Video clip encryption method based on discrete sine memristor Rulkov mapping

Abstract

The invention discloses a video clip encryption method based on discrete sine memristor Rulkov mapping, local security protection of video data is realized by selecting a continuous key frame containing key information or key events in a video time dimension for encryption, most of the area of a phase space is filled with DSM-RM through attractor analysis, and therefore, the security of the video clip is greatly improved. A special digital circuit is constructed to capture an attractor of DSM-RM, so that not only is a theoretical analysis result verified, but also the application potential of the technology in the industrial field is shown. Experimental results show that the method provided by the invention maintains a relatively high encryption security level while reducing the consumption of computing resources, and provides a brand-new and effective solution for real-time video secure transmission.

Description

Technical Field

[0001] The present invention relates to the fields of communication, cryptography, and information technology, and particularly relates to a video clip encryption method based on a discrete sine memristive Rulkov map. Background Art

[0002] With the wide application of information acquisition tools such as smart phones and monitoring devices, the generation volume of video data has increased exponentially. However, the rapid growth of video data has also brought increasingly severe security challenges. Different from images, videos are composed of continuous image frames, with rich visual information and significant temporal correlation, which makes videos more vulnerable to attacks such as theft, tampering, or illegal access during transmission. Existing image encryption methods mainly target static two-dimensional data, while video encryption needs to process both the spatial information of image frames and the temporal information between frames simultaneously. Therefore, video encryption faces higher complexity. Traditional methods mostly adopt a frame-by-frame encryption method, treating the video as an image sequence and encrypting each frame separately. For example, some researchers have designed parallel encryption methods based on fractional-order chaotic maps, or improved the security by constructing dynamic S-boxes combined with pixel scrambling and diffusion. Although frame-by-frame encryption can improve security, it has a high computational burden in resource-constrained environments, especially when encrypting redundant background information in videos, with low efficiency.

[0003] For this reason, some researchers have proposed an encryption method based on the region of interest (ROI) of the video, only encrypting the target area to improve efficiency. However, when dealing with multi-category and continuous motion scenarios, the ROI method is difficult to accurately define the encryption range, especially in terms of protecting key frames in the time dimension, and it is easy to cause encryption vulnerabilities.

[0004] To address the above problems, this patent proposes a video encryption method based on action segmentation. Through action recognition and segmentation, continuous segments carrying key information in the video are extracted and encrypted centrally, thereby improving the security of key frames and reducing overall resource consumption.

[0005] In terms of key stream generation, chaotic systems are widely used in the field of encryption due to their initial value sensitivity and non-linear characteristics. In recent years, memristors, as a new type of non-linear circuit element, have gradually become an important component in the design of chaotic systems because they can retain the historical resistance state in the power-off state and can trigger complex dynamic behaviors. Compared with traditional chaotic maps, memristor-based systems have higher unpredictability and stronger dynamic retention capabilities. However, the complexity of existing memristive maps is limited, making it difficult to meet the requirements for the quality of key streams in high-security encryption systems. The Rulkov neuron model simulates the firing behavior of biological neurons under stimulation and has high sensitivity and strong non-linear characteristics. Existing research has proposed a memristor-improved Rulkov mapping model, but it still shows linear correlation in some phase spaces, which may lead to encryption vulnerabilities. Therefore, this patent designs a discrete sine memristive Rulkov mapping (DSM-RM). By introducing two memristors into the original model to enhance the dynamic complexity of the system, it can generate high-quality large-scale key streams and is suitable for video encryption tasks with high security requirements.

[0006] In summary, there are still bottlenecks in existing video encryption technologies in dealing with key frame protection and key stream generation.

[0007] The present invention combines the idea of action segment encryption with the design of a highly complex memristive chaotic system, and proposes a new video segment encryption method to achieve efficient and highly secure video protection.

[0008] The MSLID-TCN algorithm is from Gao S, Wu R, Liu S, et al. MSLID-TCN: multi-stage linear-index dilated temporal convolutional network for temporal action segmentation [J]. International Journal of Machine Learning and Cybernetics, 2025, 16: 567–581. Summary of the Invention

[0009] The object of the present invention is to solve the problem that in the prior art, videos are composed of continuous image frames, with rich visual information and significant temporal correlation, which makes videos more vulnerable to attacks such as theft, tampering, or illegal access during transmission.

[0010] To solve the above problems, the present invention provides a video segment encryption method based on discrete sine memristive Rulkov mapping, including:

[0011] S1: Construct a discrete sine memristive Rulkov mapping DSM-RM model by coupling a discrete memristor with the Rulkov mapping.

[0012] The formula for a class of discrete memristor models is:

[0013]

[0014] Among them, represents the memductance function, represents the independent variable of the memductance function. When , the equation becomes a discrete sine memristor model, and the formula is:

[0015]

[0016] Among them, i n and v n represent the current and voltage of the discrete sine memristor model respectively. a, b, and γ are the control parameters of the discrete sine memristor. When a = 0.2, b = 0.5, and γ = 0.9, they are the optimal parameters;

[0017] The mathematical expression of the Rulkov model is as follows:

[0018]

[0019] Among them, δ1, δ2, and δ3 are the control parameters of the Rulkov model. x n represents the fast variable, and y n represents the slow variable. Among them, the control parameter δ1 varies within the interval (4.5, 5.5), and δ2 = 0.2, δ3 = 0.01 are the optimal parameters;

[0020] The DSM-RM model introduces a discrete sine memristor into the Rulkov model. The expression of the DSM-RM model is:

[0021]

[0022] Among them, ε1 and ε2 are the adjustment parameters of the DSM-RM model, and ε1 and ε2 are set to vary within the interval (0.05, 0.45);

[0023] S2: Construct a dedicated digital circuit to capture the attractor of the DSM-RM model.

[0024] Adopt a DSP hardware platform, use a 32-bit floating-point DSP processor TMS320F28335 with a maximum clock frequency of 150 MHz and a dual-channel voltage output digital-to-analog converter DAC8552 with a 16-bit conversion accuracy; program and transmit the chaotic sequence generated by the PC to the DSP platform by setting parameters, and then convert it into an analog signal and output the attractor phase diagram of different phase planes through an oscilloscope.

[0025] S3: Extract key frames in the video using the temporal action segmentation algorithm;

[0026] Adopt the MSLID-TCN algorithm to separate the key frames of the video. The video is denoted as V with a length of T, then the video is represented as V 1:T = v1, v2, v3,..., v T , where v i represents the original information of each frame;

[0027] After extracting the features of each frame, the video is represented as F 1:T = f1, f2, f3,..., f T , where f i represents the features of the corresponding frame;

[0028] Use the MSLID-TCN algorithm to obtain the frame labels C of the video 1:T = c1, c2, c3,..., c T , and select the key frames of the frame labels of the video

[0029] S4: Generate the key stream of the encryption system based on the DSM-RM model;

[0030] The input in each encryption system is denoted as P, and the size of P is P1×P2. Calculate the encryption key k of each encryption system. The formula is:

[0031]

[0032] Use the key k to generate the parameters of the DSM-RM model and the initial values x(0) = 0.3060 + k, y(0) = 0.1996 + k, z(0) = 0.169 + k, δ1 = 5 + k, ε1 = 0.4 + k, ε2 = 0.4 + k. Fix a = 0.2, b = 0.5, γ = 0.9, δ2 = 0.2, δ3 = 0.01. Through iterative discrete sine memristive Rulkov mapping, generate the key streams X, Y, Z, and intercept the lengths of the key streams X, Y, Z. The formula is:

[0033] X = X(1:P1)

[0034] Y = Y(1:P2)

[0035] Z = Z(1:P1×P2)

[0036] Take the intercepted key stream X as the input, perform element sorting, and generate a position number matrix according to the sorting result. In the position number matrix, the value of each element represents the new position number where the corresponding element in the original matrix is located after sorting. Finally, obtain the row scrambling matrix F X; Using the same intercepted key stream Y as the input, a column scrambling matrix F with a new position number is obtained Y ;

[0037] The intercepted key stream Z is used to obtain the diffusion matrix F Z , and the formula is:

[0038] F Z = floor(Z × 10 10 ) mod 256

[0039] where floor is the floor function and mod is the modulo function, mapping the value of Z to the range [0, 255], and converting the dimension of Z from 1 × P1·P2 to P1 × P2;

[0040] S5: Encrypt the extracted key frames in combination with the key stream;

[0041] S5-1: According to the row scrambling matrix F X , the column scrambling matrix F Y , and the diffusion matrix F Z , the input P is encrypted to obtain the encrypted video frame P C , and the formula is:

[0042] When i = 1, j = 1:

[0043]

[0044] When i = 1, j = 2:P2:

[0045]

[0046] When i = 2:P1, j = 1:P2:

[0047]

[0048] S5-2: Encrypt the key frames in the same way as in step S5-1, and finally obtain the encrypted video.

[0049] In a preferred manner, the method in step S5-1 is to rearrange the encrypted frames and the unencrypted frames according to the frame numbers in the original video, so that the frame order of the synthesized video is the same as that of the original video, thus forming a complete encrypted video.

[0050] The beneficial effects of the present invention are as follows: 1. The present invention designs a video segment encryption algorithm EAS. Aiming at the problem that traditional video encryption methods cannot effectively encrypt key segments in the time dimension of videos, it extracts and encrypts consecutive key frames by means of action segmentation to achieve key content protection, significantly improving the security of the overall video. 2. The present invention constructs a new discrete sine memristive Rulkov map (DSM-RM) and systematically analyzes its dynamic characteristics. The results show that the system has good chaotic characteristics and high complexity. By implementing its attractor in a digital circuit, it is verified that the system has good engineering feasibility and is applicable to actual encryption application scenarios. 3. The present invention conducts experimental verification based on the public dataset 50Salads. The results show that the proposed encryption algorithm can save encryption time while ensuring security, especially applicable to environments with limited computing resources, and has good practical value and application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flowchart of the method of the present invention;

[0052] Figure 2 It is a structural diagram of the discrete sine memristive Rulkov map of the present invention;

[0053] Figure 3 It is a schematic diagram of the attractor analysis of the discrete sine memristive Rulkov map of the present invention with δ1 = 5, ε1 = 0.4, and ε2 = 0.4;

[0054] Figure 4 It is a schematic diagram of the attractor analysis of the discrete sine memristive Rulkov map of the present invention with δ1 = 5.1, ε1 = 0.05, and ε2 = 0.2;

[0055] Figure 5 It is a schematic diagram of the DSP implementation of the discrete sine memristive Rulkov map of the present invention;

[0056] Figure 6 It is a diagram of the video segment encryption result provided by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0057] Example 1:

[0058] The present invention proposes a video segment encryption method based on discrete sine memristive Rulkov mapping, aiming to solve the deficiencies of traditional video encryption technologies in terms of key frame protection, encryption efficiency, and key stream complexity. This method first extracts continuous action segments containing key information from the original video through action segmentation technology, and performs encryption processing on these segments, thereby avoiding ineffective encryption of redundant background frames and improving resource utilization. To enhance the security of the encryption system, the present invention designs a discrete sine memristive Rulkov mapping (DSM-RM). This system constructs a chaotic dynamic system with high complexity by introducing two memristors combined with sine functions to generate a high-quality pseudo-random key stream. Combining the key stream generated by DSM-RM, this paper constructs a segment encryption algorithm to perform pixel-level scrambling and diffusion operations on key frames in both spatial and temporal dimensions, achieving efficient and reliable video encryption. Experimental results show that this method has good security performance while significantly reducing the encryption overhead, and is applicable to various application scenarios with high requirements for encryption efficiency and security. As Figure 1 shown, a video segment encryption method based on discrete sine memristive Rulkov mapping includes:

[0059] S1: Coupling discrete memristors with Rulkov mapping to construct a discrete sine memristive Rulkov mapping DSM-RM model;

[0060] By combining discrete memristors with the Rulkov neuron model and introducing sine non-linear characteristics, the chaotic characteristics of the mapping are enhanced, laying a foundation for subsequent complex dynamic analysis and applications.

[0061] Since memristors are relatively convenient to implement in hardware applications, they have been widely used in the design of pseudo-random signal generators. A formula for a class of discrete memristor models is:

[0062]

[0063] where represents the memductance function, represents the independent variable of the memductance function. When , the equation becomes a discrete sine memristor model, and the formula is:

[0064]

[0065] where i n and v n respectively represent the current and voltage of the discrete sine memristor model, and a, b, and γ are the control parameters of the discrete sine memristor;

[0066] The Rulkov model can simulate the firing behavior of neurons. The fast variable x in the Rulkov modeln The interaction with the slow variable y n results in complex dynamic characteristics. Under appropriate parameter conditions, the Rulkov model can exhibit chaotic bursting behavior similar to biological neurons. The mathematical expression of the Rulkov model is as follows:

[0067]

[0068] where δ1, δ2, and δ3 are the control parameters of the Rulkov model, and x n represents the fast variable, and y n represents the slow variable;

[0069] The DSM-RM model introduces a discrete sine memristor into the Rulkov model, integrating the memory characteristics of the memristor with the nonlinear dynamic behavior of the Rulkov model, thereby being able to simulate the influence of magnetic feedback on neuron behavior. The structure of the DSM-RM is as Figure 2 shown. According to Figure 2 the DSM-RM structure shown, the expression of the DSM-RM model is:

[0070]

[0071] where ε1 and ε2 are the adjustment parameters of the DSM-RM model;

[0072] S2: Analyze the attractor of the discrete sine memristive Rulkov map. Through numerical simulation and phase space reconstruction methods, study the dynamic behavior of the system under different parameter conditions, and draw the attractor diagram to reveal its chaotic characteristics and attractor structure.

[0073] The initial state of the DSM-RM model is fixed as x(0) = 0.3060, y(0) = 0.1996, z(0) = 0.169. Observe the attractor behavior of the DSM-RM: δ1 = 5, ε1 = 0.4, ε2 = 0.4 and δ1 = 5.1, ε1 = 0.05, ε2 = 0.2; it shows that the attractor of the DSM-RM model covers a large area in the phase space;

[0074] Analyze the attractor of the DSM-RM model. Through numerical simulation and phase space reconstruction methods, study the dynamic behavior of the system under different parameter conditions, and draw the attractor diagram to reveal its chaotic characteristics and attractor structure.

[0075] Analyze the dynamic behavior of DSM-RM. The control parameters of the discrete sine memristor are fixed as a = 0.2, b = 0.5, and γ = 0.9. The control parameters of the Rulkov model are set as follows: δ1 varies within the interval (4.5, 5.5), δ2 = 0.2, δ3 = 0.01, and the adjustment parameters ε1 and ε2 of DSM-RM are set to vary within the interval (0.05, 0.45);

[0076] Attractors provide an intuitive way to observe the dynamic behavior of chaotic systems, revealing the evolution trajectory of the system's state in the phase space. By fixing the initial state of DSM-RM as x(0) = 0.3060, y(0) = 0.1996, z(0) = 0.169, observe the attractor behavior of DSM-RM under two different sets of parameters: δ1 = 5, ε1 = 0.4, ε2 = 0.4 and δ1 = 5.1, ε1 = 0.05, ε2 = 0.2, as Figure 3 and Figure 4 shown.

[0077] Attractor analysis shows that the attractors of DSM-RM cover a large area in the phase space, indicating that the system has extremely rich and complex dynamic behavior. Such a characteristic means that its key space is extremely broad, thus significantly increasing the difficulty of cracking passwords through brute force and exhaustive search.

[0078] S3: Construct a dedicated digital circuit to capture the attractor of the DSM-RM model;

[0079] Construct a dedicated digital circuit to capture the attractor of the discrete sine memristive Rulkov map. Based on digital hardware platforms such as FPGA or DSP, realize the real-time capture of the chaotic attractor generated by the discrete sine memristive Rulkov map to verify its feasibility and application potential.

[0080] In specific implementation, as a preferred implementation manner of the present invention, the step S3 specifically includes: To verify the feasibility of the proposed chaotic map in engineering applications, this study uses a DSP hardware platform to implement the phase diagram of the chaotic map and verify its authenticity. Using Figure 3 the parameters and initial values in Figure 5 shows the implementation process of DSM-RM on the DSP platform. The TMS320F28335 used is a commonly used 32-bit floating-point DSP processor with a maximum clock frequency of up to 150 MHz and powerful digital signal processing capabilities; DAC8552 is a dual-channel voltage output digital-to-analog converter with a 16-bit conversion accuracy. By programming and transmitting the chaotic sequence generated by the PC to the DSP platform, and then converting it into an analog signal and outputting the attractor phase diagrams of different phase planes through an oscilloscope, which is highly consistent with the simulation results, further verifying the practicality of DSM-RM and its application potential on the hardware platform.

[0081] S4: Extract the key frames in the video using the temporal action segmentation algorithm;

[0082] Using the temporal action segmentation algorithm, extract the key frames in the video. Perform time series analysis on the input video stream to extract the key frames reflecting the main motion changes, providing refined data for subsequent encryption processing.

[0083] Adopt the MSLID-TCN algorithm to separate the key frames of the video;

[0084] If the video is denoted as V with a length of T, then the video is represented as V 1:T = v1, v2, v3,..., v T , where v i represents the original information of each frame;

[0085] After extracting the features of each frame, the video is represented as F 1:T = f1, f2, f3,..., f T , where f i represents the features of the corresponding frame;

[0086] Use the MSLID-TCN algorithm to obtain the frame labels C of the video 1:T = c1, c2, c3,..., c T , and select the key frames of the frame labels of the video for input to the encryption system.

[0087] S5: Generate the key stream of the encryption system based on the DSM-RM model;

[0088] Generate the key stream of the cryptosystem using the DSM-RM model. Based on the chaotic characteristics of the DSM-RM model, construct a key stream with high randomness and high sensitivity to ensure its applicability to secure communication and encryption applications.

[0089] Based on the MSLID-TCN model, separate the key frames of the video and select the key frames of the frame labels of the video for encryption.

[0090] For each key frame Encrypt each independent key frame separately. The input in each encryption system is denoted as P, and the size of P is P1×P2;

[0091] Based on the input P, calculate the encryption key k for each encryption system. The formula is:

[0092]

[0093] Generate the parameters of the DSM-RM model and the initial values \(x(0)=0.3060 + k\), \(y(0)=0.1996 + k\), \(z(0)=0.169 + k\), \(\delta_1 = 5 + k\), \(\varepsilon_1 = 0.4 + k\), \(\varepsilon_2 = 0.4 + k\). Fix \(a = 0.2\), \(b = 0.5\), \(\gamma = 0.9\), \(\delta_2 = 0.2\), \(\delta_3 = 0.01\). Generate the key streams \(X\), \(Y\), \(Z\) through iterative discrete sine memristive Rulkov mapping;

[0094] Intercept the key streams \(X\), \(Y\), \(Z\) so that their lengths are suitable for the encryption system. The formula is:

[0095] \(X = X(1:P1)\)

[0096] \(Y = Y(1:P2)\)

[0097] \(Z = Z(1:P1\times P2)\)

[0098] Take the intercepted key stream \(X\) as the input, perform element sorting, and generate a position number matrix according to the sorting result. In the position number matrix, the value of each element represents the new position number where the corresponding element in the original matrix is sorted. Finally, obtain the row scrambling matrix \(F\); X ; Similarly, take the intercepted key stream \(Y\) as the input to obtain the column scrambling matrix \(F\) of the new position numbers; Y .

[0099] Take the intercepted key stream \(Z\) to obtain the diffusion matrix \(F\); Z , the formula is:

[0100] \(F\) Z = floor(Z\times10 10 ) mod 256

[0101] where floor is the flooring function and mod is the modulo function, which maps the value of \(Z\) to the range \([0, 255]\) and converts the dimension of \(Z\) from \(1\times P1\cdot P2\) to \(P1\times P2\);

[0102] S6: Encrypt the extracted key frames in combination with the key stream;

[0103] Adopt the stream cipher mechanism and use the generated key stream to perform pixel-level encryption on the key frames to ensure the confidentiality and anti-attack ability of the video content and achieve efficient and secure multimedia data protection.

[0104] S6-1: According to the row scrambling matrix \(F\) X , the column scrambling matrix \(F\) Y , the diffusion matrix \(F\) Z , encrypt the input \(P\) to obtain the encrypted video frame \(P\) C , the formula is:

[0105] When \(i = 1\), \(j = 1\):

[0106]

[0107] When i = 1, j = 2: P2:

[0108]

[0109] When i = 2: P1, j = 1: P2:

[0110]

[0111] S6-2: For the key frames Encrypt in the manner of step S6-1, and finally obtain the encrypted video. That is, rearrange the encrypted frames and the unencrypted frames according to the frame numbers in the original video to ensure that the frame order of the synthesized video is the same as that of the original video, and a complete encrypted video can be formed. Taking the rgb-06-2 video in the 50Salads video dataset as an example, the encryption result is as Figure 6 shown, where the "peel cucumber" action is regarded as an important action. To verify the effectiveness of the method of the present invention, the method of the present invention was verified through simulation experiments, and the simulation results are as Figure 6 shown.

[0112] Figure 3 In: (a) X-Y plane attractor of the discrete sine memristive Rulkov map; (b) X-Z plane attractor of the discrete sine memristive Rulkov map; (c) Y-Z plane attractor of the discrete sine memristive Rulkov map;

[0113] Figure 4 In: (a) X-Y plane attractor of the discrete sine memristive Rulkov map; (b) X-Z plane attractor of the discrete sine memristive Rulkov map; (c) Y-Z plane attractor of the discrete sine memristive Rulkov map.

[0114] Figure 6 Taking the rgb-06-2 video in the 50Salads video dataset as an example, the "peel cucumber" action is regarded as an important action.

[0115] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of the present invention claimed is defined by the appended claims and their equivalents.

Claims

1. A video clip encryption method based on a discrete sine memristive Rulkov map, characterized in that Including: S1: Coupling discrete memristors with the Rulkov map to construct a discrete sine memristive Rulkov map DSM-RM model; The formula of a class of discrete memristor models is: Among them, represents the memductance function, represents the independent variable of the memductance function. When , the equation becomes the discrete sine memristor model, and the formula is: where, i n and v n respectively represent the current and voltage of the discrete sine memristor model, and a, b, and γ are the control parameters of the discrete sine memristor. When a = 0.2, b = 0.5, and γ = 0.9, they are the optimal parameters of the discrete sine memristor model; The mathematical expression of the Rulkov model is as follows: Among them, δ1, δ2, and δ3 are the control parameters of the Rulkov model, and x n represents the fast variable, and y n represents the slow variable, where the control parameter δ1 varies within the interval (4.5, 5.5), δ2 = 0.2, and δ3 = 0.01 are the preferred parameters of the Rulkov model; The DSM-RM model introduces discrete sine memristors into the Rulkov model, and the expression of the DSM-RM model is: Among them, ε1 and ε2 are the adjustment parameters of the DSM-RM model, and ε1 and ε2 are set to vary within the interval (0.05, 0.45); S2: Construct a dedicated digital circuit to capture the attractor of the DSM-RM model; Adopt a DSP hardware platform, use a 32-bit floating-point DSP processor TMS320F28335 with a maximum clock frequency of 150 MHz and a dual-channel voltage output digital-to-analog converter DAC8552 with a 16-bit conversion accuracy; program and transmit the chaotic sequence generated by the PC to the DSP platform by setting parameters, and then convert it into an analog signal and output the attractor phase diagram of different phase planes through an oscilloscope; S3: Use the temporal action segmentation algorithm to extract key frames in the video; Using the MSLID-TCN algorithm, the key frames of the video are separated. The video is denoted as V with a length of T, so the video is represented as V 1:T = v1, v2, v3,..., v T , where v i represents the original information of each frame; After feature extraction for each frame, the video is represented as F 1:T = f1, f2, f3,..., f T , where f i represents the features of the corresponding frame; Obtain the frame label C of the video using the MSLID-TCN algorithm 1:T = c1, c2, c3,..., c T , and select the key frames of the frame labels of the video S4: Generate the key stream of the encryption system based on the DSM-RM model; The input in each encryption system is denoted as P, and the size of P is P1×P2. Calculate the encryption key k of each encryption system. The formula is: Use the key k to generate the parameters of the DSM-RM model and the initial values x(0)=0.3060 + k, y(0)=0.1996 + k, z(0)=0.169 + k, δ1 = 5 + k, ε1 = 0.4 + k, ε2 = 0.4 + k. Fix a = 0.2, b = 0.5, γ = 0.9, δ2 = 0.2, δ3 = 0.

01. By iterating the discrete sine memristive Rulkov map, generate the key streams X, Y, Z, and intercept the lengths of the key streams X, Y, Z. The formula is: X = X(1:P1) Y = Y(1:P2) Z = Z(1:P1×P2) Taking the intercepted key stream X as the input, perform element sorting and generate a position number matrix according to the sorting result. In the position number matrix, the value of each element represents the new position number where the corresponding element in the original matrix is located after sorting, and finally obtain the row scrambling matrix F X ; Similarly, taking the intercepted key stream Y as the input, obtain the column scrambling matrix F of the new position numbers Y ; The intercepted key stream Z is used to obtain the diffusion matrix F Z , and the formula is: F Z = floor(Z × 10 10 ) mod 256 Among them, floor is the rounding function, mod is the remainder function, map the value of Z to the range [0, 255], and convert the dimension of Z from 1×P1·P2 to P1×P2; S5: Encrypt the extracted key frames in combination with the key stream; S5-1: According to the row scrambling matrix F X , the column scrambling matrix F Y , the diffusion matrix F Z , encrypt the input P to obtain the encrypted video frame P C , the formula is: When i = 1, j = 1: When i = 1, j = 2:P2: When i = 2:P1, j = 1:P2: S5-2: For key frames Encrypt them in the manner of step S5-1, and finally obtain the encrypted video.

2. The video segment encryption method based on the discrete sine memristive Rulkov map according to claim 1, characterized in that The method in step S5-1 is to rearrange the encrypted frames and the unencrypted frames according to the frame numbers in the original video so that the frame order of the synthesized video is the same as that of the original video, and then a complete encrypted video can be formed.

Images