process. When some information of the key or initial parameters is leaked, the overall encryption security decreases.

[0080] Compared to existing technologies, the advantages are as follows: By linearizing and mapping multidimensional image data and introducing a diffusion jump scale dynamically controlled by a Logistic mapping chaotic system, each pixel corresponds to a different diffusion span during the diffusion process. This breaks through the traditional local diffusion mode based on adjacent pixels or fixed step size, establishing diffusion correlations between distant pixels within the global scope of multidimensional image data. The diffusion jump scale is jointly determined by the chaotic state value and the maximum diffusion jump range, causing the diffusion path to dynamically change with the iterative state of the chaotic system. The diffusion relationship between different pixels is no longer fixed, thereby improving the nonlinear characteristics and unpredictability of the diffusion process and effectively enhancing the encryption of images. It possesses resistance to differential and statistical analysis attacks; by establishing a global diffusion mapping relationship based on the diffusion jump scale, the pixel value of the current pixel is not only related to itself during the diffusion process, but also coupled with the diffusion results of other pixels selected by the chaotic system. This allows minute numerical changes in any pixel to propagate to pixel positions at different distances during the diffusion process, achieving multi-scale global diffusion of multi-dimensional image data; since the diffusion process takes place globally within the multi-dimensional image data, it can effectively break the original correlations of multi-dimensional images in spatial, channel, and hierarchical dimensions, structurally reducing the identifiability of statistical features of image data and improving the overall security and robustness of encrypted images.

[0081] Methods for generating encrypted image data include:

[0082] For each pixel value in the diffused multidimensional image arranged in a predetermined scanning order, a chaotic perturbation sequence corresponding one-to-one with the pixel index is generated based on a key-driven chaotic system, and the chaotic perturbation sequence is introduced into the pixel transformation process as a quantization offset factor.

[0083] During the pixel transformation process, information loss transformation is performed on the pixel values ​​in the diffused multidimensional image data. A proportional modulation related to the compression or quantization intensity is applied to the current pixel value, and a controlled numerical compression or quantization operation is performed after superimposing chaotic perturbations. This causes the pixel values ​​to produce limited information loss during the encryption stage, thereby obtaining encrypted image data.

[0084] Information loss type transformation is: ;in, This represents the result obtained after performing an information loss type transformation. Encrypted pixel values ​​for each pixel; This indicates compression or quantization control parameters. It is a real or rational number greater than 1, used to adjust the compression strength of pixel values ​​during information loss transformation, thereby controlling the degree of information loss generated during the encryption stage; This represents the perturbation term generated by the key-driven chaotic system, in relation to the pixel index. One-to-one correspondence is used to introduce key-related numerical offsets during pixel transformation, making information-lossy transformations unpredictable and key-sensitive.

[0085] The compression or quantization intensity is determined by preset parameters, which controls the degree of information loss, making it impossible to accurately deduce the pixel value distribution after diffusion under unknown key conditions; the chaotic perturbation sequence works together with the compression or quantization operation to make the information loss process have key correlation and inter-pixel correlation.

[0086] Preferably, the method for outputting and storing encrypted image data includes:

[0087] Encrypted image data is organized and encapsulated according to a preset data format to ensure that encrypted pixel values ​​are not modified during storage and transmission; encapsulation can be implemented using existing image file formats (such as PNG, BMP, TIFF, etc.) while maintaining the pixel arrangement order;

[0088] Encrypted image data can be stored on local storage devices, servers, or remote cloud storage systems, supporting the saving, cross-platform access, or retrieval of encrypted image data. During output and storage, encrypted image data can be transmitted through standardized interfaces, while integrity checks (such as hash, CRC, etc.) ensure that the encrypted image data has not been tampered with.

[0089] Methods for recovering multidimensional image data include:

[0090] During the decryption phase, the same chaotic system parameters as those in the encryption phase are invoked to perform synchronous initialization and iterative calculations on the chaotic system, and a new chaotic perturbation sequence corresponding one-to-one with the pixel index in the encryption phase is generated to compensate for the quantization or rounding errors introduced by the information loss-type pixel transformation in the encryption phase.

[0091] For each pixel value in the encrypted image data, a pixel recovery operation based on chaos compensation is performed using a chaotic perturbation sequence and compression or quantization parameters consistent with the encryption stage. This reverse recovery process is then applied to the encrypted pixel values ​​to recover the multidimensional image data.

[0092] Pixel recovery operation is as follows: ;in, This indicates the number of decryption stages that were recovered. The pixel value of the nth pixel, under ideal compensation conditions, is equivalent to the pixel value of the th pixel after diffusion. Pixel value of each pixel However, given the rounding errors inherent in actual rounding or quantization, and There are controllable, minute deviations between them; This indicates a compensation correction term, used to correct rounding errors caused by rounding or quantization operations during the encryption phase;

[0093] This solution addresses the following technical problems in existing technologies: Existing multidimensional image encryption methods typically rely on fixed step sizes or local diffusion strategies, causing pixel value changes to propagate only within local neighborhoods, making it difficult to achieve global nonlinear coupling. Information-depleting operations are often fixed quantization or compression, lacking key correlation, making the ciphertext susceptible to partial cracking under statistical analysis or differential attacks. During decryption, traditional methods struggle to accurately recover pixel values, failing to effectively compensate for errors introduced by quantization or rounding, resulting in uncontrollable distortion in the decrypted image. Existing methods have weak dependence on the key during encryption; even if an attacker intercepts the ciphertext, they may be able to partially reconstruct the image content by analyzing pixel patterns. Traditional diffusion is limited to adjacent pixels, failing to achieve nonlinear, multi-scale pixel coupling globally, thus reducing resistance to differential attacks.

[0094] The advantages over existing technologies include: Utilizing chaotic sequences to dynamically control pixel jump scales allows diffusion operations to cover the entire global range of the multidimensional image, achieving cross-distance, nonlinear, and multi-scale pixel coupling. Even minor local changes in image content can propagate rapidly globally, enhancing resistance to differential attacks and statistical analysis security. In the encryption stage, a key-driven chaotic perturbation sequence is introduced as a quantization offset factor, ensuring that information-loss-type transformations are not only controlled by compression / quantization parameters but also have key correlation. The combined effect of the chaotic perturbation sequence and compression operations enhances the unpredictability of the ciphertext, making it impossible to accurately deduce the original pixel distribution under unknown key conditions. In the decryption stage, the encrypted pixel values ​​are reverse-recovered using chaotic perturbation terms, compression parameters, and compensation correction terms consistent with the encryption stage. This compensates for information loss and rounding errors; under known key conditions, the recovered pixel values ​​are ideally identical to the diffused pixels, guaranteeing the reversibility of image decryption. The combination of chaotic-controlled global diffusion and information-loss-type pixel transformations makes the encrypted image both statistically unpredictable and accurately recoverable under legitimate key conditions. It effectively solves the problems of local diffusion, insufficient key sensitivity, and irreversible information loss in traditional multidimensional image encryption.

[0095] The preset pixel change ratio threshold is set by staff based on historical data analysis results. This historical analysis process includes the system collecting multiple pixel change ratios and calculating their average value as a reference to obtain the preset pixel change ratio threshold. Similarly, preset abnormal statistical feature indicator judgment thresholds, preset statistical feature indicator thresholds, preset risk thresholds, and preset similarity thresholds are also set by staff based on the system's historical operating data and specific application scenario requirements. These preset thresholds can be adjusted by staff during system operation according to actual conditions.

[0096] In this embodiment, by linearizing and mapping multidimensional image data and introducing a diffusion jump scale dynamically controlled by a Logistic mapping chaotic system, each pixel corresponds to a different diffusion span during the diffusion process. This breaks through the traditional local diffusion mode based on adjacent pixels or fixed step size, establishing diffusion correlations between distant pixels within the global scope of the multidimensional image data. The diffusion jump scale is jointly determined by the chaotic state value and the maximum diffusion jump range, causing the diffusion path to dynamically change with the iterative state of the chaotic system. The diffusion relationship between different pixels is no longer fixed, thereby improving the nonlinear characteristics and unpredictability of the diffusion process and effectively enhancing the encryption image's resistance to differential attacks. The system enhances resistance to attacks and statistical analysis attacks. By establishing a global diffusion mapping relationship based on the diffusion jump scale, the pixel value of the current pixel is not only related to itself during the diffusion process, but also coupled with the diffusion results of other pixels selected by the chaotic system. This allows minute numerical changes in any pixel to propagate to pixel positions at different distances during the diffusion process, achieving multi-scale global diffusion of multi-dimensional image data. Since the diffusion process takes place globally within the multi-dimensional image data, it can effectively break the original correlations of multi-dimensional images in spatial, channel, and hierarchical dimensions, structurally reducing the identifiability of statistical features of image data and improving the overall security and robustness of encrypted images.

[0097] By dynamically controlling the pixel jump scale using chaotic sequences, the diffusion operation covers the entire global range of the multidimensional image, achieving cross-distance, nonlinear, and multi-scale pixel coupling. Even if there are minor changes in the local image content, they can be rapidly propagated globally, enhancing resistance to differential attacks and the security of statistical analysis. In the encryption stage, a key-driven chaotic perturbation sequence is introduced as a quantization offset factor, making the information-loss transformation not only controlled by compression / quantization parameters but also key-dependent. The chaotic perturbation sequence and compression operation work together to enhance the unpredictability of the ciphertext, making it impossible to accurately deduce the original pixel distribution under unknown key conditions. In the decryption stage, the encrypted pixel values ​​are reversed using the same chaotic perturbation term, compression parameters, and compensation correction term as in the encryption stage. This compensates for information loss and rounding errors; under known key conditions, the recovered pixel values ​​are ideally identical to the diffused pixels, ensuring the reversibility of image decryption. The combination of chaotic-controlled global diffusion and information-loss pixel transformation makes the encrypted image both statistically unpredictable and accurately recoverable under legitimate key conditions. It effectively solves the problems of local diffusion, insufficient key sensitivity, and irreversible information loss in traditional multidimensional image encryption.

[0098] Example 2

[0099] Please see Figure 2As shown, parts not described in detail in this embodiment are described in Embodiment 1. A multidimensional image encryption system based on chaotic coding and global diffusion is provided, including:

[0100] The image acquisition and representation module is used to acquire the target image data to be encrypted and to perform a structured representation of the target image data, organizing the image data into multi-dimensional image data containing spatial dimension, channel dimension and hierarchical dimension;

[0101] The parameter initialization module is used to select a Logistic map chaotic system, initialize the initial state of the chaotic system based on a preset key parameter, and iteratively generate a sequence of chaotic state values ​​that match the multidimensional image data.

[0102] The global diffusion jump module is used to perform global nonlinear rearrangement and diffusion processing on the position of each pixel during the encryption process of multidimensional image data. It dynamically determines the jump scale of pixel diffusion based on the chaotic state value sequence to obtain the diffused multidimensional image data.

[0103] The information loss encryption module is used to process the information loss of the diffused multidimensional image data. By performing controlled numerical compression or quantization on the pixel values, information loss is formed on the surface during the encryption stage to generate encrypted image data.

[0104] The encrypted output storage module is used to output and store encrypted image data, and in the decryption stage, it calls the corresponding chaotic coding parameters and compensation rules to reverse process the encrypted image data to recover the multidimensional image data.

[0105] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0106] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A multidimensional image encryption method based on chaotic coding and global diffusion, characterized in that, include: S1. Obtain the target image data to be encrypted, and perform a structured representation of the target image data, organizing the image data into multi-dimensional image data containing spatial dimension, channel dimension and hierarchical dimension; S2. Select a Logistic mapping chaotic system and initialize the initial state of the chaotic system based on preset key parameters, and iteratively generate a sequence of chaotic state values ​​that match the multidimensional image data. S3. In the process of encrypting multidimensional image data, the position of each pixel is subjected to global nonlinear rearrangement and diffusion processing. The jump scale of pixel diffusion is dynamically determined according to the chaotic state value sequence to obtain the diffused multidimensional image data. The method for obtaining the diffused multidimensional image data includes: Multidimensional image data is linearized and mapped according to a preset dimensional expansion rule, and a unique linear index number is assigned to each pixel in the multidimensional image data to uniformly describe the positional relationship of each pixel in the global diffusion process; based on the chaotic state value generated by the Logistic mapping chaotic system in the corresponding iteration step, the corresponding diffusion jump scale is dynamically determined for each pixel. The diffusion jump scale is determined by the chaotic state value and the preset maximum diffusion jump range, so that different pixels have different diffusion spans during the diffusion process, and the diffusion span covers the global range of the multidimensional image data; after determining the diffusion jump scale of the pixel, a global diffusion mapping relationship between pixels is established based on the diffusion jump scale. By using the diffusion jump scale and global diffusion mapping relationship dynamically controlled by the Logistic map chaotic system, the numerical change of any pixel in the multidimensional image data can be propagated to pixel positions at different distances during the diffusion process, thus obtaining the diffused multidimensional image data. S4. Information loss processing is performed on the diffused multidimensional image data. By performing controlled numerical compression or quantization operations on pixel values, information loss is formed on the surface during the encryption stage to generate encrypted image data. S5. Output and store the encrypted image data, and in the decryption stage, call the corresponding chaotic coding parameters and compensation rules to reverse process the encrypted image data in order to recover the multidimensional image data. The method for restoring multidimensional image data includes: During the decryption phase, the same chaotic system parameters as those in the encryption phase are invoked to perform synchronous initialization and iterative calculations on the chaotic system, and a new chaotic perturbation sequence corresponding one-to-one with the pixel index in the encryption phase is generated to compensate for the quantization or rounding errors introduced by the information loss-type pixel transformation in the encryption phase. For each pixel value in the encrypted image data, a pixel recovery operation based on chaos compensation is performed using a chaotic perturbation sequence and compression or quantization parameters consistent with the encryption stage. This reverse recovery process is then applied to the encrypted pixel values ​​to recover the multidimensional image data.

2. The multidimensional image encryption method based on chaotic coding and global diffusion according to claim 1, characterized in that, The method for obtaining the target image data to be encrypted includes: The target image data is acquired from image acquisition devices or image files, including digital cameras, scanners and medical imaging devices, and image files including BMP, PNG, JPEG and TIFF digital image formats.

3. The multidimensional image encryption method based on chaotic coding and global diffusion according to claim 2, characterized in that, The method for acquiring the multidimensional image data includes: The target image data is parsed through the image reading interface and converted into a digital pixel matrix that can be processed by a computer; the grayscale value of each pixel in the target image is digitally represented and organized into a two-dimensional or three-dimensional array according to the row and column order of the target image. The digital pixel matrix is ​​represented in a structured way, and the row and column coordinates of the target image are organized as two-dimensional spatial dimensions. A unique spatial index is assigned to each pixel to describe the positional relationship of the pixels in two-dimensional space. The color channels of the target image are organized as channel dimensions, and a unique index is assigned to each pixel in the channel dimension. Then, the different frames or levels of the target image are organized as level dimensions, and a unique index is assigned to each pixel in the level dimension. Finally, a unique index is generated for each pixel in the target image data, and an index mapping relationship is established for spatial dimension, channel dimension and hierarchical dimension to obtain multidimensional image data containing spatial dimension, channel dimension and hierarchical dimension.

4. The multidimensional image encryption method based on chaotic coding and global diffusion according to claim 3, characterized in that, The method for initializing the initial state of the chaotic system includes: Logistic mapping is chosen as the basic model of chaotic system. The initial state of chaotic system is initialized based on preset key parameters, which include initial state values ​​and system control parameters. Initial state seeds and control parameter seeds are extracted from the preset key parameters. The initial state seed is mapped to the effective initial value range of the Logistic map chaotic system to obtain the initial state value of the chaotic system; the control parameter seed is mapped to the effective range of the control parameters of the chaotic system to obtain the system control parameters; the initial state value and the system control parameters are mapped to the effective numerical range of the chaotic system by the key.

5. A multidimensional image encryption method based on chaotic coding and global diffusion according to claim 4, characterized in that, The method for obtaining the chaotic state value sequence includes: The length of the chaotic state value sequence is determined based on the total number of pixels or the matrix size of the multidimensional image data. By iterating the Logistic mapping formula, a chaotic state value sequence with the same number of pixels as the multidimensional image data is generated iteratively, with each chaotic state value corresponding to a pixel in the multidimensional image data.

6. A multidimensional image encryption method based on chaotic coding and global diffusion according to claim 5, characterized in that, The method for generating encrypted image data includes: For each pixel value in the diffused multidimensional image arranged in a predetermined scanning order, a chaotic perturbation sequence corresponding one-to-one with the pixel index is generated based on a key-driven chaotic system, and the chaotic perturbation sequence is introduced into the pixel transformation process as a quantization offset factor. During the pixel transformation process, information loss transformation is performed on the pixel values ​​in the diffused multidimensional image data. A proportional modulation related to the compression or quantization intensity is applied to the current pixel value, and a controlled numerical compression or quantization operation is performed after superimposing chaotic perturbations. This causes the pixel values ​​to produce limited information loss during the encryption stage, thereby obtaining encrypted image data.

7. A multidimensional image encryption method based on chaotic coding and global diffusion according to claim 6, characterized in that, The method for outputting and storing encrypted image data includes: The encrypted image data is organized and encapsulated according to a preset data format. The encapsulation can be implemented using existing image file formats while maintaining the pixel arrangement order. Encrypted image data can be stored on local storage devices, servers, or remote cloud storage systems, supporting the saving, cross-platform access, or retrieval of encrypted image data. During output and storage, encrypted image data can be transmitted through standardized interfaces, while integrity verification ensures that the encrypted image data has not been tampered with.

8. A multidimensional image encryption system based on chaotic coding and global diffusion, used to implement the multidimensional image encryption method based on chaotic coding and global diffusion as described in any one of claims 1 to 7, characterized in that, include: The image acquisition and representation module is used to acquire the target image data to be encrypted and to perform a structured representation of the target image data, organizing the image data into multi-dimensional image data containing spatial dimension, channel dimension and hierarchical dimension; The parameter initialization module is used to select a Logistic map chaotic system, initialize the initial state of the chaotic system based on a preset key parameter, and iteratively generate a sequence of chaotic state values ​​that match the multidimensional image data. The global diffusion jump module is used to perform global nonlinear rearrangement and diffusion processing on the position of each pixel during the encryption process of multidimensional image data. It dynamically determines the jump scale of pixel diffusion based on the chaotic state value sequence to obtain the diffused multidimensional image data. The information loss encryption module is used to process the information loss of the diffused multidimensional image data. By performing controlled numerical compression or quantization on the pixel values, information loss is formed on the surface during the encryption stage to generate encrypted image data. The encrypted output storage module is used to output and store encrypted image data, and in the decryption stage, it calls the corresponding chaotic coding parameters and compensation rules to reverse process the encrypted image data to recover the multidimensional image data.

Images