在视频点播等应用中,为了刺激消费者购买高质量的视频,需要对视频进行特别的感知加密,使视频加密后仍然部分可见。该文提出一种Context-based Adaptive Binary Arithmetic Coding(CABAC)的视觉质量可控的快速感知加密算法。利用2D超混沌系统生成的流密码对重要的语法元素和敏感的编码元素异或加密。加密内容包括运动矢量差的符号位、非零系数符号位和语法元素significant_coeff_flag。分析及实验结果表明,提出的感知加密算法对码率无影响,编码时间仅增加7‰左右,视频质量随着质量因子的改变而变化,满足视频感知加密的应用需求。
In Video on Demand (VoD) applications it is desired that the encrypted multimedia data are still partially perceptible after encryption in order to stimulate the purchase of the high-quality versions of the multimedia products. This perceptual encryption requires specific algorithms for encrypting the video data. In this paper, a Context-based Adaptive Binary Arithmetic Coding (CABAC) video perceptual encryption scheme is designed. The video quality of this encryption scheme is controllable. The important syntax elements and sensitive coded elements are chosen to encrypt using mathematical XOR operations with stream ciphers generated by 2D hyper chaos system. The encryption scheme is composed of encrypting Motion Vector Differences (MVD) sign, non-zero coefficients sign and significant_coeff flag. Theoretical analysis and experimental results show that the proposed scheme has no impact on bit rate. With the 7‰ encoding time increasing, the video quality can be controlled by the change of the quality factor, which meets the requirement of the video perceptual encryption.