基于压缩感知和GHM多小波变换的信息隐藏算法

发布时间：2019-06-14 04:55

【摘要】：针对基于秘密信息置乱方法等类型的信息隐藏算法不可见性低和抗攻击性弱这一问题,提出了一种基于压缩感知和GHM多小波变换的信息隐藏算法。首先,将载体图像进行一次GHM多小波变换,再对所得到的中间能量区域进行一次小波变换得到HH分量,将HH分量进行奇异值分解;其次,将秘密图像进行小波变换,将得到的小波系数进行压缩感知得到观测矩阵,再对观测矩阵元素进行奇异值分解;最后,利用秘密图像的奇异值替换掉载体图像的奇异值来完成秘密信息的嵌入。实验结果表明,相比两种加密算法,算法不可见性(PSNR值)分别提高5.99%和22.11%;对低通滤波、椒盐噪声、高斯噪声、JPEG压缩等常见攻击具有良好的鲁棒性,相关系数(NC)平均增强了4.11%和11.53%。
[Abstract]:In order to solve the problems of low invisibility and weak anti-aggression of information hiding algorithms based on secret information scrambling method, an information hiding algorithm based on compressed perception and GHM multi-wavelet transform is proposed. Firstly, the carrier image is transformed by GHM multiwavelet transform, and then the HH component is obtained by wavelet transform to the intermediate energy region, and the HH component is decomposed by singular value decomposition. Secondly, the secret image is transformed by wavelet transform, and the wavelet coefficient is compressed and sensed to obtain the observation matrix, and then the elements of the observation matrix are decomposed by singular value decomposition. Finally, the singular value of the secret image is replaced by the singular value of the carrier image to complete the embedding of the secret information. The experimental results show that compared with the two encryption algorithms, the invisibility (PSNR value) of the algorithm is increased by 5.99% and 22.11%, respectively, and it is robust to low-pass filtering, salt and pepper noise, Gao Si noise, JPEG compression and other common attacks, and the correlation coefficients (NC) are enhanced by 4.11% and 11.53% on average.
【作者单位】：长安大学电子与控制工程学院;长安大学信息工程学院;
【基金】：国家自然科学基金资助项目(61402052) 陕西省自然科学基础研究计划项目(2014JM2-6105) 中国博士后科学基金资助项目(2015M572510) 陕西省博士后科学基金资助项目西藏自治区自然科学基金项目(2015ZR-14-20) 中央高校基本科研业务费专项资金资助项目(310832151092) 国家级大学生创新创业训练计划项目(201510710044) 2017年中央高校教育教学改革专项(研究生卓越人才培养计划项目)(310624176303)~~
【分类号】：TN911.73;TP309

【相似文献】