中文摘要：

通过对现有视频隐写算法的分析与研究,针对其嵌入效率低,隐藏性和安全性差等缺点,提出一种基于余数分类的视频隐写算法.该算法在量化后的离散余弦变换域的中低频区中选取8个系数,对其进行模5运算求余,然后根据余数的不同来进行分类.最后统计每个类别的个数,根据其奇偶性来嵌入4bit信息.实验结果证明:该算法在对载体系数做出少量修改的情况下,可以嵌入大容量的信息.与其他同域的隐写算法比较,在信息嵌入率,峰值信噪比,安全隐蔽性等方面都有明显的优势.

英文摘要：

Through analysis and research on steganography, in view of the low efficiency of the embedded, hidden and safety bad, a steganographic algorithm was proposed based on remainder classification of the video. The algorithm selected 8 coefficients in DCT (discrete cosine transform) quantized in the low frequency region. The 8 coefficients are carried out the module 5 operations for remainders., and then classified them according to the different. At last, we statistics the number of each class of remainders, and embed the 4 bit information according to the parity of the amount of each class. Experiments show that the proposed algorithm can embed large capacity information in the case of a few modifications to the carrier coefficient. Compared with other similar steganography algorithms, it has obvious advantages in information embedding rate, PSNR (peak signal to noise ratio), security concealment and so on. ? 2016, Editorial Board of Journal of Huazhong University of Science and Technology. All right reserved.