控制关键帧选择的H.264熵编码加密算法

发布时间：2018-03-07 14:57

本文选题：关键帧　切入点：熵编码　出处：《中国图象图形学报》2014年03期 　论文类型：期刊论文

【摘要】：目的常见的H.264熵编码加密是一种将加密方案与熵编码方式融合的视频安全算法,这种算法不仅加解密速度快且可保持数据格式不变,是一种兼容性的加密方案,但需要处理帧内所有数据。方法针对熵编码加密过程中存在的不足,提出了一种基于关键帧选择与熵编码相结合的视频加密方法,通过获取一帧中所有8×8子宏块的运动矢量,构造一个用于表示每一帧内容变化情况的向量F,然后将其欧几里德范数与阈值K比较,如果大于或等于阈值,则确定当前帧是关键帧,则在下一帧熵编码时加密其运动矢量差与残差系数值,否则不参与任何加密运算。结果通过控制阈值K的大小来决定关键帧的选择,并结合混沌序列加密关键帧数据来保证视频信息的安全性,与熵编码加密算法相比,加密数据平均减少数据量约为39.78%。结论实验结果表明该加密方法在保持编码后数据总量不变的情况下,以较小的时间代价获得了良好的加密效果,且与常规的熵编码加密算法相比大幅度减少了加密数据量。
[Abstract]:Objective H.264 Entropy Encoding encryption is a kind of video security algorithm which combines encryption scheme with entropy encoding method. This algorithm not only has high encryption and decryption speed, but also keeps the same data format, and is a compatible encryption scheme. However, we need to deal with all the data in the frame. Aiming at the shortcomings of entropy encoding encryption, a video encryption method based on key frame selection and entropy coding is proposed. The motion vectors of all 8 脳 8 sub-macroblocks in a frame are obtained. A vector F is constructed to represent the change in the content of each frame, and its Euclidean norm is compared with the threshold K. if the threshold is greater than or equal to the threshold, it is determined that the current frame is a key frame. Then the value of motion vector difference and residual coefficient is encrypted in the next frame entropy coding, otherwise it does not participate in any encryption operation. The result is to control the size of threshold K to determine the choice of key frame. The security of video information is ensured by using chaotic sequence to encrypt the key frame data, which is compared with the entropy encoding encryption algorithm. The experimental results show that the encryption method can achieve a good encryption effect at a small time cost while keeping the total amount of the encoded data unchanged. And compared with the conventional entropy encoding encryption algorithm, the amount of encrypted data is greatly reduced.
【作者单位】：江西理工大学信息工程学院;
【基金】：国家自然科学基金项目(11062002,61363076) 江西省教育厅重点科技项目(GJJ13435) 江西省研究生创新专项资金项目(YC2012-S092)
【分类号】：TN919.81

【参考文献】