Multi-objective optimization H.264 video steganography method with constraint conditions
A multi-objective optimization and constraint technology, applied in the field of multi-objective optimization H.264 video steganography, can solve problems such as poor performance and not considering local optimality
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0131] Such as figure 1 , figure 2 as well as image 3 As shown, a multi-objective optimization H.264 video steganography method with constraints includes the following steps:
[0132] Step S1: Establishing a three-dimensional set of motion vectors according to the temporal and spatial positions of the motion vectors in a video, and determining the correlation measurement factor of the motion vectors on the three-dimensional set of motion vectors;
[0133] Step S2: Confirm the local optimal motion vector set Ω in the video S and Ω T , and modify the locally optimal motion vector set Ω S and Ω T ;
[0134] Step S3: Confirm the horizontal component forbidden mode set F of each motion vector in the video h and the forbidden pattern set F of the vertical component v ;
[0135] Step S4: According to Ω S , Ω T , F h and F v These four sets solve the locally optimal constraints of any one motion vector in the three-dimensional set of motion vectors;
[0136] Step S5: C...
Embodiment 2
[0243] Such as Figure 4 , as shown in Table 1, Figure 4 Under different encoding parameters, the security of the algorithm UEDWR proposed by Aly [1], Yao [2], Zhang [3] and this patent against the steganalysis algorithm based on the local optimal characteristics of the motion vector is demonstrated. The detection algorithm is NPEFLO (Near Perfect Estimation For Local Optimality), see the paper [6] for details. Load is measured in average embedded bytes per motion vector - bpmv (bits per motion vector). in Figure 4 (1) The experimental parameter condition is QP=28, the frame structure is IPPPP, and the motion search method is EPZS [7]; Figure 4 (2) The experimental parameter condition is QP=28, the frame structure is IBBBP, and the motion search method is EPZS; Figure 4 (3) The experimental parameter condition is QP=28, the frame structure is IBBBP, and the motion search method is HEX [8]; Figure 4 The experimental parameter condition of (4) is QP=18, the frame struc...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


