The Research Of Three-dimensional Video Information Hiding Technology And Its Application

Due to the three-dimensional(3D)video can provide users with stereoscopy and telepresence,3D video technology has been under rapid development.With the maturity of network technology and the open internet,3D video can be easily accessed and processed,the security of 3D video has become the great challenge to 3D product commercialization.Information hiding technology can complete the steganography and copyright authentication of multimedia data effectively,and it is identified as one of effective approaches to resolve 3D video information security issue.Owing to the large amount of high-definition video,it is imperative to compress the video effectively for the sake of saving transmission bandwidth and storage space.The High Efficiency Video Coding(HEVC)standard is published by the ITU-T Video Coding Experts Group and ISO/IEC Moving Picture Experts Group,3D-HEVC based on the 3D expansion of HEVC is the latest 3D video compression coding standard.For aiming at video's information security problem,video information hiding technology has been studied under the HEVC and 3D-HEVC compression coding standard in this paper.1.For the video stream copyright issues of HEVC,a new zero-watermarking algorithm with robustness to re-quantization transcoding is proposed.Firstly,from statistics analysis about re-quantization transcoding,it is found that Coding Unit(CU)depths information have strong stability and obvious transfer rule;Then,combining with the transfer rule of depths information,the CU depths are divided into two groups according to the stability rules and the balance rules of watermarking,and mapped into binary watermark information;Finally,the watermark information is encrypted,and ‘xor' operation is performed between the binary watermark information encrypted and the copyright information scrambled by using the chaotic algorithm,the outcome with the timestamp acts as the ultimate registered zero-watermarking.Experimental results show that the watermark extraction accuracy of proposed algorithm is at least 85% to re-quantization transcoding attacks within a certain range and other common signal attacks,the proposed algorithm has strong robustness.2.For security problems of depth video,using a new carrier embedded information,an information hiding algorithm of depth video in 3D-HEVC based on single depth intra mode is proposed.Firstly,a candidate list is established about the CU block which use single depth intra mode coding,and the two neighborhood pixels are compared in the candidate list;Then,in order to further increase the security of the algorithm and avoid error drift problem,in the case of equal pixel values and unequal pixel values,choosing the CU block being embedded accurately by the key and the prediction mode range of the subsequent block,respectively;Finally,the index value of the selected pixel in the candidate list of the selected CU block is modulated by the secret information.Experimental results show that in comparison to original HTM13.0,the average decrease of PSNR and SSIM of the rendered virtual view image based reconstructed depth video which embedded secret information are 1.41×10-3dB and 6×10-6,respectively and the bitrate of video stream increases 0.12% on average.The experimental results show that the proposed algorithm has a slight impact on the subjective perception quality of 3D video and bitrate.3.Due to be short of the research on human visual perception characteristics in the conventional video information hiding algorithm,an inter information hiding algorithm based on visual perception in 3D-HEVC is proposed.Firstly,the stereoscopic video image is divided into salient region and non-salient region through stereoscopic significant model and the otsu threshold;Then,the mapping relationship is established between secret information and Quantization Parameter(QP),and in case of salient region and non-salient region,the QP of P frame or B frame is modulated by the secret information using Remainder Equal Principle,respectively;Finally,using the QP embedded secret information to code.Experimental results show that the average decrease of PSNR and SSIM of reconstructed stereoscopic video in the proposed algorithm compared with the original HTM13.0 are 3.14×10-2 dB and 5.8×10-5,the bitrate of video stream increases 0.0741% on average,and the secret information can be extract blindly.