3D H.264 Video Steganography Algorithm Based On The Optimal Coefficient Matching

Digital steganography is refers to hide the secret information to the public media by the use of the human eye’s perception of redundancy and carrier data redundancy,so as to achieve passing information secretly.Diamond coding is an image hiding algorithm which hides information by modifying the pixel.However, the modify amplitude of the pixel and the DCT coefficient is not in a measure, so image hiding algorithm DE cannot be directly applied to Video Steganography.Matrix encoding is a mature video information hiding algorithm. But with the increase of the number of embedding bits at one time, its embedding rate rapidly decreases.We improve diamond coding and its improved algorithm adaptive pixel pair matching to propose a new data-hiding method multidimensional quantized DCT coefficient matching based on 3D H.264/AVC(MCM).That is to select a plurality of DCT coefficients as variable to construct a linear function and compute the difference between value of the function and secret information to be embedded. Then we do a smallest increase or decrease to DCT coefficients according to the value of difference to embed secret information.This algorithm greatly improves the security of information hiding. The proposed algorithm compared with DE and APPM get a better experimental result in embedding efficiency、embedding rate、PSNR 、SSIM 、 bit rate increase and so on.Efficient Steganographic Embedding by Exploiting Modification Direction(EMD) greatly improves the efficiency of the modulo operation. We take a full advantage of EMD algorithm to improve the matrix encoding and propose a new data-hiding method of high embedding rate matrix-encoding based on 3D H.264/AVC(HEME).The basic idea of this algorithm is to get the optimal transformation matrix with the overall average minimum modifier by controlling the modify amplitude of coefficients, and construct a carrier vector by choosing several middle frequency QDCT coefficients.Calculate multiplication of transformation matrix and carrier vector, and then compute the difference between multiplication vector and secret information vector to be embedded. At last we adjust DCT coefficients according to the value of difference vector to implement the mapping relationship between carrier vector and secret data vector.Experimental results show that it can achieve a high embedding efficiency and a high embedding rate compared to the original matrix coding(F5 algorithm).