摘要
针对Tian提出的差值扩展可逆隐写算法(DE算法),以及该算法变形延伸算法存在的位置图比特流简单压缩方法导致隐写容量变小的问题,为了提高位置图压缩效率进而提高隐写容量,通过分析图像差值矩阵相关性对位置图比特流中连续0、1分布的影响,提出一种利用图像差值相关性来提高隐写容量的改进算法。实验结果表明,改进算法显著提高了隐写容量,且对位置图压缩效率约可到提高35%,PSNR平均提高1 d B。
An improved algorithm based on the reversible difference expansion （DE） algorithm put forward by Tian is proposed by analyzing the ways that how the host image pixels＇ differences correlation influence the continua- tion of 0, 1 of location map. The proposed algorithm takes advantage of the pixels＇ differences correlation to improve the efficiency of location-map compressing and to solve capacity limits caused by crude compression of location map. The experiment results show that the proposed algorithm can notably improve capacity with a improved location-map compression efficiency by - 35% and a PSNR by 1 dB on average.
出处
《电子科技》
2015年第11期33-36,共4页
Electronic Science and Technology
关键词
差值扩展
隐写容量
位置图
压缩
difference expansion （DE）
steganography capacity
location-map
compression
