RGB-D video encryption and decryption method based on hyperchaos technology and octley fractional order Hartley transformation

Abstract

The invention discloses an RGB-D video encryption and decryption method based on a hyper-chaos technology and octlet fractional order Hartley transformation, and the encryption method comprises the following steps: S1, constructing a two-dimensional hyper-chaos system ICCHM based on ICMIC chaos mapping and Chebyshev chaos mapping, and generating a random matrix and a random sequence required by encryption by using the two-dimensional hyper-chaos system ICCHM; s2, defining octlet fractional order Hartley transformation and inverse transformation of the octlet fractional order Hartley transformation; and S3, carrying out RGB-D video encryption processing. According to the method disclosed by the invention, integral encryption of RGB-D is realized, two types of data with difference can be processed at the same time, and relatively high video encryption efficiency, relatively large key space and relatively high security are achieved; in addition, the encryption key management and transmission of the method are very convenient, and common cryptographic attacks can be effectively resisted.

Description

Technical Field

[0001] The present invention belongs to the technical field of information security, and particularly relates to an RGB-D video encryption and decryption method based on hyperchaos technology and octonion fractional-order Hartley transform. Background Technique

[0002] RGB-D videos containing color and depth information can reconstruct realistic 3D scenes and have been widely used in fields such as autonomous driving systems, robot navigation, virtual and augmented reality. With the generalization of network threats and the normalization of unauthorized access to sensitive information, and since videos usually contain rich and important information, ensuring the confidentiality and integrity of transmitted video data has become a key issue, and video encryption technology has thus attracted extensive attention in the academic field and the industrial community. However, different from traditional videos, RGB-D videos consist of two core data parts, RGB video data and depth video data. To accurately calculate distances, depth data needs to be stored in a single channel (usually 16 bits) with a larger storage space, while RGB data is still stored in three channels (8 bits / channel, 24 bits in total). The difference in data formats between depth video data and RGB video data makes traditional video encryption algorithms unable to be directly applied to RGB-D videos.

[0003] Due to the characteristics of chaotic systems such as sensitivity to initial conditions, pseudo-randomness, and unpredictability, they have become powerful tools in the field of encryption. Chaotic systems are usually divided into one-dimensional (1D) and high-dimensional (HD) systems. Among them, one-dimensional chaotic systems are characterized by simple structures, low requirements for control parameters, and easy implementation. However, they have disadvantages such as limited key spaces and narrow chaos ranges, making video / image encryption schemes relying on one-dimensional chaotic maps more vulnerable to attacks and cryptanalysis. High-dimensional chaotic systems, especially hyperchaotic maps, exhibit more state variables, more complex structures, wider chaos ranges, and complex chaotic behaviors. However, the high computational complexity of high-dimensional systems has restricted their applications. Therefore, to solve the above problems, two-dimensional (2D) hyperchaotic maps with simple structures have attracted great attention. Summary of the Invention

[0004] The main purpose of the present invention is to overcome the disadvantages and deficiencies of the prior art and propose an RGB-D video encryption and decryption method based on hyperchaos technology and octonion fractional-order Hartley transform.

[0005] To achieve the above purpose, the present invention adopts the following technical solutions:

[0006] An RGB-D video encryption method based on hyperchaos technology and octonion fractional-order Hartley transform includes the following steps:

[0007] S1. Construct a two-dimensional hyperchaotic system ICCHM based on the ICMIC chaotic map and the Chebyshev chaotic map, and use it to generate the random matrix and random sequence required for encryption;

[0008] S2. Define the octonion fractional-order Hartley transform and its inverse transform;

[0009] S3. Encryption processing of RGB-D video.

[0010] Furthermore, step S1 specifically includes:

[0011] S11. Construct a new two-dimensional hyperchaotic map ICCHM by cascading the ICMIC map and the Chebyshev map:

[0012] x(t) = sin(ω / y(t - 1)×cos(q×arccos(x(t - 1))))

[0013] y(t) = sin(ω / x(t)×cos(q×arccos(y(t - 1))))

[0014] where ω and q are control parameters, ω ∈ (0, +∞), q ∈ (0, +∞);

[0015] S12. Using ω1 and q1 as control parameters, x1 and y1 as initial variables, generate two one-dimensional chaotic sequences SX1 and SY1 of length M×N using the ICCHM map, where M and N are the height and width of the video frame respectively, ω1 ∈ (0, +∞), q1 ∈ (0, +∞), and x1 and y1 are both random numbers with values in (-1, 1); then obtain a one-dimensional random chaotic sequence CS1 with element values in [0, 1] according to the formula SXY1 = (|SX1| + |SY1|) / 2, expand CS1 into a two-dimensional matrix of size M×N, and finally quantize all element values of this two-dimensional matrix to integers between [0, 255] to obtain a two-dimensional random integer matrix RM;

[0016] S13. Using ω2 and q2 as control parameters, x2 and y2 as initial variables, generate two chaotic sequences SX2 and SY2 of length M×N using the ICCHM map, ω2 ∈ (0, +∞), q2 ∈ (0, +∞), and x2 and y2 are both random numbers with values in (-1, 1); then apply the formula SXY2 = (SX2 + SY2) / 2 to obtain a random chaotic sequence CS2 with element values in [-1, 1]; sort CS2 to get [SCS2, Ind] = sort(CS2), where SCS2 represents the sorted sequence and Ind is the index value of SCS2, and sort() is the sorting operation.

[0017] Furthermore, step S2 specifically includes:

[0018] S21. Define the octonion fractional - order Hartley transform:

[0019]

[0020] where \(p_1\), \(p_2\) are the transform orders; cas = cos + sin; \(i\), \(j\), \(k\), \(l\), \(m\), \(n\), \(o\) are 7 imaginary units and satisfy \(i\ 2 =j 2 =k 2 =l 2 =m 2 =n 2 =o 2 =-1, \(\xi=\alpha i+\beta j+\gamma k+\mu i+\eta j+\lambda k+\rho k\), \(\xi\) is a unit pure octonion, satisfying the constraint \(\xi\ 2 =-1, \(\alpha\), \(\beta\), \(\gamma\), \(\mu\), \(\eta\), \(\lambda\) and \(\rho\) are all real numbers; \(h(x,y)=h_0(x,y)+h_1(x,y)i + h_2(x,y)j+h_3(x,y)k+h_4(x,y)l+h_5(x,y)m+h_6(x,y)n+h_7(x,y)o\) is an octonion, and \(h_0(x,y)\), \(h_1(x,y)\), \(h_2(x,y)\), \(h_3(x,y)\), \(h_4(x,y)\), \(h_5(x,y)\), \(h_6(x,y)\), \(h_7(x,y)\) are the 8 components of \(h(x,y)\); \(H(u,v)\) is the octonion fractional - order Hartley transform spectrum of \(h(x,y)\), and \(H_0(u,v)\), \(H_1(u,v)\), \(H_2(u,v)\), \(H_3(u,v)\), \(H_4(u,v)\), \(H_5(u,v)\), \(H_6(u,v)\), \(H_7(u,v)\) are the 8 components of \(H(u,v)\); OFrHT p1,p2 () represents the octonion fractional - order Hartley transform with orders \(p_1\), \(p_2\);

[0021] S22. By taking the transform orders as \((-p_1,-p_2)\), obtain the inverse transform of the octonion fractional - order Hartley transform with orders \((p_1,p_2)\):

[0022] h(x,y)=IOFrHT p1,p2 [H(u,v)] = OFrHT -p1,-p2 [H(u,v)]

[0023] =h_0(x,y)+h_1(x,y)+h_2(x,y)+h_3(x,y)+h_4(x,y)+h_5(x,y)+h_6(x,y)+h_7(x,y)

[0024] where \(IOFrHT\ p1,p2 () represents the inverse octonion fractional - order Hartley transform with orders \(p_1\), \(p_2\).

[0025] Further, step S3 includes:

[0026] S31. Sequentially read the (2t - 1)-th and 2t-th video frames from the RGB information video, denoted as rgb1 and rgb2 respectively, and then read the t-th depth video frame from the depth video DV, denoted as df, where t = 1, 2,......, L / 2, and the total number of frames of the RGB video and the depth video DV is both L;

[0027] S32. Since the depth video frame df is a 16-bit single channel, while the RGB video frames rgb1 and rgb2 are both 24-bit three channels, preprocess them as follows:

[0028] Convert df into two 8-bit single channels, denoted as dh and dl respectively, where dh is the high 8-bit channel and dl is the low 8-bit channel; then decompose the color video frames rgb1 and rgb2 into three 8-bit single channels, and denote the red, green, and blue 3-channel components of rgb1 as r1, g1, and b1, and the red, green, and blue 3-channel components of rgb2 as r2, g2, and b2; after processing, dh, dl, r1, g1, b1, r2, g2, and b2 are all two-dimensional 8-bit single channels;

[0029] S33. Perform exclusive OR operations on dh, dl, r1, g1, b1, r2, g2, and b2 respectively with the two-dimensional random integer matrix RM calculated in step S12, and perform normalization processing on all exclusive OR results to obtain h0, h1, h2, h3, h4, h5, h6, and h7;

[0030] S34. Use h0, h1, h2, h3, h4, h5, h6, and h7 calculated in step S33 as the 8 components of the octonion to represent a depth video frame and two RGB video frames in an overall manner with the octonion h: h = h0 + h1i + h2j + h3k + h4l + h5m + h6n + h7o;

[0031] S35. Perform two-dimensional octonion fractional-order Hartley transform on h with transformation orders p1 and p2 to obtain H = OFrHT p1,p2 (h) = H0 + H1i + H2j + H3k + H4l + H5m + H6n + H7o;

[0032] S36. Perform the inverse two-dimensional octonion fractional-order Hartley transform on H, that is, perform the fractional-order Hartley transform with transformation orders -p1 and -p2, and perform scrambling operation based on ICCHM on the intermediate calculation results obtained during the inverse transform calculation process.

[0033] Further, step S36 specifically includes:

[0034] S361. Perform inverse fractional-order Hartley transforms of order p1 and p2 on H0, H1, H2, H3, H4, H5, H6, and H7 respectively, that is, perform fractional-order Hartley transforms of order -p1 and -p2 to obtain eight complex matrices IH0, IH1, IH2, IH3, IH4, IH5, IH6, and IH7;

[0035] S362. First, reduce all of IH0, IH1, IH2, IH3, IH4, IH5, IH6, and IH7 from two dimensions to one dimension to obtain eight one-dimensional sequences: RIH0, RIH1, RIH2, RIH3, RIH4, RIH5, RIH6, and RIH7;

[0036] Then use Ind calculated in step S13 and operation PRIH s = RIH s (Ind) to scramble the eight one-dimensional sequences, where s = 0, 1,......, 7;

[0037] Ascend the eight scrambled sequences from one dimension to two dimensions to obtain eight corresponding scrambled matrices: PRIH0, PRIH1, PRIH2, PRIH3, PRIH4, PRIH5, PRIH6, and PRIH7;

[0038] S363. Apply formula (1) to the scrambled matrices PRIH0, PRIH1, PRIH2, PRIH3, PRIH4, PRIH5, PRIH6, and PRIH7 to calculate the reconstructed scrambled octonion h' after the inverse octonion fractional-order Hartley transform, h' = h'0 + h'1i + h'2j + h'3k + h'4l + h'5m + h'6n + h'7o, and formula (1) is:

[0039]

[0040] where real(x) and imag(x) are operations to return the real part and imaginary part of x respectively, and h'0, h'1, h'2, h'3, h'4, h'5, h'6, h'7 are the eight components of h'.

[0041] Further, after step S36, it also includes:

[0042] S37. Reconstruct the encrypted 16-bit depth video frame df' using h'0 and h'1 as the high 8-bit channel and the low 8-bit channel respectively; Recombine h'2, h'3 and h'4 as the red, green and blue channels respectively to obtain the encrypted information video frame rgb1'; Recombine h'5, h'6 and h'7 as the red, green and blue channels respectively to obtain the encrypted information video frame rgb2'.

[0043] S38. Repeat steps S31 to S37 to encrypt all RGB information video frames and L / 2 depth video frames, combine all the encrypted information video frames to obtain the encrypted RGB video RGB', and combine all the encrypted depth video frames to obtain the encrypted depth video DV'. Among them, the video RGB' contains L encrypted information video frames, and DV' contains L / 2 encrypted depth video frames.

[0044] S39. Encrypt the remaining L / 2 depth video frames, specifically including:

[0045] S391. Read 4 consecutive remaining depth video frames in sequence, and decompose them all into the high 8-bit channel and the low 8-bit channel to obtain 8 two-dimensional 8-bit single channels: dh1, dl1, dh2, dl2, dh3, dl3, dh4, dl4.

[0046] S392. Use the methods of steps S33 to S37 to encrypt the 4 depth video frames, and append these 4 encrypted depth video frames to the encrypted depth video DV' in the original order.

[0047] S393. Repeat steps S391 to S392 to encrypt all the remaining depth video frames and append them to the encrypted depth video DV'.

[0048] RGB' and DV' are the final encrypted RGB information video and encrypted depth video, and ω1, q1, x1, y1, ω2, q2, x2, y2, p1 and p2 are the keys of the RGB-D video encryption method.

[0049] The present invention also includes an RGB-D video decryption method for decrypting the video encrypted by the RGB-D video encryption method provided by the present invention, including the following steps:

[0050] S41. Read the (2t - 1)-th and 2t-th encrypted video frames from the encrypted information video RGB' in sequence, denoted as rgb1' and rgb2' respectively. Then decompose them into red, green, and blue channels and perform normalization to obtain r1', g1', b1', r2', g2', and b2'. Next, read the t-th encrypted depth video frame from the encrypted depth video DV', denoted as df', and convert it into two 8-bit channels and perform normalization, denoted as dh' and dl' respectively, where dh' is the high 8-bit channel and dl' is the low 8-bit channel.

[0051] S42. First, represent an encrypted depth video frame and two encrypted RGB video frames obtained through the processing in step S41 as a whole using octonions: h’ = dh' + dl'i + r1'j + g1'k + b1'l + r2'm + g2'n + b2'o. Perform a two-dimensional octonion fractional-order Hartley transform on h’ using the keys p1 and p2 to obtain H’ = OFrHT p1,p2 (h’) = H’0 + H’1i + H’2j + H’3k + H’4l + H’5m + H’6n + H’7o;

[0052] S43. Perform an inverse two-dimensional octonion fractional-order Hartley transform on H’ with transform orders p1 and p2, and perform inverse scrambling decryption based on ICCHM on the intermediate calculation results using the decryption keys ω2, q2, x2, and y2.

[0053] Further, step S43 specifically includes:

[0054] S431. Perform inverse fractional-order Hartley transforms with transform orders p1 and p2 on H’0, H’1, H’2, H’3, H’4, H’5, H’6, and H’7 respectively to obtain IH’0, IH’1, IH’2, IH’3, IH’4, IH’5, IH’6, and IH’7;

[0055] S432. Use ω2, q2, x2, and y2 as decryption keys and calculate Ind using the method in step S13.

[0056] Then reduce all of IH’0, IH’1, IH’2, IH’3, IH’4, IH’5, IH’6, and IH’7 from two-dimensional to one-dimensional to obtain RIH’0, RIH’1, RIH’2, RIH’3, RIH’4, RIH’5, RIH’6, and RIH’7;

[0057] Use the operation DPRIH’ s (Ind) = RIH’ s Perform inverse scrambling on these 8 one-dimensional sequences, where s = 0, 1,......, 7;

[0058] All of DPRIH’0, DPRIH’1, DPRIH’2, DPRIH’3, DPRIH’4, DPRIH’5, DPRIH’6, and DPRIH’7 obtained by inverse scrambling are raised from one - dimensional to two - dimensional to obtain IPIH'0, IPIH'1, IPIH'2, IPIH'3, IPIH'4, IPIH'5, IPIH'6, and IPIH'7;

[0059] S433. First, apply formula (1) to IPIH'0, IPIH'1, IPIH'2, IPIH'3, IPIH'4, IPIH'5, IPIH'6, and IPIH'7 to calculate the reconstructed octonion h” after inverse scrambling, h” = h”0 + h”1i + h”2j + h”3k + h”4l + h”5m + h”6n + h”7o; where h”0, h”1, h”2, h”3, h”4, h”5, h”6, h”7 are the 8 components of h”; map the values of the elements of these 8 reconstructed components to integers between [0, 255].

[0060] Furthermore, after step S43, it further includes:

[0061] S44. Using ω1, q1, x1, and y1 as the decryption keys, calculate the random integer matrix RM using the method of step S12, and perform XOR operations with h”0, h”1, h”2, h”3, h”4, h”5, h”6, h”7 respectively to obtain dh0, dh1, dh2, dh3, dh4, dh5, dh6, dh7;

[0062] S45. Recover the decrypted 16 - bit depth video frame df using dh0 and dh1 as the high 8 - bit channel and low 8 - bit channel; recombine dh2, dh3, and dh4 as the red, green, and blue channels to obtain the decrypted RGB information video frame rgb1; recombine dh5, dh6, and dh7 as the red, green, and blue channels to obtain the decrypted RGB information video frame rgb2;

[0063] S46. Repeat steps S41 to S45 to obtain all the decrypted information video frames and depth video frames, and then recombine the obtained information video frames and depth video frames respectively to obtain the decrypted RGB information video dRGB and the decrypted depth video dDV; at this time, the total number of frames of the decrypted RGB information video is L, and the total number of frames of the decrypted depth video is L / 2;

[0064] S47. Decrypt the remaining encrypted depth video frames.

[0065] Furthermore, step S47 specifically includes:

[0066] S471. Read the remaining 4 encrypted depth video frames in sequence, decompose them all into high 8-bit channels and low 8-bit channels to obtain 8 two-dimensional 8-bit single channels: dh1’, dl1’, dh2’, dl2’, dh3’, dl3’, dh4’, dl4’, and then perform normalization processing on them;

[0067] S472. Use the method of steps S42 to S45 to decrypt these 4 encrypted depth video frames, and append these 4 decrypted depth video frames to the decrypted depth video dDV in the original order;

[0068] S473. Repeat steps S471 to S472 to decrypt all the remaining encrypted depth video frames and append them to the decrypted depth video dDV;

[0069] dRGB and dDV are the final decrypted RGB information video and decrypted depth video.

[0070] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0071] 1. The present invention uses the octonion fractional-order Hartley transform to process two RGB information video frames and one depth video frame as a vector whole, realizing the simultaneous processing of two types of different data, effectively improving the encryption efficiency of RGB-D videos, effectively avoiding crosstalk noise during decryption, improving the image quality of the decrypted video, and further improving the security of the method with the transformation order as the key; moreover, the present invention uses the designed hyperchaotic system to encrypt and decrypt RGB-D videos, expanding the key space and enhancing pseudo-randomness, making the RGB-D video encryption method of the present invention have extremely high security and effectively resisting common cryptographic attacks. Description of the Drawings

[0072] Figure 1 is the flowchart of the encryption process of the present invention;

[0073] Figure 2 is the flowchart of the decryption process of the present invention;

[0074] Figure 3 is the bifurcation diagram of the hyperchaotic system ICCHM in the embodiment when q = 10, ω ∈ (0, 15), x(1) = 0.3 and y(1) = 0.6;

[0075] Figure 4 is the bifurcation diagram of the hyperchaotic system ICCHM in the embodiment when ω = 6, q ∈ (0, 15), x(1) = 0.3 and y(1) = 0.6;

[0076] Figure 5is the Lyapunov exponent spectrum of the hyperchaotic system ICCHM in the embodiment;

[0077] Figure 6 is the plaintext RGB-D video in the embodiment;

[0078] Figure 7 is the encrypted RGB-D video in the embodiment;

[0079] Figure 8 is the decryption result when all keys in the embodiment are correct;

[0080] Figure 9 is the key ω1 = ω1 + 10 in the embodiment -15 when the decrypted RGB-D video;

[0081] Figure 10 is the key q1 = q1 + 10 in the embodiment -15 when the decrypted RGB-D video;

[0082] Figure 11 is the key x1 = x1 + 10 in the embodiment -16 when the decrypted RGB-D video;

[0083] Figure 12 is the key y1 = y1 + 10 in the embodiment -16 when the decrypted RGB-D video;

[0084] Figure 13 is the key ω2 = ω2 + 10 in the embodiment -14 when the decrypted RGB-D video;

[0085] Figure 14 is the key q2 = q2 + 10 in the embodiment -15 when the decrypted RGB-D video;

[0086] Figure 15 is the key x2 = x2 + 10 in the embodiment -16 when the decrypted RGB-D video;

[0087] Figure 16 is the key y2 = y2 + 10 in the embodiment -16 when the decrypted RGB-D video;

[0088] Figure 17 is the key p1 = p1 + 5 × 10 in the embodiment -4 when the decrypted RGB-D video;

[0089] Figure 18 is the key p2 = p2 + 5 × 10 in the embodiment -4 when the decrypted RGB-D video. Detailed implementation manners

[0090] The present invention will be further described in detail below in conjunction with embodiments and the accompanying drawings, but the implementation manners of the present invention are not limited thereto.

[0091] Embodiment

[0092] As Figure 1 shown, an RGB-D video encryption method based on hyperchaos technology and octonion fractional Hartley transform includes:

[0093] I. Construct a two-dimensional hyperchaos system ICCHM based on ICMIC chaos mapping and Chebyshev chaos mapping, and apply it to generate a random matrix and a random sequence required for encryption; in this embodiment, specifically:

[0094] S1A. First, obtain the number of rows M = 480 and the number of columns N = 640 from an information video frame or a depth video frame, then randomly generate x1 = 0.280339004407621 and y1 = 0.652629816750457, and then use x1 and y1 as the initial parameters of the designed ICCHM mapping, with ω1 = 2 and q1 = 3 as the control parameters, and use step S12 to calculate a random integer matrix RM uniformly distributed in the interval [0, 255], with a size of 480×640;

[0095] S1B. Randomly generate x2 = 0.585367013307612 and y2 = 0.976282376516561, and then use x2 and y2 as the initial parameters of the ICCHM mapping, with ω2 = 10 and q2 = 20 as the control parameters, and use step S13 to calculate an index sequence Ind after sorting a random sequence, with a length of 307200.

[0096] II. Define the octonion fractional Hartley transform and its inverse transform;

[0097] III. RGB-D video encryption processing, in this embodiment, includes:

[0098] S3A. Sequentially read the (2t - 1)-th and 2t-th video frames from the RGB information video cabinet, denoted as rgb1 and rgb2 respectively, and then read the t-th depth video frame from the depth video depth_cabinet, denoted as df, where t = 1, 2,......, L / 2, and the total number of video frames of the cabinet video and the depth video depth_cabinet is both L = 952 frames.

[0099] S3B. Perform a preprocessing operation of converting and channel decomposing df, rgb1, and rgb2 according to step S32 to obtain 8 8-bit single channels: dh, dl, r1, g1, b1, r2, g2, and b2.

[0100] S3C. Perform exclusive OR operations on dh, dl, r1, g1, b1, r2, g2, and b2 respectively with the matrix RM calculated in step S1A, and perform normalization processing on all exclusive OR results to obtain h0, h1, h2, h3, h4, h5, h6, and h7.

[0101] S3D. Use h0, h1, h2, h3, h4, h5, h6, and h7 as the 8 components of the octonion to represent a depth video frame and two RGB video frames in an overall manner with the octonion h: h = h0 + h1i + h2j + h3k + h4l + h5m + h6n + h7o.

[0102] S3E. Take the transformation orders p1 = 0.912 and p2 = 0.876, and perform the octonion fractional-order Hartley transform on h to obtain H = OFrHT p1,p2 (h) = H0 + H1i + H2j + H3k + H4l + H5m + H6n + H7o.

[0103] S3F. Obtain the reconstructed scrambled octonion h' by applying the inverse two-dimensional octonion fractional-order Hartley transform with transformation orders p1 = 0.912 and p2 = 0.876 and the scrambling operation based on ICCHM to H through the following steps:

[0104] (1) Perform the inverse fractional-order Hartley transform with transformation orders p1 = 0.912 and p2 = 0.876 on H0, H1, H2, H3, H4, H5, H6, and H7 respectively to obtain 8 complex matrices IH0, IH1, IH2, IH3, IH4, IH5, IH6, and IH7;

[0105] (2) Reduce the dimensions of all of IH0, IH1, IH2, IH3, IH4, IH5, IH6, and IH7 to obtain 8 one-dimensional sequences: RIH0, RIH1, RIH2, RIH3, RIH4, RIH5, RIH6, and RIH7; then use the scrambling operation in step S362 and Ind calculated in step S1B to scramble the above 8 one-dimensional sequences, and then increase the dimensions of these 8 scrambled sequences to obtain the corresponding scrambled matrices PRIH0, PRIH1, PRIH2, PRIH3, PRIH4, PRIH5, PRIH6, and PRIH7;

[0106] (3) Take α = β = γ = μ = η = λ = ρ = (1 / 7) 1 / 2, apply formula (1) to PRIH0, PRIH1, PRIH2, PRIH3, PRIH4, PRIH5, PRIH6, and PRIH7 to calculate the reconstructed scrambled octonion h' = h'0 + h'1i + h'2j + h'3k + h'4l + h'5m + h'6n + h'7o after the inverse octonion fractional Fourier transform:

[0107] S3G. Use h'0 and h'1 as the high 8-bit channel and low 8-bit channel to reconstruct the encrypted 16-bit depth video frame df'; combine h'2, h'3, and h'4 to obtain the encrypted information video frame rgb1'; combine h'5, h'6, and h'7 to obtain the encrypted information video frame rgb2'.

[0108] S3H. Repeat steps S3A to S3G to encrypt all the information video frames and 476 depth video frames, and combine all the encrypted information video frames to obtain the encrypted RGB video RGB', and combine all the encrypted depth video frames to obtain the encrypted depth video DV'. Among them, the video RGB' contains 952 encrypted information video frames, and DV' contains 476 encrypted depth video frames.

[0109] S3I. Encrypt the remaining 476 depth video frames using a process similar to the above, specifically including the following steps:

[0110] (1) Read 4 consecutive remaining depth video frames in sequence and decompose them all into high 8-bit channels and low 8-bit channels to obtain 8 8-bit single channels: dh1, dl1, dh2, dl2, dh3, dl3, dh4, dl4;

[0111] (2) Use the methods of steps S3C to S3G to encrypt these 4 depth video frames, and append these 4 encrypted depth video frames to the encrypted depth video DV' in the original order;

[0112] (3) Repeat steps (1) to (2) in S3I to encrypt all the remaining depth video frames and append them to the encrypted depth video DV'.

[0113] RGB' and DV' are the final encrypted RGB information video and encrypted depth video.

[0114] In another embodiment, a method for decrypting an RGB-D video is also provided, which is used to decrypt the video encrypted by the RGB-D video encryption method of the above embodiment, as Figure 2 shown, specifically including:

[0115] S4A. Read the (2t - 1)-th and 2t-th encrypted video frames from the encrypted information video RGB' in sequence, and denote them as rgb1' and rgb2' respectively. Then decompose both of them into three RGB channels and perform normalization to obtain r1', g1', b1', r2', g2', and b2'. Next, read the t-th encrypted depth video frame from the encrypted depth video DV' and denote it as df'. Then convert it into two 8-bit channels and perform normalization, denoted as dh' and dl' respectively, where dh' is the high 8-bit channel and dl' is the low 8-bit channel.

[0116] S4B. Represent the obtained 8 8-bit channels in an overall manner with octonions: h’ = dh' + dl'i + r1'j + g1'k + b1'l + r2'm + g2'n + b2'o. Then perform a two-dimensional octonion fractional-order Hartley transform on h’ using the keys p1 = 0.912 and p2 = 0.876 to obtain H’ = OFrHT p1,p2 (h’) = H’0 + H’1i + H’2j + H’3k + H’4l + H’5m + H’6n + H’7o.

[0117] S4C. Apply the inverse two-dimensional octonion fractional-order Hartley transform with transformation orders p1 and p2 to H’ through the following steps, and perform inverse scrambling decryption based on ICCHM on the intermediate calculation results using the decryption keys ω2, q2, x2, and y2:

[0118] (1) Perform inverse fractional-order Hartley transforms with transformation orders p1 = 0.912 and p2 = 0.876 on H’0, H’1, H’2, H’3, H’4, H’5, H’6, and H’7 respectively to obtain I H’0, I H’1, I H’2, I H’3, I H’4, I H’5, I H’6, and I H’7;

[0119] (2) Use ω2 = 10, q2 = 20, x2 = 0.585367013307612, and y2 = 0.976282376516561 as the decryption keys, and calculate Ind using the method in step S13. Then reduce the dimensions of I H’0, I H’1, I H’2, I H’3, I H’4, I H’5, I H’6, and I H’7 to obtain 8 one-dimensional sequences: R I H’0, R I H’1, R I H’2, R I H’3, R I H’4, R I H’5, R I H’6, and R I H’7. Then perform inverse scrambling on the above 8 one-dimensional sequences using the inverse scrambling operation and Ind in step S432, and then increase the dimensions of the obtained 8 inverse-scrambled sequences to obtain the corresponding inverse-scrambled matrices I P I H'0, I P I H'1, I P I H'2, I P I H'3, I P I H'4, I P I H'5, I P I H'6, and I P I H'7;

[0120] (3) Apply formula (1) to IPIH'0, IPIH'1, IPIH'2, IPIH'3, IPIH'4, IPIH'5, IPIH'6, and IPIH'7 to calculate the reconstructed octonion h” after inverse scrambling: h” = h”0 + h”1i + h”2j + h”3k + h”4l + h”5m + h”6n + h”7o, where h”0, h”1, h”2, h”3, h”4, h”5, h”6, and h”7 are the 8 components of h”; then map the values of the elements of these 8 reconstructed components to integers between [0, 255].

[0121] S4D. Using ω1 = 2, q1 = 3, x1 = 0.280339004407621, and y1 = 0.652629816750457 as the decryption key, calculate the random integer matrix RM using the method in step S12, and perform XOR operations on it with h”0, h”1, h”2, h”3, h”4, h”5, h”6, and h”7 respectively to obtain dh0, dh1, dh2, dh3, dh4, dh5, dh6, and dh7.

[0122] S4E. Using dh0 and dh1 as the high 8-bit channel and low 8-bit channel respectively, recover the decrypted 16-bit depth video frame df; combine dh2, dh3, and dh4 to obtain the decrypted RGB information video frame rgb1; combine dh5, dh6, and dh7 to obtain the decrypted RGB information video frame rgb2.

[0123] S4F. First, repeat steps S4A to S4E to obtain all the decrypted information video frames and depth video frames, and then recombine the obtained information video frames and depth video frames respectively to obtain the decrypted RGB information video dRGB and the decrypted depth video dDV. At this time, the total number of frames of the decrypted RGB information video is 952 frames, and the total number of frames of the decrypted depth video is 476 frames.

[0124] S4G. Decrypt the remaining encrypted depth video frames using a process similar to the above, which specifically includes the following steps:

[0125] (1) Read 4 remaining encrypted depth video frames in sequence, and decompose them all into a high 8-bit channel and a low 8-bit channel to obtain 8 two-dimensional 8-bit single channels: dh1’, dl1’, dh2’, dl2’, dh3’, dl3’, dh4’, and dl4’, and then perform normalization processing on them;

[0126] (2) Use the methods in steps S4B to S4E to decrypt these 4 encrypted depth video frames, and append these 4 decrypted depth video frames to the decrypted depth video dDV in the original order;

[0127] (3) Repeat steps (1) to (2) in step S4G to decrypt all the remaining encrypted depth video frames and append them to the decrypted depth video dDV. dRGB and dDV are the final decrypted RGB information video and decrypted depth video.

[0128] As Figure 3 and Figure 4 shown, Figure 3 and Figure 4 are the bifurcation diagrams of the hyperchaotic system ICCHM when q = 10, ω ∈ (0, 15), x(1) = 0.3 and y(1) = 0.6, and when ω = 6, q ∈ (0, 15), x(1) = 0.3 and y(1) = 0.6, respectively; where (a) is the bifurcation diagram of the variable x varying with ω, and (b) is the bifurcation diagram of the variable y varying with ω.

[0129] As Figure 5 shown, is the Lyapunov exponent spectrum of the hyperchaotic system ICCHM when q = 10, ω ∈ (0, 15), x(1) = 0.3 and y(1) = 0.6, and when ω = 6, q ∈ (0, 15), x(1) = 0.3 and y(1) = 0.6.

[0130] The Lyapunov exponent is an important quantitative index to measure the dynamic characteristics of a system, which characterizes the average exponential rate of convergence or divergence between adjacent orbits in the phase space of the system. A positive Lyapunov exponent indicates that in the phase space of the system, no matter how small the distance between the initial two trajectories is, their difference will increase exponentially with the evolution of time until an unpredictable chaotic phenomenon is reached. A negative Lyapunov exponent indicates that the system is insensitive to the initial conditions and is stable. The hyperchaotic system has at least two positive Lyapunov exponents and has more complex dynamic behaviors compared with ordinary chaotic systems. This means that the image cryptosystem based on the hyperchaotic system has excellent performance. When fixing q = 10, the bifurcation diagrams of the variables x and y of the ICCHM system varying with ω ( Figure 3 ) and the exponential spectrum diagram of the Lyapunov exponent varying with ω ( Figure 5 (a)) are obtained; when fixing ω = 6, the bifurcation diagrams of the variables x and y varying with q ( Figure 4 ) and the exponential spectrum diagram of the Lyapunov exponent varying with ω ( Figure 5 (b)) are obtained.

[0131] As shown in the Lyapunov exponent spectrum diagram and the system bifurcation diagram, when the system is at q = 10 and ω ∈ (0.1, 0.12) ∪ (0.36, 0.41) ∪ (0.77, 0.79) ∪ (1.20, 1.22) ∪ (1.57, 1.58), both Lyapunov exponents are less than 0, indicating that the system is in a non-chaotic state; when ω ∈ (1.8, 1.81) ∪ (4.19, 4.20), one Lyapunov exponent is positive and the other is negative, indicating that the system has transformed into a chaotic state; in addition to the above cases, both Lyapunov exponents are positive, and the system exhibits hyperchaotic behavior. When the system is at ω = 6 and q ∈ (0, 0.01), both Lyapunov exponents are less than 0, indicating that the system is in a non-chaotic state; in addition to this case, both Lyapunov exponents are positive, and the system exhibits hyperchaotic behavior. From Figure 3 and Figure 4 it can be seen that the ICCHM system has a large and continuous parameter range, and the iterative sequences are almost randomly distributed throughout the space.

[0132] In this embodiment, an RGB-D video with a video frame size of 480×640 is selected for encryption. In this RGB-D video, both the RGB information video and the depth video contain 952 video frames, as Figure 6 shown. Figure 6 In, (a) and (b) are respectively the first frame and the second frame of the RGB information video cabinet, (c) is the first frame of the depth video depth_cabinet (the depth video frame corresponding to the first frame of cabinet), and (d) is the 3D visualization image (azimuth angle is -45 degrees, elevation angle is 70 degrees) obtained by converting the information video frame (a) and the depth video frame (c) into point cloud data. As Figure 7 shown, it is the encrypted RGB-D video, where (a) and (b) are respectively the first frame and the second frame of the encrypted cabinet video, and (c) is the first frame of the encrypted depth_cabinet video. From Figure 7 it can be seen that the information of the plaintext video is hidden in the noise-like video frames, indicating that the encryption of the RGB-D video using the encryption method of the present invention is successful.

[0133] Using the decryption method of the present invention, the plaintext RGB-D video is restored from the encrypted video. When all the keys are correct, the results are as Figure 8 shown, Figure 8 in, (a) and (b) are respectively the first frame and the second frame of the decrypted cabinet video; (c) is the first frame of the decrypted depth_cabinet video; (d) is the 3D visualization image obtained by converting the decrypted video frame (a) and the decrypted depth video frame (c) into point cloud data; fromFigure 8 (a)-8(d) It can be seen that when all keys are correct, the plaintext RGB-D video can be completely restored, indicating that the encryption and decryption of the RGB-D video using the present invention are successful.

[0134] In addition, when one key is incorrect and the other keys are correct, the decryption result is as Figures 9 to 18 shown Figures 9 to 18 in which, (a) and (b) are the first and second frames of the decrypted cabinet video respectively; (c) is the first frame of the decrypted depth_cabinet video; from Figures 9 to 18 the decryption results shown, it can be seen that even when only one key undergoes a very small change and the other keys are all correct, the resulting decrypted RGB-D video is incorrect and cannot provide any valid information. Thus, the security of the encryption method of the present invention can be guaranteed.

[0135] It should also be noted that in this specification, terms such as "including", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0136] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. RGB-D video encryption method based on hyperchaotic technology and octonion fractional Hartley transform, characterized in that, It includes the following steps: S1. Construct a two-dimensional hyperchaotic system ICCHM based on the ICMIC chaotic map and the Chebyshev chaotic map, and apply it to generate the random matrix and random sequence required for encryption; S2. Define the octonion fractional-order Hartley transform and its inverse transform; S3. Perform RGB-D video encryption processing.

2. The RGB-D video encryption method according to claim 1, wherein Step S1 specifically includes: S11. Construct a new two-dimensional hyperchaotic map ICCHM by cascading the ICMIC map and the Chebyshev map: x(t) = sin(ω / y(t - 1)×cos(q×arccos(x(t - 1)))) y(t) = sin(ω / x(t)×cos(q×arccos(y(t - 1)))) where ω and q are control parameters, ω ∈ (0, +∞), q ∈ (0, +∞); S12. Using ω1 and q1 as control parameters and x1 and y1 as initial variables, generate two one-dimensional chaotic sequences SX1 and SY1 of length M×N using the ICCHM map, where M and N are the height and width of the video frame respectively, ω1 ∈ (0, +∞), q1 ∈ (0, +∞), and x1 and y1 are both random numbers with values in (-1, 1); then obtain a one-dimensional random chaotic sequence CS1 with element values between [0, 1] according to the formula SXY1 = (|SX1| + |SY1|) / 2, up-dimension CS1 to a two-dimensional matrix of size M×N, and finally quantize all element values of this two-dimensional matrix to integers between [0, 255] to obtain a two-dimensional random integer matrix RM; S13. Using ω2 and q2 as control parameters and x2 and y2 as initial variables, generate two chaotic sequences SX2 and SY2 of length M×N using the ICCHM map, ω2 ∈ (0, +∞), q2 ∈ (0, +∞), and x2 and y2 are both random numbers with values in (-1, 1); then apply the formula SXY2 = (SX2 + SY2) / 2 to obtain a random chaotic sequence CS2 with element values between [-1, 1]; sort CS2 to get [SCS2, Ind] = sort(CS2), where SCS2 represents the sorted sequence and Ind is the index value of SCS2, and sort() is the sorting operation.

3. The RGB-D video encryption method according to claim 1, wherein, Step S2 specifically includes: S21. Define the octonion fractional-order Hartley transform: Among them, p1 and p2 are transformation orders. cas = cos + sin; i, j, k, l, m, n, o are 7 imaginary units and satisfy i 2 = j 2 = k 2 = l 2 = m 2 = n 2 = o 2 = -1, ξ = αi + βj + γk + μi + ηj + λk + ρk, ξ is a unit pure octonion and satisfies the constraint ξ 2 = -1, α, β, γ, μ, η, λ, and ρ are all real numbers; h(x, y) = h0(x, y) + h1(x, y)i + h2(x, y)j + h3(x, y)k + h4(x, y)l + h5(x, y)m + h6(x, y)n + h7(x, y)o is an octonion, and h0(x, y), h1(x, y), h2(x, y), h3(x, y), h4(x, y), h5(x, y), h6(x, y), h7(x, y) are the 8 components of h(x, y); H(u, v) is the octonion fractional-order Hartley transform spectrum of h(x, y), and H0(u, v), H1(u, v), H2(u, v), H3(u, v), H4(u, v), H5(u, v), H6(u, v), H7(u, v) are the 8 components of H(u, v); OFrHT p1,p2 () represents the octonion fractional-order Hartley transform with fractional orders p1 and p2. S22. By taking the transform order as (-p1, -p2), obtain the inverse transform of the octonion fractional-order Hartley transform with fractional order (p1, p2): h(x,y) = IOFrHT p1,p2 [H(u,v)] = OFrHT -p1,-p2 [H(u,v)] = h0(x, y) + h1(x, y) + h2(x, y) + h3(x, y) + h4(x, y) + h5(x, y) + h6(x, y) + h7(x, y) Among them, IOFrHT p1,p2 () represents the inverse octonion fractional-order Hartley transform with transformation orders p1 and p2.

4. The RGB-D video encryption method according to claim 1, characterized in that Step S3 includes: S31. Sequentially read the (2t - 1)-th and 2t-th video frames from the RGB information video, denoted as rgb1 and rgb2 respectively, and then read the t-th depth video frame from the depth video DV, denoted as df, where t = 1, 2,......, L / 2, and the total number of frames of the RGB video and the depth video DV is both L; S32. Since the depth video frame df is a 16-bit single channel, while the RGB video frames rgb1 and rgb2 are both 24-bit three channels, preprocessing is performed on them first. Specifically: Convert df into two 8-bit single channels, denoted as dh and dl respectively, where dh is the high 8-bit channel and dl is the low 8-bit channel; then decompose the color video frames rgb1 and rgb2 into three 8-bit single channels, and use r1, g1, and b1 to represent the red, green, and blue channel components of rgb1, and use r2, g2, and b2 to represent the red, green, and blue channel components of rgb2; after processing, dh, dl, r1, g1, b1, r2, g2, and b2 are all two-dimensional 8-bit single channels; S33. Perform exclusive OR operations on dh, dl, r1, g1, b1, r2, g2, and b2 respectively with the two-dimensional random integer matrix RM calculated in step S12, and perform normalization processing on all exclusive OR results to obtain h0, h1, h2, h3, h4, h5, h6, and h7; S34. Use h0, h1, h2, h3, h4, h5, h6, and h7 calculated in step S33 as the 8 components of the octonion, and represent a depth video frame and two RGB video frames in an overall manner with the octonion h: h = h0 + h1i + h2j + h3k + h4l + h5m + h6n + h7o; S35. Perform a two-dimensional octonion fractional-order Hartley transform on h with transformation orders p1 and p2 to obtain H = OFrHT p1,p2 (h) = H0 + H1i + H2j + H3k + H4l + H5m + H6n + H7o; S36. Perform the inverse two-dimensional octonion fractional-order Hartley transform on H with transform orders p1 and p2, that is, perform the fractional-order Hartley transform with transform orders -p1 and -p2, and perform scrambling operations based on ICCHM on the intermediate calculation results obtained during the inverse transform calculation process.

5. The RGB-D video encryption method according to claim 4, wherein Step S36 specifically includes: S361. Perform the inverse fractional-order Hartley transform with transform orders p1 and p2 on H0, H1, H2, H3, H4, H5, H6, and H7 respectively, that is, perform the fractional-order Hartley transform with transform orders -p1 and -p2, to obtain 8 complex matrices IH0, IH1, IH2, IH3, IH4, IH5, IH6, and IH7; S362. First, reduce IH0, IH1, IH2, IH3, IH4, IH5, IH6, and IH7 from two-dimensional to one-dimensional, to obtain 8 one-dimensional sequences: RIH0, RIH1, RIH2, RIH3, RIH4, RIH5, RIH6, and RIH7; Then use Ind and operation PRIH calculated in step S13 s = RIH s (Ind) scramble the 8 one-dimensional sequences, where s = 0, 1,......, 7; Raise the 8 scrambled sequences from one-dimensional to two-dimensional, to obtain 8 corresponding scrambled matrices: PRIH0, PRIH1, PRIH2, PRIH3, PRIH4, PRIH5, PRIH6, and PRIH7; S363. Apply formula (1) to the permutation matrices PRIH0, PRIH1, PRIH2, PRIH3, PRIH4, PRIH5, PRIH6, and PRIH7 to calculate the reconstructed permuted octonion h' after the inverse octonion fractional Hartley transform. h' = h'0 + h'1i + h'2j + h'3k + h'4l + h'5m + h'6n + h'7o, and formula (1) is as follows: where real(x) and imag(x) are operations that return the real and imaginary parts of x respectively, and h'0, h'1, h'2, h'3, h'4, h'5, h'6, h'7 are the 8 components of h'.

6. The RGB-D video encryption method according to claim 5, wherein After step S36, it further includes: S37. Use h'0 and h'1 as the high 8-bit channel and low 8-bit channel to reconstruct the encrypted 16-bit depth video frame df'; use h'2, h'3, and h'4 as the red, green, and blue channels respectively, and recombine them to obtain the encrypted information video frame rgb1'; use h'5, h'6, and h'7 as the red, green, and blue channels respectively, and recombine them to obtain the encrypted information video frame rgb2'; S38. Repeat steps S31 to S37 to encrypt all RGB information video frames and L / 2 depth video frames, combine all the encrypted information video frames to obtain the encrypted RGB video RGB', and combine all the encrypted depth video frames to obtain the encrypted depth video DV'. Among them, the video RGB' contains L encrypted information video frames, and DV' contains L / 2 encrypted depth video frames; S39. Encrypt the remaining L / 2 depth video frames, specifically including: S391. Read 4 consecutive remaining depth video frames in sequence, decompose them into high 8-bit channels and low 8-bit channels respectively, to obtain 8 two-dimensional 8-bit single channels: dh1, dl1, dh2, dl2, dh3, dl3, dh4, dl4; S392. Use the methods of steps S33 to S37 to encrypt the 4 depth video frames, and append these 4 encrypted depth video frames to the encrypted depth video DV' in the original order; S393. Repeat steps S391 to S392 to encrypt all the remaining depth video frames and append them to the encrypted depth video DV'; RGB' and DV' are the final encrypted RGB information video and encrypted depth video, and ω1, q1, x1, y1, ω2, q2, x2, y2, p1, and p2 are the keys of the RGB-D video encryption method.

7. An RGB-D video decryption method for decrypting a video encrypted by the RGB-D video encryption method according to any one of claims 1-6, characterized in that, It includes the following steps: S41. Read the (2t - 1)th and 2tth encrypted video frames from the encrypted information video RGB' in sequence, and represent them as rgb1' and rgb2' respectively; then decompose them into red, green, and blue channels and perform normalization processing to obtain r1', g1', b1', r2', g2', and b2'; then read the tth encrypted depth video frame from the encrypted depth video DV', represent it as df', and convert it into two 8-bit channels and perform normalization processing, represented as dh' and dl' respectively, where dh' is the high 8-bit channel and dl' is the low 8-bit channel; S42. First, represent an encrypted depth video frame and two encrypted RGB video frames obtained after the processing in step S41 as a whole with octonions: h’ = dh'+dl'i+r1'j+g1'k+b1'l+r2'm+g2'n+b2'o; perform a two-dimensional octonion fractional-order Hartley transform on h’ using keys p1 and p2 to obtain H’ = OFrHT p1,p2 (h’) = H’0+H’1i+H’2j+H’3k+H’4l+H’5m+H’6n+H’7o; S43. Perform the inverse two-dimensional octonion fractional-order Hartley transform of order p1 and p2 on H', and perform inverse scrambling decryption based on ICCHM on the intermediate calculation results using the decryption keys ω2, q2, x2, and y2.

8. The RGB-D video decryption method according to claim 7, wherein Step S43 specifically includes: S431. Perform the inverse fractional-order Hartley transform of order p1 and p2 on H'0, H'1, H'2, H'3, H'4, H'5, H'6, and H'7 respectively to obtain I'H'0, I'H'1, I'H'2, I'H'3, I'H'4, I'H'5, I'H'6, and I'H'7; S432. Use ω2, q2, x2, and y2 as decryption keys and calculate Ind using the method of step S13; Then reduce all of I'H'0, I'H'1, I'H'2, I'H'3, I'H'4, I'H'5, I'H'6, and I'H'7 from two dimensions to one dimension to obtain RI'H'0, RI'H'1, RI'H'2, RI'H'3, RI'H'4, RI'H'5, RI'H'6, and RI'H'7; Using operation DPRIH’ s (Ind) = RIH’ s Perform inverse scrambling on these 8 one-dimensional sequences, where s = 0, 1,......, 7; Ascend all of DPRI'H'0, DPRI'H'1, DPRI'H'2, DPRI'H'3, DPRI'H'4, DPRI'H'5, DPRI'H'6, and DPRI'H'7 obtained through inverse scrambling from one dimension to two dimensions to obtain IP'I'H'0, IP'I'H'1, IP'I'H'2, IP'I'H'3, IP'I'H'4, IP'I'H'5, IP'I'H'6, and IP'I'H'7; S433. First, apply formula (1) to IP'I'H'0, IP'I'H'1, IP'I'H'2, IP'I'H'3, IP'I'H'4, IP'I'H'5, IP'I'H'6, and IP'I'H'7 to calculate the reconstructed octonion h'' after inverse scrambling, h'' = h''0 + h''1i + h''2j + h''3k + h''4l + h''5m + h''6n + h''7o; where h''0, h''1, h''2, h''3, h''4, h''5, h''6, and h''7 are the 8 components of h''; map the values of the elements of these 8 reconstructed components to integers between [0, 255].

9. The RGB-D video decryption method according to claim 8, characterized in that, After step S43, it also includes: S44. Use ω1, q1, x1, and y1 as decryption keys, calculate the random integer matrix RM using the method of step S12, and perform XOR operations with h''0, h''1, h''2, h''3, h''4, h''5, h''6, and h''7 respectively to obtain dh0, dh1, dh2, dh3, dh4, dh5, dh6, and dh7; S45. Recover the decrypted 16-bit depth video frame df using dh0 and dh1 as the high 8-bit channel and the low 8-bit channel; recombine dh2, dh3, and dh4 as the red, green, and blue channels to obtain the decrypted RGB information video frame rgb1; recombine dh5, dh6, and dh7 as the red, green, and blue channels to obtain the decrypted RGB information video frame rgb2; S46. Repeat steps S41 to S45 to obtain all the decrypted information video frames and depth video frames, and then recombine the obtained information video frames and depth video frames respectively to obtain the decrypted RGB information video dRGB and the decrypted depth video dDV; at this time, the total number of frames of the decrypted RGB information video is L, and the total number of frames of the decrypted depth video is L / 2; S47. Decrypt the remaining encrypted depth video frames.

10. The RGB-D video decryption method according to claim 9, wherein, Step S47 specifically includes: S471. Read 4 remaining encrypted depth video frames in sequence, decompose them all into a high 8-bit channel and a low 8-bit channel to obtain 8 two-dimensional 8-bit single channels: dh1’, dl1’, dh2’, dl2’, dh3’, dl3’, dh4’, dl4’, and then perform normalization processing on them; S472. Use the methods of steps S42 to S45 to decrypt these 4 encrypted depth video frames, and append these 4 decrypted depth video frames to the decrypted depth video dDV in the original order; S473. Repeat steps S471 to S472 to decrypt all the remaining encrypted depth video frames and append them to the decrypted depth video dDV; dRGB and dDV are the final decrypted RGB information video and decrypted depth video.

Images