Multi-face image encryption method based on non-adjacent coupled mapping

Abstract

The application provides a multi-face image encryption method based on non-adjacent coupling mapping. The method comprises the following steps: face detection, face information extraction, bit-level cyclic shift, privacy information and non-privacy information integration, pixel-level cyclic shift and three-dimensional double-index diffusion, so that multiple face images with different sizes can be encrypted each time, and the encrypted privacy degree is high.

Description

Technical Field

[0001] This invention relates to the field of image encryption technology, and more particularly to a method for encrypting multi-face images based on non-adjacent coupling mapping. Background Technology

[0002] The face is more private than other areas, and current encryption algorithms mainly focus on single-image encryption containing a single face, while research on encryption of multi-face images is relatively weak.

[0003] Existing image encryption methods employ cyclic shifting techniques, most of which involve pixel shifting or bit shifting on a plane, while indexing techniques are mostly used in a two-dimensional plane.

[0004] In view of this, a multi-face image encryption method based on non-adjacent coupling mapping is proposed. Summary of the Invention

[0005] To address the aforementioned technical problems with encryption of multiple face images, this invention provides a method for encrypting multiple face images based on non-adjacent coupled mapping (NCMLWDP). This invention primarily utilizes NCMLWDP to design an encryption scheme for protecting multiple face images. This scheme can simultaneously process grayscale or color face images of different sizes. This simultaneous cross-plane encryption ensures a certain degree of coupling between the encrypted images. The scheme includes: face detection, face information extraction, bit-level cyclic shifting, integration of private and non-privacy information, pixel-level cyclic shifting, and three-dimensional double-index diffusion. Therefore, each encryption can encrypt multiple face images of different sizes, and the encrypted images have a high degree of privacy.

[0006] The technical means employed in this invention are as follows:

[0007] This invention provides a method for encrypting multi-face images based on non-adjacent coupling mapping, comprising:

[0008] Face information and non-face information of the original image are extracted using face detection technology. The non-face information is then used to fill in the face information, and the information is sorted to obtain privacy information.

[0009] A key is obtained by processing the original image and the privacy information using a hash function;

[0010] The parameters and initial values ​​of the non-adjacent coupled mapping lattice are obtained according to the key. The non-adjacent coupled mapping lattice is iterated according to the parameters and the initial values ​​to obtain the first to the fifteenth chaotic sequences.

[0011] The encrypted privacy information is obtained by performing bit-interleaved cyclic shifting on the privacy information according to the first to fourth chaotic sequences;

[0012] The encrypted privacy information is integrated with the non-face information, and the gaps are filled with the number 0 to obtain a three-dimensional matrix. According to the 5th to 8th chaotic sequences, the three-dimensional matrix is ​​subjected to pixel-level interleaved cyclic shift to obtain the final matrix.

[0013] Based on the 13th to the 15th chaotic sequences and the face detection technology, face location information is obtained. Based on the face location information, the initial value and the number of iterations of the one-dimensional chaotic mapping system are obtained. Based on the initial value and the number of iterations, the one-dimensional chaotic mapping system is iterated to obtain the 1st to the 4th new chaotic sequences.

[0014] Three-dimensional information is obtained by performing three-dimensional diffusion on the final matrix based on the 9th to 12th chaotic sequences and the 1st to 4th new chaotic sequences;

[0015] The three-dimensional information is converted into encrypted images layer by layer.

[0016] Further, the step of performing bit-interleaved cyclic shifting on the privacy information according to the first to fourth chaotic sequences to obtain the encrypted privacy information includes:

[0017] The first to fourth chaotic sequences are divided into a first sequence group. The first chaotic sequence is sorted to obtain the first row index sequence. The second chaotic sequence is sorted to obtain the second row index sequence. The third and fourth row value sequences are obtained based on the third and fourth chaotic sequences.

[0018] The encrypted privacy information is obtained by performing a bit-level cyclic shift on the privacy information based on the first row index sequence, the second row index sequence, the third row value sequence, and the fourth row value sequence.

[0019] Furthermore, the third row value sequence and the fourth row value sequence obtained from the third chaotic sequence and the fourth chaotic sequence are calculated according to the following formula:

[0020]

[0021] Wherein, BP is the layer number of the privacy information, By1 is the value sequence of the third row, By2 is the value sequence of the fourth row, floor is the floor function, mod is the modulo function, and AI is the first sequence group.

[0022] Further, the step of performing pixel-level interleaved cyclic shifting on the three-dimensional matrix according to the 5th to 8th chaotic sequences to obtain the final matrix includes:

[0023] The 5th to 8th chaotic sequences are divided into a second sequence group. The 5th chaotic sequence is sorted to obtain the 5th row index sequence. The 6th chaotic sequence is sorted to obtain the 6th row index sequence. The 7th and 8th row value sequences are obtained based on the 7th and 8th chaotic sequences.

[0024] The final matrix is ​​obtained by performing pixel-level interleaved cyclic shifting on the three-dimensional matrix based on the index sequence of the 5th row, the index sequence of the 6th row, the value sequence of the 7th row, and the value sequence of the 8th row.

[0025] Furthermore, the 7th row value sequence and the 8th row value sequence obtained from the 7th and 8th chaotic sequences are calculated according to the following formula:

[0026]

[0027] Wherein, IIA is the three-dimensional matrix, Ay1 is the value sequence of the 7th row, Ay2 is the value sequence of the 8th row, BI is the second sequence group, AP is the layer number of the three-dimensional matrix, sizeof is the function to calculate the data size, floor is the floor function, and mod is the modulo function.

[0028] Further, the face location information is obtained based on the 13th to 15th chaotic sequences and the face detection technology. The initial iteration value and iteration number of the one-dimensional chaotic mapping system are then obtained based on the face location information, calculated according to the following formula:

[0029]

[0030] Wherein, β1, β2, β3, and β4 constitute the initial values ​​of the iteration, cct is the iteration number, L1 is the iteration number, sum is the summation function, C1 is the 13th chaotic sequence, C2 is the 14th chaotic sequence, C3 is the 15th chaotic sequence, X and Y represent the length and width of each face, and M and N represent the length and width of each initial image, respectively.

[0031] Further, the step of iterating the one-dimensional chaotic mapping system according to the initial iteration value and the number of iterations to obtain the first to fourth new chaotic sequences includes:

[0032] The one-dimensional chaotic mapping system is iterated using the initial iteration value and the number of iterations, and the results of the first 19 iterations are discarded to obtain the first to the fourth new chaotic sequences.

[0033] Further, the step of obtaining three-dimensional information by three-dimensional diffusion of the final matrix based on the 9th to 12th chaotic sequences and the 1st to 4th new chaotic sequences includes:

[0034] The third and fourth value sequences are obtained based on the third and fourth new chaotic sequences, and are calculated according to the following formula:

[0035]

[0036] Wherein, F1 is the third new chaotic sequence, F2 is the fourth new chaotic sequence, F3 is the third value sequence, F4 is the fourth value sequence, sum is the summation function, and H and W are face size information.

[0037] Furthermore, the three-dimensional information is obtained by performing three-dimensional diffusion on the final matrix based on the 9th to 12th chaotic sequences and the 1st to 4th new chaotic sequences, calculated according to the following formula:

[0038]

[0039] Wherein, CC is the ciphertext matrix of the three-dimensional information, IA is the final matrix, CX, CY, and CZ are the position index sequences of the three directions of the ciphertext matrix, AX, AY, and AZ are the position index sequences of the three directions of the final matrix, and tmp is the previous iteration value of the ciphertext matrix in each iteration.

[0040] Compared with the prior art, the present invention has the following advantages:

[0041] 1. The multi-face image encryption method based on non-adjacent coupling mapping provided by this invention proposes a new non-adjacent coupling mapping lattice, which has a wider range of parameters compared with other spatiotemporal chaotic models, and each lattice can achieve an extremely ideal chaotic state.

[0042] 2. The multi-face image encryption method based on non-adjacent coupling mapping provided by this invention proposes an encryption scheme that combines face detection and chaos, so that multiple face images of different sizes can be encrypted for each encryption, and each face image can contain a face.

[0043] 3. The multi-face image encryption method based on non-adjacent coupling mapping provided by this invention proposes an image obfuscation method that combines bit-level scrambling and pixel-level scrambling. Bit-level scrambling is used to enhance the privacy of face information after encryption.

[0044] 4. The multi-face image encryption method based on non-adjacent coupling mapping provided by this invention proposes a three-dimensional cyclic shift scrambling method, which realizes pixel-level cyclic shifting between multiple images.

[0045] 5. The multi-face image encryption method based on non-adjacent coupling mapping provided by this invention proposes three-dimensional double-index diffusion, which uses chaotic sequences to link the positional relationships before and after diffusion, thereby improving the encryption effect under the same time complexity. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 This is a flowchart of a multi-face image encryption method based on non-adjacent coupling mapping provided by the present invention.

[0048] Figure 2 A schematic diagram for key generation.

[0049] Figure 3 This is a schematic diagram of a cyclic shift technique.

[0050] Figure 4 This is another schematic diagram of the cyclic shift technique. Detailed Implementation

[0051] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0052] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0053] Combination Figure 1 , Figure 2 and Figure 3 , Figure 1 This is a flowchart of a multi-face image encryption method based on non-adjacent coupling mapping provided by the present invention. Figure 2 A schematic diagram for key generation. Figure 3 This is a schematic diagram of a cyclic shift technique. Figure 4 The diagram below illustrates another cyclic shift technique, demonstrating a specific embodiment of the multi-face image encryption method based on non-adjacent coupling mapping provided by the present invention, including:

[0054] Using face detection technology, extract the original images P1, P2, P3, ..., P... n The facial information and non-facial information are used to fill in the facial information with non-facial information and then arranged to obtain the privacy information IRB;

[0055] The key K is obtained by processing the original image and privacy information IRB using a hash function;

[0056] Specifically, refer to Figure 2 The key K consists of a first key K1 generated by the hash function SHA-384 and information related to a second key K2 obtained by the face detection algorithm. The first key K1 is a fixed-length sequence of 384 bits, and the second key K2 is a non-fixed-length sequence. The second key K2 consists of four parts: avg_1 and avg_2 represent the average value of all image pixels and the average value of face pixels, respectively; (M i N i (Xj, Yj, Wj, Lj) represents the size of each image; (Xj, Yj, Wj, Lj) represents the location of all faces in each image and the size of each face. The first key K1 and the second key K2 are used to generate the parameters of the Non-Adjacent Coupled Mapping Grid (NCMLWDP) and the One-Dimensional Chaotic Mapping System (1D-ISAL), as well as the initial iteration values ​​of the Non-Adjacent Coupled Mapping Grid (NCMLWDP).

[0057] The parameters (e, d, μ, γ) and initial value A of the non-adjacent coupled mapping lattice (NCMLWDP) are obtained based on the key K. The non-adjacent coupled mapping lattice (NCMLWDP) is iterated based on the parameters (e, d, μ, γ) and initial value A to obtain the first to the fifteenth chaotic sequences.

[0058] The parameters (e, d, μ, γ) and initial value A of the non-adjacent coupled mapping lattice (NCMLWDP) are obtained based on the key K and calculated in the following manner:

[0059]

[0060] Specifically, the non-adjacent coupled mapping lattice (NCMLWDP) is iterated based on parameters (e, d, μ, γ) and initial value A to obtain chaotic sequences 1 to 15, including:

[0061] The initial value A and the parameters (e, d, μ, γ) are substituted into the non-adjacent coupled mapping lattice (NCMLWDP) and iterated L1 times, L1 = max{M, N} + 48. The first 48 iterations are to eliminate the influence of the initial value and make the chaotic system tend to stabilize.

[0062] The encrypted privacy information IB is obtained by performing bit-interleaved cyclic shifting on the privacy information IRB based on the first chaotic sequence AI1 to the fourth chaotic sequence AI4, including:

[0063] The first chaotic sequence AI1 to the fourth chaotic sequence AI4 are divided into the first sequence group. Sort the first chaotic sequence AI1 to obtain the first row index sequence B. x Sort the second chaotic sequence AI2 to obtain the second row index sequence B. y The third row value sequence B is obtained based on the third chaotic sequence AI3 and the fourth chaotic sequence AI4. y1 and the value sequence B in row 4 y2 ;

[0064] Based on the index sequence B in row 1 x The second row index sequence B y The third row of values, sequence B y1 and the value sequence B in row 4 y2 The encrypted privacy information IB is obtained by performing a bit-level cyclic shift on the privacy information IRB.

[0065] Specifically, the third row value sequence B is obtained based on the third chaotic sequence AI3 and the fourth chaotic sequence AI4. y1 and the value sequence B in row 4 y2 Calculate according to the following formula:

[0066]

[0067] Where BP is the layer number of privacy information, By1 is the value sequence of the 3rd row, By2 is the value sequence of the 4th row, floor is the floor function, mod is the modulo function, and AI is the first sequence group.

[0068] Specifically, based on the index sequence B of row 1 x The second row index sequence B y The third row of values, sequence B y1 and the value sequence B in row 4 y2 The encrypted privacy information IB is obtained by performing a bit-level cyclic shift on the privacy information IRB, and is calculated as follows:

[0069]

[0070] The encrypted privacy information IB is integrated with non-facial information, and gaps are filled with the digit 0 to obtain a three-dimensional matrix IIA. Based on the 5th chaotic sequence BI1 to the 8th chaotic sequence BI4, the three-dimensional matrix IIA is subjected to pixel-level interleaved cyclic shifting to obtain the final matrix IA, including:

[0071] The fifth chaotic sequence BI1 to the eighth chaotic sequence BI4 are divided into the second sequence group. Sort the fifth chaotic sequence BI1 to obtain the fifth row index sequence IA. x Sort the 6th chaotic sequence BI2 to obtain the 6th row index sequence IA. y The 7th row value sequence A is obtained based on the 7th chaotic sequence BI3 and the 8th chaotic sequence BI4. y1 and the value sequence A in row 8 y2 ;

[0072] Based on the index sequence IA in row 5 x 6th row index sequence IA y The 7th row value sequence A y1 and the value sequence A in row 8 y2 The final matrix IA is obtained by performing pixel-level interleaved cyclic shifting on the three-dimensional matrix IIA.

[0073] Specifically, the 7th row value sequence A is obtained based on the 7th chaotic sequence BI3 and the 8th chaotic sequence BI4. y1 and the value sequence A in row 8 y2 Calculate according to the following formula:

[0074]

[0075] Where IIA is a three-dimensional matrix, Ay1 is the value sequence of the 7th row, Ay2 is the value sequence of the 8th row, BI is the second sequence group, AP is the layer number of the three-dimensional matrix, sizeof is the function to calculate the data size, floor is the floor function, and mod is the modulo function. Specifically, based on the index sequence IA of the 5th row... x 6th row index sequence IA y The 7th row value sequence A y1 and the value sequence A in row 8 y2 The algorithm for obtaining the final matrix IA by performing pixel-level interleaved cyclic shifting on the three-dimensional matrix IIA is described in Algorithm2.

[0076] Based on the 13th chaotic sequence C1 to the 15th chaotic sequence C3 and the face detection technology, the face location information is obtained. Based on the face location information, the initial values ​​(β1, β2, β3, β4) and the number of iterations cct of the one-dimensional chaotic mapping system (1D-ISAL) are obtained. Based on the initial values ​​(β1, β2, β3, β4) and the number of iterations cct, the one-dimensional chaotic mapping system (1D-ISAL) is iterated to obtain the 1st new chaotic sequence E1 to the 4th new chaotic sequence F2.

[0077] Specifically, face location information is obtained based on chaotic sequences C1 (13th) to C3 (15th) and face detection technology. Based on this face location information, the initial iteration values ​​(β1, β2, β3, β4) and the iteration count cct of the one-dimensional chaotic mapping system (1D-ISAL) are calculated using the following formula:

[0078]

[0079] Where β1, β2, β3, and β4 constitute the initial values ​​of the iteration, cct is the iteration number, L1 is the iteration number, sum is the summation function, C1 is the 13th chaotic sequence, C2 is the 14th chaotic sequence, C3 is the 15th chaotic sequence, X and Y represent the length and width of each face, and M and N represent the length and width of each initial image, respectively.

[0080] Specifically, the one-dimensional chaotic mapping system (1D-ISAL) is iterated based on the initial iteration values ​​(β1, β2, β3, β4) and the iteration number cct to obtain the first new chaotic sequence E1 to the fourth new chaotic sequence F2, including:

[0081] The initial values ​​(β1, β2, β3, β4) and the number of iterations cct are used to iterate the one-dimensional chaotic mapping system (1D-ISAL), and the results of the first 19 iterations are discarded to obtain the first new chaotic sequence E1, the second new chaotic sequence E2, the third new chaotic sequence F1, and the fourth new chaotic sequence F2.

[0082] Based on the chaotic sequences from the 9th to the 12th and the first to the fourth new chaotic sequences E1, the final matrix IA is subjected to three-dimensional diffusion to obtain three-dimensional information, including:

[0083] Based on the first new chaotic sequence E1 and the second new chaotic sequence E2, the first value sequence AZ and the second value sequence CZ are obtained. Specifically, E1 and E2 are indexed and sorted every AP positions to obtain two index sequences AZ and CZ.

[0084] Based on the third new chaotic sequence F1 and the fourth new chaotic sequence F2, the third value sequence F3 and the fourth value sequence F4 are obtained, and calculated according to the following formula:

[0085]

[0086] Where F1 is the third new chaotic sequence, F2 is the fourth new chaotic sequence, F3 is the third value sequence, F4 is the fourth value sequence, sum is the summation function, and H and W are the face size information.

[0087] The final matrix IA is subjected to three-dimensional diffusion based on the 9th to 12th chaotic sequences and the 1st to 4th new chaotic sequences E1 to F2 to obtain three-dimensional information, which is calculated according to the following formula:

[0088]

[0089] Where CC is the ciphertext matrix of the three-dimensional information, IA is the final matrix, CX, CY, and CZ are the position index sequences of the three directions of the ciphertext matrix, AX, AY, and AZ are the position index sequences of the three directions of the final matrix, and tmp is the previous iteration value of the ciphertext matrix in each iteration.

[0090] The 3D information CC is converted into encrypted images layer by layer.

[0091] The multi-face image encryption method based on non-adjacent coupling mapping provided by this invention encrypts face images that resemble noise; without decryption, the receiver cannot discern the image content, thus protecting the image. When the receiver uses the correct key and corresponding algorithm for decryption, the encrypted image is decrypted to reveal an image indistinguishable from the original. Key space and sensitivity tests prove that the key fully meets encryption requirements. Tests on multiple indicators of the encrypted image, including histogram, correlation, NPCR, and UACI, show that the encryption effect of the multi-face image encryption method based on non-adjacent coupling mapping provided by this invention is excellent. Furthermore, robustness, known-plaintext attack, and chosen-plaintext attack tests were conducted, and the decryption results show that the encrypted image of the multi-face image encryption method based on non-adjacent coupling mapping provides good decryption visual quality even under these attacks. In summary, the multi-face image encryption method based on non-adjacent coupling mapping provided by this invention has excellent encryption performance on face images.

[0092] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0093] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for encrypting multi-face images based on non-adjacent coupling mapping, characterized in that, include: Face information and non-face information of the original image are extracted using face detection technology. The non-face information is then used to fill in the face information, and the information is sorted to obtain privacy information. A key is obtained by processing the original image and the privacy information using a hash function; The parameters and initial values ​​of the non-adjacent coupled mapping lattice are obtained according to the key. The non-adjacent coupled mapping lattice is iterated according to the parameters and the initial values ​​to obtain the first to the fifteenth chaotic sequences. The encrypted privacy information is obtained by performing bit-interleaved cyclic shifting on the privacy information according to the first to fourth chaotic sequences; The encrypted privacy information is integrated with the non-face information, and the gaps are filled with the number 0 to obtain a three-dimensional matrix. According to the 5th to 8th chaotic sequences, the three-dimensional matrix is ​​subjected to pixel-level interleaved cyclic shift to obtain the final matrix. Based on the 13th to the 15th chaotic sequences and the face detection technology, face location information is obtained. Based on the face location information, the initial value and the number of iterations of the one-dimensional chaotic mapping system are obtained. Based on the initial value and the number of iterations, the one-dimensional chaotic mapping system is iterated to obtain the 1st to the 4th new chaotic sequences. Three-dimensional information is obtained by performing three-dimensional diffusion on the final matrix based on the 9th to 12th chaotic sequences and the 1st to 4th new chaotic sequences, including: The third and fourth value sequences are obtained based on the third and fourth new chaotic sequences, and are calculated according to the following formula: ； in, For the third new chaotic sequence, For the fourth new chaotic sequence, For the third value sequence, For the fourth value sequence, The summation function is used, where H and W represent face size information. ； in, This is the ciphertext matrix of the three-dimensional information. For the final matrix, , , These are the position index sequences for the three directions of the ciphertext matrix. , , The final matrix is ​​a sequence of position indices for its three directions. This represents the value of the ciphertext matrix in the previous iteration for each iteration; The three-dimensional information is converted into encrypted images layer by layer.

2. The method for encrypting multi-face images based on non-adjacent coupling mapping according to claim 1, characterized in that, The step of performing bit-interleaved cyclic shifting on the privacy information according to the first to fourth chaotic sequences to obtain the encrypted privacy information includes: The first to fourth chaotic sequences are divided into a first sequence group. The first chaotic sequence is sorted to obtain the first row index sequence. The second chaotic sequence is sorted to obtain the second row index sequence. The third and fourth row value sequences are obtained based on the third and fourth chaotic sequences. The encrypted privacy information is obtained by performing a bit-level cyclic shift on the privacy information based on the first row index sequence, the second row index sequence, the third row value sequence, and the fourth row value sequence.

3. The method for encrypting multi-face images based on non-adjacent coupling mapping according to claim 2, characterized in that, The third and fourth row value sequences are obtained based on the third and fourth chaotic sequences, and are calculated according to the following formula: ； in, The number of layers of the privacy information. For the value sequence of the third row, For the value sequence of the 4th row, This is the floor function. For modulo function, This is the first sequence group.

4. The method for encrypting multi-face images based on non-adjacent coupling mapping according to claim 1, characterized in that, The step of performing pixel-level interleaved cyclic shifting on the three-dimensional matrix according to the 5th to 8th chaotic sequences to obtain the final matrix includes: The 5th to 8th chaotic sequences are divided into a second sequence group. The 5th chaotic sequence is sorted to obtain the 5th row index sequence. The 6th chaotic sequence is sorted to obtain the 6th row index sequence. The 7th and 8th row value sequences are obtained based on the 7th and 8th chaotic sequences. The final matrix is ​​obtained by performing pixel-level interleaved cyclic shifting on the three-dimensional matrix based on the index sequence of the 5th row, the index sequence of the 6th row, the value sequence of the 7th row, and the value sequence of the 8th row.

5. The method for encrypting multi-face images based on non-adjacent coupling mapping according to claim 4, characterized in that, The values ​​for the 7th and 8th rows, obtained from the 7th and 8th chaotic sequences, are calculated using the following formula: ； in, For the three-dimensional matrix, For the value sequence of the 7th row, For the value sequence of the 8th row, For the second sequence group, The number of layers in the three-dimensional matrix. A function for calculating data size. This is the floor function. This is the modulo function.

6. The method for encrypting multi-face images based on non-adjacent coupling mapping according to claim 1, characterized in that, The face location information is obtained by combining the chaotic sequences from the 13th to the 15th with the face detection technology. Based on the face location information, the initial value and the number of iterations of the one-dimensional chaotic mapping system are obtained, calculated according to the following formula: ； in, , , , The initial values ​​constitute the iteration. The number of iterations, For the number of iterations, For the summation function, C 1 represents the 13th chaotic sequence. C 2 is the 14th chaotic sequence. C 3 represents the 15th chaotic sequence, where X and Y represent the length and width of each face, and M and N represent the length and width of each initial image, respectively.

7. The method for encrypting multi-face images based on non-adjacent coupling mapping according to claim 1, characterized in that, The step of iterating the one-dimensional chaotic mapping system according to the initial iteration value and the number of iterations to obtain the first to fourth new chaotic sequences includes: The one-dimensional chaotic mapping system is iterated using the initial iteration value and the number of iterations, and the results of the first 19 iterations are discarded to obtain the first to the fourth new chaotic sequences.

Images