A Color Image Encryption Method, System and Device Based on Chaotic Neural Network

Abstract

The present invention discloses a color image encryption method, system and device based on a chaotic neural network, including extracting the R component, G component and B component of a color image, respectively generating an R component scrambling transformation key, a G component scrambling transformation key and a B component scrambling transformation key based on a three-dimensional chaotic neural network, performing row and column random scrambling on the color image according to the scrambling transformation key to obtain a scrambled image, splitting the scrambled image into multiple parts by columns, respectively generating diffusion keys based on the three-dimensional chaotic neural network, diffusing all parts of the split image according to the diffusion keys to obtain all parts of the diffused image, scrambling and splicing all parts of the diffused image by using an array splicing function to obtain a spliced image, synthesizing the components of the spliced image to obtain an encrypted image of the color image, strengthening randomness, enhancing the anti-attack ability and improving security.

Description

Technical Field

[0001] The present invention relates to the technical field of image encryption, and in particular, to a color image encryption method, system and device based on a chaotic neural network. Background Art

[0002] With the rapid development of digital technology and network technology, more and more multimedia data has been generated and transmitted through the network and stored on platforms such as cloud servers. Among them, digital images contain a large amount of information. For example, through a picture of a military oil depot, not only its size and quantity can be obtained, but also its approximate location can be obtained; a face photo can not only expose his or her appearance, but also give an approximate age and physical condition. Therefore, in medical image systems, military image systems and video conferences, the security of protecting image data has attracted wide attention.

[0003] Currently, there are mainly two problems in the color image encryption method based on a chaotic neural network: one is that the method has a low correlation with the plaintext and poor anti-attack ability; the other is that some methods do not consider the correlation between the R, G, and B components, and the proposed methods are easily cracked by statistical analysis. Summary of the Invention

[0004] The present invention aims to at least solve the technical problems existing in the prior art. For this purpose, the present invention provides a color image encryption method, system and device based on a chaotic neural network, which can improve randomness and enhance anti-attack ability.

[0005] In the first aspect of the present invention, a color image encryption method based on a chaotic neural network is provided, including the following steps:

[0006] Extract the R component, G component and B component of the color image;

[0007] Based on a three-dimensional chaotic neural network, generate an R-component scrambling transformation key, a G-component scrambling transformation key and a B-component scrambling transformation key respectively. According to the R-component scrambling transformation key, perform row and column random scrambling on the R component of the color image to obtain the scrambled R component. According to the G-component scrambling transformation key, perform row and column random scrambling on the G component of the color image to obtain the scrambled G component. According to the B-component scrambling transformation key, perform row and column random scrambling on the B component of the color image to obtain the scrambled B component;

[0008] Respectively divide the scrambled R component, the scrambled G component and the scrambled B component into multiple parts by columns;

[0009] Based on the three-dimensional chaotic neural network, an R-component diffusion key, a G-component diffusion key, and a B-component diffusion key are respectively generated. All parts of the segmented R-component are diffused according to the R-component diffusion key to obtain all parts of the diffused R-component. All parts of the segmented G-component are diffused according to the G-component diffusion key to obtain all parts of the diffused G-component. All parts of the segmented B-component are diffused according to the B-component diffusion key to obtain all parts of the diffused B-component;

[0010] All parts of the diffused R-component are scrambled and spliced using an array splicing function to obtain a spliced R-component. All parts of the diffused G-component are scrambled and spliced using an array splicing function to obtain a spliced G-component. All parts of the diffused B-component are scrambled and spliced using an array splicing function to obtain a spliced B-component;

[0011] The spliced R-component, the spliced G-component, and the spliced B-component are synthesized to obtain the encrypted image of the color image.

[0012] According to the embodiments of the present invention, there are at least the following technical effects:

[0013] This method extracts the R component, G component, and B component of a color image, generates the scrambling transformation key for the R component, the scrambling transformation key for the G component, and the scrambling transformation key for the B component respectively based on a three-dimensional chaotic neural network. It performs row and column random scrambling on the R component of the color image according to the scrambling transformation key for the R component to obtain the scrambled R component, performs row and column random scrambling on the G component of the color image according to the scrambling transformation key for the G component to obtain the scrambled G component, and performs row and column random scrambling on the B component of the color image according to the scrambling transformation key for the B component to obtain the scrambled B component. The scrambled R component, the scrambled G component, and the scrambled B component are respectively split into multiple parts by columns, and the diffusion key for the R component, the diffusion key for the G component, and the diffusion key for the B component are generated respectively based on a three-dimensional chaotic neural network. All parts of the split R component are diffused according to the diffusion key for the R component to obtain all parts of the diffused R component, all parts of the split G component are diffused according to the diffusion key for the G component to obtain all parts of the diffused G component, and all parts of the split B component are diffused according to the diffusion key for the B component to obtain all parts of the diffused B component. By using the key stream generated by a three-dimensional chaotic neural network to perform scrambling and diffusion operations on the three channels of the color image, the randomness is enhanced and the security is improved. All parts of the diffused R component are scrambled and spliced using an array splicing function to obtain the spliced R component, all parts of the diffused G component are scrambled and spliced using an array splicing function to obtain the spliced G component, and all parts of the diffused B component are scrambled and spliced using an array splicing function to obtain the spliced B component. The spliced R component, the spliced G component, and the spliced B component are synthesized to obtain the encrypted image of the color image. The anti-attack ability is enhanced by scrambling again using the array splicing function, making the encryption more secure.

[0014] According to some embodiments of the present invention, the extraction of the R component, G component, and B component of the color image includes:

[0015] Using the array separation of matlab to extract the R component, G component, and B component of the color image, where the calculation formulas for the R component, G component, and B component of the color image are;

[0016]

[0017] where P is the color image, PR is the R component of the color image, PG is the G component of the color image, and PB is the B component of the color image.

[0018] According to some embodiments of the present invention, the generation of the scrambling transformation key for the R component based on a three-dimensional chaotic neural network, and the row and column random scrambling of the R component of the color image according to the scrambling transformation key for the R component to obtain the scrambled R component, includes:

[0019] Obtain the M×N two-dimensional matrix of the R component of the color image, where M is the number of rows of the pixel points of the R component of the color image, and N is the number of columns of the pixel points of the R component of the color image;

[0020] Obtain two groups of chaotic sequences y 1 (z), y 2 (z) through a three-dimensional chaotic neural network, and generate a scrambling transformation key according to the chaotic sequences, where the scrambling transformation key is a random non-repeating sequence, and the calculation formula for generating the scrambling transformation key according to the chaotic sequences is:

[0021]

[0022] where RandM(i) is the scrambling transformation key for the i-th row of the R component, RandN(j) is the scrambling transformation key for the j-th column of the R component, and floor() is the floor function;

[0023] Construct a key pair according to the scrambling transformation key and an increasing sequence, where the calculation formula for constructing the key pair according to the scrambling transformation key and the increasing sequence is:

[0024]

[0025] where Mchange is a 2×M two-dimensional array key pair, and Nchange is a 2×N two-dimensional array key pair;

[0026] Perform row and column random scrambling on the R component of the color image according to the key pair to obtain the scrambled R component, where the calculation formula for performing row and column random scrambling on the R component of the color image according to the key pair is:

[0027]

[0028] According to some embodiments of the present invention, the splitting the scrambled R component, the scrambled G component, and the scrambled B component into multiple parts by column respectively includes:

[0029] Split the scrambled R component into three parts by column to obtain the first part of the R component, the second part of the R component, and the third part of the R component, where the first part of the R component is the left one-third of the R component, the second part of the R component is the middle one-third of the R component, and the third part of the R component is the right one-third of the R component;

[0030] The scrambled G component is split into three parts column by column to obtain the first part of the G component, the second part of the G component, and the third part of the G component. Among them, the first part of the G component is the left one-third of the G component, the second part of the G component is the middle one-third of the G component, and the third part of the G component is the right one-third of the G component;

[0031] The scrambled B component is split into three parts column by column to obtain the first part of the B component, the second part of the B component, and the third part of the B component. Among them, the first part of the B component is the left one-third of the B component, the second part of the B component is the middle one-third of the B component, and the third part of the B component is the right one-third of the B component.

[0032] According to some embodiments of the present invention, an R-component diffusion key is generated based on a three-dimensional chaotic neural network, and all parts of the segmented R component are diffused by using the R-component diffusion key to obtain all parts of the diffused R component, including:

[0033] Obtain two-dimensional arrays of the first part of the R component, the second part of the R component, and the third part of the R component. Among them, the calculation formula for obtaining the two-dimensional arrays of the first part of the R component, the second part of the R component, and the third part of the R component is:

[0034]

[0035] Among them, LP is the two-dimensional array of the first part of the R component, CP is the two-dimensional array of the second part of the R component, and RP is the two-dimensional array of the third part of the R component;

[0036] Generate three different chaotic sequences y 1 , y 2 , y 3 through the three-dimensional chaotic neural network, and calculate the R-component diffusion key through the chaotic sequences. Among them, the calculation formula for calculating the R-component diffusion key through the chaotic sequences is:

[0037]

[0038] Among them, k(n 1 ) is the diffusion key of the first part of the R component, k(n 2 ) is the diffusion key of the second part of the R component, and k(n 3 ) is the diffusion key of the third part of the R component;

[0039] XOR encrypt the first part of the R component using the two-dimensional array of the first part of the R component and the diffusion key of the first part of the R component to obtain the first part of the diffused R component. XOR encrypt the second part of the R component using the two-dimensional array of the second part of the R component and the diffusion key of the second part of the R component to obtain the second part of the diffused R component. XOR encrypt the third part of the R component using the two-dimensional array of the third part of the R component and the diffusion key of the third part of the R component to obtain the third part of the diffused R component. Among them, the calculation formula for XOR encrypting the first part of the R component using the two-dimensional array of the first part of the R component and the diffusion key of the first part of the R component to obtain the first part of the diffused R component, XOR encrypting the second part of the R component using the two-dimensional array of the second part of the R component and the diffusion key of the second part of the R component to obtain the second part of the diffused R component, and XOR encrypting the third part of the R component using the two-dimensional array of the third part of the R component and the diffusion key of the third part of the R component to obtain the third part of the diffused R component is as follows:

[0040]

[0041] Among them, CLR is the first part of the diffused R component, CCR is the second part of the diffused R component, CRR is the third part of the diffused R component, and bixor(P,k) is the XOR function.

[0042] According to some embodiments of the present invention, the calculation formula for scrambling and splicing all parts of the diffused R component using the array splicing function is as follows:

[0043] CR = [CCR, CLR, CRR]

[0044] Among them, CR is the spliced R component.

[0045] According to some embodiments of the present invention, the calculation formula for synthesizing the spliced R component, the spliced G component, and the spliced B component is as follows:

[0046] C = cat(3, CR, CG, CB)

[0047] Among them, C is the encrypted image, CG is the spliced G component, CB is the spliced B component, and cat() is the splicing function.

[0048] According to some embodiments of the present invention, the specific form of the third-order neural network is as follows:

[0049]

[0050]

[0051] Among them, X(t), Y(t), and Z(t) respectively represent the state variables of three neurons, and the initial values of the neuron state variables are set as X(0), Y(0), and Z(0).

[0052] In a second aspect of the present invention, a color image encryption system based on a chaotic neural network is provided. The color image encryption system based on a chaotic neural network includes:

[0053] A component extraction module for extracting the R component, G component, and B component of a color image;

[0054] A component scrambling module for respectively generating an R-component scrambling transformation key, a G-component scrambling transformation key, and a B-component scrambling transformation key based on a three-dimensional chaotic neural network, randomly scrambling the rows and columns of the R component of the color image according to the R-component scrambling transformation key to obtain a scrambled R component, randomly scrambling the rows and columns of the G component of the color image according to the G-component scrambling transformation key to obtain a scrambled G component, and randomly scrambling the rows and columns of the B component of the color image according to the B-component scrambling transformation key to obtain a scrambled B component;

[0055] A component segmentation module for respectively dividing the scrambled R component, the scrambled G component, and the scrambled B component into multiple parts by columns;

[0056] A component diffusion module for respectively generating an R-component diffusion key, a G-component diffusion key, and a B-component diffusion key based on the three-dimensional chaotic neural network, diffusing all parts of the segmented R component according to the R-component diffusion key to obtain all parts of the diffused R component, diffusing all parts of the segmented G component according to the G-component diffusion key to obtain all parts of the diffused G component, and diffusing all parts of the segmented B component according to the B-component diffusion key to obtain all parts of the diffused B component;

[0057] A component splicing module for scrambling and splicing all parts of the diffused R component by using an array splicing function to obtain a spliced R component, scrambling and splicing all parts of the diffused G component by using an array splicing function to obtain a spliced G component, and scrambling and splicing all parts of the diffused B component by using an array splicing function to obtain a spliced B component;

[0058] A component synthesis module for synthesizing the spliced R component, the spliced G component, and the spliced B component to obtain an encrypted image of the color image.

[0059] This system extracts the R component, G component, and B component of a color image, and respectively generates an R-component scrambling transformation key, a G-component scrambling transformation key, and a B-component scrambling transformation key based on a three-dimensional chaotic neural network. According to the R-component scrambling transformation key, the R component of the color image is randomly scrambled row by row and column by column to obtain the scrambled R component. According to the G-component scrambling transformation key, the G component of the color image is randomly scrambled row by row and column by column to obtain the scrambled G component. According to the B-component scrambling transformation key, the B component of the color image is randomly scrambled row by row and column by column to obtain the scrambled B component. The scrambled R component, the scrambled G component, and the scrambled B component are respectively divided into multiple parts by columns, and an R-component diffusion key, a G-component diffusion key, and a B-component diffusion key are respectively generated based on a three-dimensional chaotic neural network. According to the R-component diffusion key, all parts of the divided R component are diffused to obtain all parts of the diffused R component. According to the G-component diffusion key, all parts of the divided G component are diffused to obtain all parts of the diffused G component. According to the B-component diffusion key, all parts of the divided B component are diffused to obtain all parts of the diffused B component. By using the key stream generated by a three-dimensional chaotic neural network to perform scrambling and diffusion operations on the three channels of the color image, the randomness is enhanced and the security is improved. All parts of the diffused R component are scrambled and spliced using an array splicing function to obtain the spliced R component. All parts of the diffused G component are scrambled and spliced using an array splicing function to obtain the spliced G component. All parts of the diffused B component are scrambled and spliced using an array splicing function to obtain the spliced B component. The spliced R component, the spliced G component, and the spliced B component are synthesized to obtain the encrypted image of the color image. The anti-attack ability is enhanced by scrambling again using the array splicing function, making the encryption more secure.

[0060] In a third aspect of the present invention, there is provided a color image encryption electronic device based on a chaotic neural network, including at least one control processor and a memory for communicatively connecting with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the above-mentioned color image encryption method based on a chaotic neural network.

[0061] In a fourth aspect of the present invention, there is provided a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the above-mentioned color image encryption method based on a chaotic neural network.

[0062] It should be noted that the beneficial effects between the second to fourth aspects of the present invention and the prior art are the same as those between the above-mentioned color image encryption system based on a chaotic neural network and the prior art, and will not be elaborated here.

[0063] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The above and / or additional aspects and advantages of the present invention will become apparent and be readily understood from the description of the embodiments in conjunction with the following drawings, where:

[0065] Figure 1 is a flowchart of a color image encryption method based on a chaotic neural network according to an embodiment of the present invention;

[0066] Figure 2 is an overall flowchart of a color image encryption method based on a chaotic neural network according to an embodiment of the present invention;

[0067] Figure 3 is an experimental comparison diagram of Lena image encryption and decryption of a color image encryption method based on a chaotic neural network according to an embodiment of the present invention;

[0068] Figure 4 is a test result diagram of the Lena image histogram of a color image encryption method based on a chaotic neural network according to an embodiment of the present invention;

[0069] Figure 5 is a test result diagram of the Lena image correlation of a color image encryption method based on a chaotic neural network according to an embodiment of the present invention;

[0070] Figure 6 is a test result diagram of the Lena image sensitivity of a color image encryption method based on a chaotic neural network according to an embodiment of the present invention;

[0071] Figure 7 is a flowchart of a color image encryption system based on a chaotic neural network according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0072] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described by referring to the drawings are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

[0073] In the description of the present invention, if the first, second, etc. are described, it is only for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features or implicitly indicating the sequence relationship of the indicated technical features.

[0074] In the description of the present invention, it should be understood that when it comes to the description of directions, such as up, down, etc., the indicated direction or positional relationship is based on the direction or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the indicated device or element must have a specific direction, be constructed and operated in a specific direction. Therefore, it cannot be understood as a limitation to the present invention.

[0075] In the description of the present invention, it should be noted that unless otherwise clearly defined, terms such as setting, installation, connection, etc. should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention in combination with the specific content of the technical solution.

[0076] Currently, there are mainly two problems with the color image encryption method based on chaotic neural networks: one is that the method has a low correlation with the plaintext and poor anti-attack ability; the other is that some methods do not consider the correlation between the R, G, and B components, and the proposed method is easily cracked by statistical analysis.

[0077] To solve the above technical defects, referring to Figure 1 and Figure 2 , the present invention also provides a color image encryption method based on chaotic neural networks, including:

[0078] Step S101: Extract the R component, G component, and B component of the color image.

[0079] Step S102: Based on a three-dimensional chaotic neural network, generate an R-component scrambling transformation key, a G-component scrambling transformation key, and a B-component scrambling transformation key respectively. Randomly scramble the rows and columns of the R component of the color image according to the R-component scrambling transformation key to obtain the scrambled R component. Randomly scramble the rows and columns of the G component of the color image according to the G-component scrambling transformation key to obtain the scrambled G component. Randomly scramble the rows and columns of the B component of the color image according to the B-component scrambling transformation key to obtain the scrambled B component.

[0080] Step S103: Divide the scrambled R component, the scrambled G component, and the scrambled B component into multiple parts by columns respectively.

[0081] Step S104: Generate the R-component diffusion key, G-component diffusion key, and B-component diffusion key respectively based on the three-dimensional chaotic neural network. Diffuse all parts of the segmented R-component according to the R-component diffusion key to obtain all parts of the diffused R-component. Diffuse all parts of the segmented G-component according to the G-component diffusion key to obtain all parts of the diffused G-component. Diffuse all parts of the segmented B-component according to the B-component diffusion key to obtain all parts of the diffused B-component.

[0082] Step S105: Scramble and splice all parts of the diffused R-component using an array splicing function to obtain the spliced R-component. Scramble and splice all parts of the diffused G-component using an array splicing function to obtain the spliced G-component. Scramble and splice all parts of the diffused B-component using an array splicing function to obtain the spliced B-component.

[0083] Step S106: Synthesize the spliced R-component, spliced G-component, and spliced B-component to obtain the encrypted image of the color image.

[0084] This method extracts the R component, G component, and B component of a color image, and respectively generates an R-component scrambling transformation key, a G-component scrambling transformation key, and a B-component scrambling transformation key based on a three-dimensional chaotic neural network. According to the R-component scrambling transformation key, the R component of the color image is randomly scrambled row by row and column by column to obtain the scrambled R component. According to the G-component scrambling transformation key, the G component of the color image is randomly scrambled row by row and column by column to obtain the scrambled G component. According to the B-component scrambling transformation key, the B component of the color image is randomly scrambled row by row and column by column to obtain the scrambled B component. The scrambled R component, the scrambled G component, and the scrambled B component are respectively split into multiple parts column by column, and an R-component diffusion key, a G-component diffusion key, and a B-component diffusion key are respectively generated based on the three-dimensional chaotic neural network. According to the R-component diffusion key, all parts of the split R component are diffused to obtain all parts of the diffused R component. According to the G-component diffusion key, all parts of the split G component are diffused to obtain all parts of the diffused G component. According to the B-component diffusion key, all parts of the split B component are diffused to obtain all parts of the diffused B component. By using the key stream generated by the three-dimensional chaotic neural network to perform scrambling and diffusion operations on the three channels of the color image, the randomness is enhanced and the security is improved. All parts of the diffused R component are scrambled and spliced using an array splicing function to obtain the spliced R component. All parts of the diffused G component are scrambled and spliced using an array splicing function to obtain the spliced G component. All parts of the diffused B component are scrambled and spliced using an array splicing function to obtain the spliced B component. The spliced R component, the spliced G component, and the spliced B component are synthesized to obtain the encrypted image of the color image. The anti-attack ability is enhanced by scrambling again using the array splicing function, making the encryption more secure.

[0085] In some embodiments, step S101 may include but is not limited to including step S201:

[0086] Step S201: Use the array separation of matlab to extract the R component, G component, and B component of the color image. Among them, the calculation formulas for the R component, G component, and B component of the color image are;

[0087]

[0088] Among them, P is the color image, PR is the R component of the color image, PG is the G component of the color image, and PB is the B component of the color image.

[0089] In some embodiments, step S102 may include but is not limited to including steps S301 to S304:

[0090] Step S301: Obtain the M×N two-dimensional matrix of the R component of the color image, where M is the number of rows of the pixel points of the R component of the color image, and N is the number of columns of the pixel points of the R component of the color image.

[0091] Step S302: Obtain two groups of chaotic sequences y 1 (z), y 2 (z) through a three-dimensional chaotic neural network according to the two-dimensional matrix, and generate a scrambling transformation key according to the chaotic sequences. The scrambling transformation key is a random non-repeating sequence. The calculation formula for generating the scrambling transformation key according to the chaotic sequences is:

[0092]

[0093] where RandM(i) is the scrambling transformation key for the i-th row of the R component, RandN(j) is the scrambling transformation key for the j-th column of the R component, and floor() is the floor function.

[0094] Step S303: Construct a key pair according to the scrambling transformation key and the increasing sequence. The calculation formula for constructing the key pair according to the scrambling transformation key and the increasing sequence is:

[0095]

[0096] where Mchange is a 2×M two-dimensional array key pair, and Nchange is a 2×N two-dimensional array key pair.

[0097] Step S304: Perform row and column random scrambling on the R component of the color image according to the key pair to obtain the scrambled R component. The calculation formula for performing row and column random scrambling on the R component of the color image according to the key pair is:

[0098]

[0099] In some embodiments, step S102 may include but is not limited to steps S401 to S404:

[0100] Step S401: Obtain the M×N two-dimensional matrix of the G component of the color image, where M is the number of rows of the pixel points of the G component of the color image, and N is the number of columns of the pixel points of the G component of the color image.

[0101] Step S402: Obtain two groups of chaotic sequences y 3 (z), y 4 (z) through a three-dimensional chaotic neural network according to the two-dimensional matrix, and generate a scrambling transformation key according to the chaotic sequences. The scrambling transformation key is a random non-repeating sequence. The calculation formula for generating the scrambling transformation key according to the chaotic sequences is:

[0102]

[0103] Among them, RandM(i) is the scrambling transformation key for the i-th row of the G component, RandN(j) is the scrambling transformation key for the j-th column of the G component, and floor() is the floor function.

[0104] Step S403: Construct key pairs according to the scrambling transformation key and the increasing sequence. The calculation formula for constructing key pairs according to the scrambling transformation key and the increasing sequence is as follows:

[0105]

[0106] Among them, Mchange is a 2×M two-dimensional array key pair, and Nchange is a 2×N two-dimensional array key pair.

[0107] Step S404: Perform row and column random scrambling on the G component of the color image according to the key pairs to obtain the scrambled G component. The calculation formula for performing row and column random scrambling on the G component of the color image according to the key pairs is as follows:

[0108]

[0109] Similarly, the scrambled B component can be obtained.

[0110] In some embodiments, step S103 may include but is not limited to steps S501 to S503:

[0111] Step S501: Divide the scrambled R component into three parts column by column to obtain the first part of the R component, the second part of the R component, and the third part of the R component. Among them, the first part of the R component is the left one-third of the R component, the second part of the R component is the middle one-third of the R component, and the third part of the R component is the right one-third of the R component.

[0112] Step S502: Divide the scrambled G component into three parts column by column to obtain the first part of the G component, the second part of the G component, and the third part of the G component. Among them, the first part of the G component is the left one-third of the G component, the second part of the G component is the middle one-third of the G component, and the third part of the G component is the right one-third of the G component.

[0113] Step S503: Divide the scrambled B component into three parts column by column to obtain the first part of the B component, the second part of the B component, and the third part of the B component. Among them, the first part of the B component is the left one-third of the B component, the second part of the B component is the middle one-third of the B component, and the third part of the B component is the right one-third of the B component.

[0114] In some embodiments, step S104 may include but is not limited to steps S601 to S603:

[0115] Step S601, obtain two-dimensional arrays of the first part, the second part, and the third part of the R component, where the calculation formula for obtaining the two-dimensional arrays of the first part, the second part, and the third part of the R component is:

[0116]

[0117] where LP is the two-dimensional array of the first part of the R component, CP is the two-dimensional array of the second part of the R component, and RP is the two-dimensional array of the third part of the R component.

[0118] Step S602, generate three different chaotic sequences y 1 , y 2 , y 3 , and calculate the R-component diffusion key through the chaotic sequences. The calculation formula for calculating the R-component diffusion key through the chaotic sequences is:

[0119]

[0120] where k(n 1 ) is the diffusion key of the first part of the R component, k(n 2 ) is the diffusion key of the second part of the R component, and k(n 3 ) is the diffusion key of the third part of the R component.

[0121] Step S603, use the two-dimensional array of the first part of the R component and the diffusion key of the first part of the R component to perform exclusive-or encryption on the first part of the R component to obtain the diffused first part of the R component, use the two-dimensional array of the second part of the R component and the diffusion key of the second part of the R component to perform exclusive-or encryption on the second part of the R component to obtain the diffused second part of the R component, and use the two-dimensional array of the third part of the R component and the diffusion key of the third part of the R component to perform exclusive-or encryption on the third part of the R component to obtain the diffused third part of the R component. The calculation formula for using the two-dimensional array of the first part of the R component and the diffusion key of the first part of the R component to perform exclusive-or encryption on the first part of the R component to obtain the diffused first part of the R component, using the two-dimensional array of the second part of the R component and the diffusion key of the second part of the R component to perform exclusive-or encryption on the second part of the R component to obtain the diffused second part of the R component, and using the two-dimensional array of the third part of the R component and the diffusion key of the third part of the R component to perform exclusive-or encryption on the third part of the R component to obtain the diffused third part of the R component is:

[0122]

[0123] Wherein, CLR is the first part of the diffused R component, CCR is the second part of the diffused R component, CRR is the third part of the diffused R component, and bixor(P,k) is an exclusive OR function.

[0124] In some embodiments, step S104 may include but is not limited to steps S701 to S703:

[0125] Step S701, obtain two-dimensional arrays of the first part of the G component, the second part of the G component, and the third part of the G component. The calculation formula for obtaining the two-dimensional arrays of the first part of the G component, the second part of the G component, and the third part of the G component is:

[0126]

[0127] Wherein, LP is the two-dimensional array of the first part of the G component, CP is the two-dimensional array of the second part of the G component, and RP is the two-dimensional array of the third part of the G component.

[0128] Step S702, generate three different chaotic sequences y 4 , y 5 , y 6 , and calculate the diffusion key of the G component through the chaotic sequences. The calculation formula for calculating the diffusion key of the G component through the chaotic sequences is:

[0129]

[0130] Wherein, k(n 4 ) is the diffusion key of the first part of the G component, k(n 5 ) is the diffusion key of the second part of the G component, and k(n 6 ) is the diffusion key of the third part of the G component.

[0131] Step S703: Use the two-dimensional array of the first part of the G component and the diffusion key of the first part of the G component to perform exclusive OR encryption on the first part of the G component to obtain the diffused first part of the G component. Use the two-dimensional array of the second part of the G component and the diffusion key of the second part of the G component to perform exclusive OR encryption on the second part of the G component to obtain the diffused second part of the G component. Use the two-dimensional array of the third part of the G component and the diffusion key of the third part of the G component to perform exclusive OR encryption on the third part of the G component to obtain the diffused third part of the G component. Among them, the formula for using the two-dimensional array of the first part of the G component and the diffusion key of the first part of the G component to perform exclusive OR encryption on the first part of the G component to obtain the diffused first part of the G component, using the two-dimensional array of the second part of the G component and the diffusion key of the second part of the G component to perform exclusive OR encryption on the second part of the G component to obtain the diffused second part of the G component, and using the two-dimensional array of the third part of the G component and the diffusion key of the third part of the G component to perform exclusive OR encryption on the third part of the G component to obtain the diffused third part of the G component is as follows:

[0132]

[0133] Among them, CLG is the diffused first part of the G component, CCG is the diffused second part of the G component, CRG is the diffused third part of the G component, and bixor(P,k) is the exclusive OR function.

[0134] Similarly, the diffused B component can be obtained.

[0135] In some embodiments, step S105 may include but is not limited to step S801:

[0136] Step S801: The formula for scrambling and splicing all parts of the diffused R component using the array splicing function is:

[0137] CR = [CCR, CLR, CRR]

[0138] Among them, CR is the spliced R component.

[0139] In some embodiments, step S105 may include but is not limited to step S901:

[0140] Step S901: The formula for scrambling and splicing all parts of the diffused G component using the array splicing function is:

[0141] CG = [CCG, CLG, CRG]

[0142] Among them, CG is the spliced G component.

[0143] Similarly, the calculation formula for scrambling and splicing all parts of the diffused B component using an array splicing function can be obtained.

[0144] In some embodiments, the calculation formula for synthesizing the spliced R component, the spliced G component, and the spliced B component is:

[0145] C = cat(3, CR, CG, CB)

[0146] where C is the encrypted image, CG is the spliced G component, CB is the spliced B component, and cat() is the splicing function.

[0147] In some embodiments, the specific form of the third-order neural network is:

[0148]

[0149]

[0150] where X(t), Y(t), and Z(t) respectively represent the state variables of 3 neurons, and the initial values of the neuron state variables are set as X(0), Y(0), and Z(0).

[0151] For the convenience of those skilled in the art to understand, a set of experimental data is provided below:

[0152] Refer to Figure 3 , it can be seen that the contour of the ciphertext image cannot be seen at all, and the pixel points are evenly distributed, making it almost impossible to obtain image information from it; and the decrypted image and the plaintext image are exactly the same, and the encryption and decryption effects are good.

[0153] Refer to Figure 4 , Figure 4 is the result graph of the histogram analysis of the Lena image. A high-security image encryption system should make the histogram of the encrypted ciphertext image as flat as possible. Looking at the histograms of R, G, and B of the plaintext image, it presents a mountain-shaped graph, and the distribution of the number of each pixel value is extremely uneven, showing a fault phenomenon at both ends. Thus, most of the image information can be obtained from this plaintext histogram. However, the histograms of R, G, and B of this ciphertext image are extremely flat, without faults and without any bulge. It shows that the encrypted image information is more secret and has a certain degree of security. Therefore, the encryption scheme of the present invention has the ability to resist statistical analysis.

[0154] The correlation coefficient results of the present invention in different directions are shown in Table 1. It can be seen that the correlation coefficient of the Lena plaintext image is almost close to 1, indicating a very strong correlation of the original image; while the ciphertext image is almost close to 0, indicating that there is almost no correlation between adjacent pixel points of the ciphertext image. The plaintext image encrypted by the image encryption system proposed in this paper has basically no correlation in the four directions of horizontal, vertical, positive diagonal, and anti-diagonal, and it is a color image encryption system based on chaotic neural network with relatively good performance.

[0155] Table 1

[0156]

[0157] Referring to Figure 5 , it can be seen that the Lena plaintext image shows a positive correlation distribution in the four directions of horizontal, vertical, positive diagonal, and anti-diagonal; however, the Lena ciphertext image shows an irregular distribution in the four directions, and the pixel values of adjacent pixel points of each pixel point can be any value. In the plaintext image, the pixel values of adjacent pixel points of each pixel point are close, and almost exist or are close to the straight line of u i = v i and the correlation is very high.

[0158] The information entropy represents the degree of chaos of the state. The larger the entropy value, the higher the degree of chaos, the higher the uncertainty of the image information, indicating that the image is more irregular. Its calculation formula is as follows.

[0159]

[0160] where P(i) represents the probability of occurrence when the pixel value is i. The ideal entropy of the R, G, and B components of a color image is 8. The present invention selects the ciphertext image obtained by the Lena plaintext image through the chaotic neural network image encryption system and analyzes the entropy of its R, G, and B components. It can be concluded from Table 2 that the information entropy of the three channels of the ciphertext image obtained by the encryption algorithm proposed in this paper is very close to the theoretical value of 8, and it can be concluded that the encryption algorithm has good security.

[0161] Table 2

[0162]

[0163] PSNR, that is, the peak signal-to-noise ratio, is an index to measure the distortion degree between the plaintext image and the ciphertext image. The lower the PSNR value, the greater the difference between the plaintext image and the ciphertext image, and the better the encryption algorithm used.

[0164] The full name of MSE is Mean squared error, that is, the mean square error, which is used to calculate the cumulative square error between the plaintext image and the ciphertext image. The larger the MSE, the better the encryption effect. The definitions of PSNR and MSE are as follows:

[0165]

[0166] From the PSNR and MSE test results in Table 3, the PSNR of the present invention is 8.6315 and the MSE is 8911.0, indicating that the encryption algorithm proposed by the present invention has very good performance.

[0167] Table 3

[0168]

[0169] Referring to Figure 6 , sensitivity analysis refers to decrypting the ciphertext image with an initial value slightly different from the original key initial value to see if the plaintext image can be restored. The Lena plaintext image is encrypted using the key initial value proposed in this paper, and the ciphertext image is decrypted using the following three groups of key initial values Y1, Y2, and Y3 with slight differences. The effect is as Figure 6 shown, and the plaintext image cannot be correctly restored. Therefore, the encryption system given by the present invention meets the requirements of key sensitivity.

[0170]

[0171] An ideal image encryption method should have its key space greater than 2 100 to ensure the security of the encryption algorithm. Assuming that the computing precision of the computer is 10 -15 and the precision of the compression ratio CR is 10 -2 . In the present invention, the calculation result of the total key space of the chaotic neural network encryption system is:

[0172] 10 2 ×10 15 ×10 14 ×10 14 ×10 14 ×10 14 ×10 14 ×10 14 ×10 14 >> 2 100

[0173] It shows that this encryption method has a large enough key space to resist brute-force attacks.

[0174] In addition, referring to Figure 7 , an embodiment of the present invention provides a color image encryption system based on a chaotic neural network, including a component extraction module 1100, a component scrambling module 1200, a component segmentation module 1300, a component diffusion module 1400, a component splicing module 1500, and a component synthesis module 1600, where:

[0175] The component extraction module 1100 is used to extract the R component, G component, and B component of a color image.

[0176] The component scrambling module 1200 is used to respectively generate an R-component scrambling transformation key, a G-component scrambling transformation key, and a B-component scrambling transformation key based on a three-dimensional chaotic neural network. According to the R-component scrambling transformation key, the R component of the color image is randomly scrambled row by row and column by column to obtain the scrambled R component. According to the G-component scrambling transformation key, the G component of the color image is randomly scrambled row by row and column by column to obtain the scrambled G component. According to the B-component scrambling transformation key, the B component of the color image is randomly scrambled row by row and column by column to obtain the scrambled B component.

[0177] The component splitting module 1300 is used to split the scrambled R component, the scrambled G component, and the scrambled B component into multiple parts column by column respectively.

[0178] The component diffusion module 1400 is used to respectively generate an R-component diffusion key, a G-component diffusion key, and a B-component diffusion key based on a three-dimensional chaotic neural network. According to the R-component diffusion key, all parts of the split R component are diffused to obtain all parts of the diffused R component. According to the G-component diffusion key, all parts of the split G component are diffused to obtain all parts of the diffused G component. According to the B-component diffusion key, all parts of the split B component are diffused to obtain all parts of the diffused B component.

[0179] The component splicing module 1500 is used to scramble and splice all parts of the diffused R component using an array splicing function to obtain the spliced R component, scramble and splice all parts of the diffused G component using an array splicing function to obtain the spliced G component, and scramble and splice all parts of the diffused B component using an array splicing function to obtain the spliced B component.

[0180] The component synthesis module 1600 is used to synthesize the spliced R component, the spliced G component, and the spliced B component to obtain the encrypted image of the color image.

[0181] This system extracts the R component, G component, and B component of a color image, and respectively generates an R-component scrambling transformation key, a G-component scrambling transformation key, and a B-component scrambling transformation key based on a three-dimensional chaotic neural network. According to the R-component scrambling transformation key, the R component of the color image is randomly scrambled row by row and column by column to obtain the scrambled R component. According to the G-component scrambling transformation key, the G component of the color image is randomly scrambled row by row and column by column to obtain the scrambled G component. According to the B-component scrambling transformation key, the B component of the color image is randomly scrambled row by row and column by column to obtain the scrambled B component. The scrambled R component, scrambled G component, and scrambled B component are respectively divided into multiple parts column by column, and an R-component diffusion key, a G-component diffusion key, and a B-component diffusion key are respectively generated based on the three-dimensional chaotic neural network. According to the R-component diffusion key, all parts of the divided R component are diffused to obtain all parts of the diffused R component. According to the G-component diffusion key, all parts of the divided G component are diffused to obtain all parts of the diffused G component. According to the B-component diffusion key, all parts of the divided B component are diffused to obtain all parts of the diffused B component. By using the key stream generated by the three-dimensional chaotic neural network to perform scrambling and diffusion operations on the three channels of the color image, the randomness is enhanced and the security is improved. All parts of the diffused R component are scrambled and spliced using an array splicing function to obtain the spliced R component. All parts of the diffused G component are scrambled and spliced using an array splicing function to obtain the spliced G component. All parts of the diffused B component are scrambled and spliced using an array splicing function to obtain the spliced B component. The spliced R component, spliced G component, and spliced B component are synthesized to obtain the encrypted image of the color image. The anti-attack ability is enhanced by scrambling again using the array splicing function, making the encryption more secure.

[0182] It should be noted that the embodiments of this system and the above-mentioned system embodiments are based on the same inventive concept. Therefore, the relevant content of the above method embodiments also applies to the embodiments of this system, and will not be elaborated here.

[0183] This application also provides a color image encryption electronic device based on a chaotic neural network, including: a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it realizes: the color image encryption method based on a chaotic neural network as described above.

[0184] The processor and the memory can be connected through a bus or other means.

[0185] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory optionally includes a memory remotely disposed relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0186] The non-transitory software programs and instructions required to implement the color image encryption method based on a chaotic neural network in the above embodiments are stored in the memory. When executed by a processor, the color image encryption method based on a chaotic neural network in the above embodiments is performed. For example, the method steps S101 to S106 described above are performed. Figure 1 in the method steps S101 to S106.

[0187] This application also provides a computer-readable storage medium storing computer-executable instructions for performing: the color image encryption method based on a chaotic neural network as described above.

[0188] The computer-readable storage medium stores computer-executable instructions, which are executed by a processor or a controller, for example, executed by a processor in the above electronic device embodiments, enabling the above processor to perform the color image encryption method based on a chaotic neural network in the above embodiments. For example, the method steps S101 to S106 described above are performed. Figure 1 in the method steps S101 to S106.

[0189] Those of ordinary skill in the art will understand that all or some of the steps and systems disclosed above can be implemented as software, firmware, hardware, and their appropriate combinations. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program units, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contains computer-readable instructions, data structures, program units, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery media.

[0190] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those of ordinary skill in the art, various changes can be made without departing from the gist of the present invention.

Claims

1. A color image encryption method based on a chaotic neural network, characterized in that, the color image encryption method based on a chaotic neural network includes: extracting the R component, G component, and B component of the color image; generating an R component scrambling transformation key, a G component scrambling transformation key, and a B component scrambling transformation key respectively based on a three-dimensional chaotic neural network, performing row and column random scrambling on the R component of the color image according to the R component scrambling transformation key to obtain the scrambled R component, performing row and column random scrambling on the G component of the color image according to the G component scrambling transformation key to obtain the scrambled G component, and performing row and column random scrambling on the B component of the color image according to the B component scrambling transformation key to obtain the scrambled B component; respectively splitting the scrambled R component, the scrambled G component, and the scrambled B component into multiple parts by columns; generating an R component diffusion key, a G component diffusion key, and a B component diffusion key respectively based on the three-dimensional chaotic neural network, diffusing all parts of the split R component according to the R component diffusion key to obtain all parts of the diffused R component, diffusing all parts of the split G component according to the G component diffusion key to obtain all parts of the diffused G component, and diffusing all parts of the split B component according to the B component diffusion key to obtain all parts of the diffused B component; performing scrambling and splicing on all parts of the diffused R component using an array splicing function to obtain the spliced R component, performing scrambling and splicing on all parts of the diffused G component using an array splicing function to obtain the spliced G component, and performing scrambling and splicing on all parts of the diffused B component using an array splicing function to obtain the spliced B component; synthesizing the spliced R component, the spliced G component, and the spliced B component to obtain the encrypted image of the color image.

2. The color image encryption method based on a chaotic neural network according to claim 1, characterized in that, the extracting the R component, G component, and B component of the color image includes: extracting the R component, G component, and B component of the color image using array separation in matlab, where the calculation formulas for the R component, G component, and B component of the color image are; where P is the color image, PR is the R component of the color image, PG is the G component of the color image, and PB is the B component of the color image.

3. The color image encryption method based on a chaotic neural network according to claim 2, characterized in that, the generating an R component scrambling transformation key based on a three-dimensional chaotic neural network, performing row and column random scrambling on the R component of the color image according to the R component scrambling transformation key to obtain the scrambled R component, includes: obtaining the M×N two-dimensional matrix of the R component of the color image, where M is the number of rows of the pixel points of the R component of the color image, and N is the number of columns of the pixel points of the R component of the color image; Two sets of chaotic sequences y 1 (z), y 2 (z) are obtained from the two-dimensional matrix by a three-dimensional chaotic neural network, and a scrambling transformation key is generated according to the chaotic sequences, where the scrambling transformation key is a random non-repeating sequence, and the calculation formula for generating the scrambling transformation key according to the chaotic sequences is: Among them, RandM(i) is the scrambling transformation key for the i-th row of the R component, RandN(j) is the scrambling transformation key for the j-th column of the R component, and floor() is the floor function; Construct a key pair according to the scrambling transformation key and the increasing sequence. Among them, the calculation formula for constructing the key pair according to the scrambling transformation key and the increasing sequence is: Among them, Mchange is a 2×M two-dimensional array key pair, and Nchange is a 2×N two-dimensional array key pair; Perform row and column random scrambling on the R component of the color image according to the key pair to obtain the scrambled R component. Among them, the calculation formula for performing row and column random scrambling on the R component of the color image according to the key pair is:

4. A color image encryption method based on a chaotic neural network according to claim 3, characterized in that The steps of respectively splitting the scrambled R component, the scrambled G component, and the scrambled B component into multiple parts by columns include: Split the scrambled R component into three parts by columns to obtain the first part of the R component, the second part of the R component, and the third part of the R component. Among them, the first part of the R component is the left one-third of the R component, the second part of the R component is the middle one-third of the R component, and the third part of the R component is the right one-third of the R component; Split the scrambled G component into three parts by columns to obtain the first part of the G component, the second part of the G component, and the third part of the G component. Among them, the first part of the G component is the left one-third of the G component, the second part of the G component is the middle one-third of the G component, and the third part of the G component is the right one-third of the G component; Split the scrambled B component into three parts by columns to obtain the first part of the B component, the second part of the B component, and the third part of the B component. Among them, the first part of the B component is the left one-third of the B component, the second part of the B component is the middle one-third of the B component, and the third part of the B component is the right one-third of the B component.

5. A color image encryption method based on a chaotic neural network according to claim 4, characterized in that Generate an R component diffusion key based on a three-dimensional chaotic neural network, and diffuse all parts of the segmented R component according to the R component diffusion key to obtain all parts of the diffused R component, including: Obtain two-dimensional arrays of the first part of the R component, the second part of the R component, and the third part of the R component. Among them, the calculation formula for obtaining two-dimensional arrays of the first part of the R component, the second part of the R component, and the third part of the R component is: Among them, LP is the two-dimensional array of the first part of the R component, CP is the two-dimensional array of the second part of the R component, and RP is the two-dimensional array of the third part of the R component; Generate three different chaotic sequences y through a three-dimensional chaotic neural network 1 , y 2 , y 3 , and calculate the R-component diffusion key through the chaotic sequences. Among them, the calculation formula for calculating the R-component diffusion key through the chaotic sequences is: where k(n 1 ) is the diffusion key of the first part of the R component, k(n 2 ) is the diffusion key of the second part of the R component, k(n 3 ) is the diffusion key of the third part of the R component; XOR encrypt the first part of the R component using the two-dimensional array of the first part of the R component and the diffusion key of the first part of the R component to obtain the first part of the diffused R component. XOR encrypt the second part of the R component using the two-dimensional array of the second part of the R component and the diffusion key of the second part of the R component to obtain the second part of the diffused R component. XOR encrypt the third part of the R component using the two-dimensional array of the third part of the R component and the diffusion key of the third part of the R component to obtain the third part of the diffused R component. Among them, the formula for XOR encrypting the first part of the R component using the two-dimensional array of the first part of the R component and the diffusion key of the first part of the R component to obtain the first part of the diffused R component, XOR encrypting the second part of the R component using the two-dimensional array of the second part of the R component and the diffusion key of the second part of the R component to obtain the second part of the diffused R component, and XOR encrypting the third part of the R component using the two-dimensional array of the third part of the R component and the diffusion key of the third part of the R component to obtain the third part of the diffused R component is as follows: Among them, CLR is the first part of the diffused R component, CCR is the second part of the diffused R component, CRR is the third part of the diffused R component, and bitxor(P,k) is the XOR function.

6. A color image encryption method based on a chaotic neural network according to claim 5, characterized in that, The formula for scrambling and splicing all parts of the diffused R component using the array splicing function is: CR = [CCR, CLR, CRR] Among them, CR is the spliced R component.

7. A color image encryption method based on a chaotic neural network according to claim 6, characterized in that, The formula for synthesizing the spliced R component, the spliced G component, and the spliced B component is: C = cat(3, CR, CG, CB) Among them, C is the encrypted image, CG is the spliced G component, CB is the spliced B component, and cat() is the splicing function.

8. A color image encryption method based on a chaotic neural network according to claim 7, characterized in that, The specific form of the three-dimensional chaotic neural network is: Among them, X(t), Y(t), and Z(t) respectively represent the state variables of 3 neurons, and the initial values of the neuron state variables are set as X(0), Y(0), and Z(0).

9. A color image encryption system based on a chaotic neural network, characterized in that, The color image encryption system based on a chaotic neural network includes: A component extraction module for extracting the R component, G component, and B component of a color image; The component scrambling module is used to generate the R-component scrambling transformation key, the G-component scrambling transformation key, and the B-component scrambling transformation key respectively based on the three-dimensional chaotic neural network, randomly scramble the rows and columns of the R-component of the color image according to the R-component scrambling transformation key to obtain the scrambled R-component, randomly scramble the rows and columns of the G-component of the color image according to the G-component scrambling transformation key to obtain the scrambled G-component, and randomly scramble the rows and columns of the B-component of the color image according to the B-component scrambling transformation key to obtain the scrambled B-component; The component splitting module is used to split the scrambled R-component, the scrambled G-component, and the scrambled B-component into multiple parts by columns respectively; The component diffusion module is used to generate the R-component diffusion key, the G-component diffusion key, and the B-component diffusion key respectively based on the three-dimensional chaotic neural network, diffuse all parts of the split R-component according to the R-component diffusion key to obtain all parts of the diffused R-component, diffuse all parts of the split G-component according to the G-component diffusion key to obtain all parts of the diffused G-component, and diffuse all parts of the split B-component according to the B-component diffusion key to obtain all parts of the diffused B-component; The component splicing module is used to scramble and splice all parts of the diffused R-component by using an array splicing function to obtain the spliced R-component, scramble and splice all parts of the diffused G-component by using an array splicing function to obtain the spliced G-component, and scramble and splice all parts of the diffused B-component by using an array splicing function to obtain the spliced B-component; The component synthesis module is used to synthesize the spliced R-component, the spliced G-component, and the spliced B-component to obtain the encrypted image of the color image.

10. A color image encryption device based on a chaotic neural network, characterized in that, it includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute the color image encryption method based on a chaotic neural network according to any one of claims 1 to 8.

Images