Contourlet-SVD域的YIQ彩色图像全息水印算法  

The YIQ color image holographic watermarking algorithm of Contourlet-SVD domain

作　　者：马婷[1] 陈农田[2] 朱建刚[1] MA Ting;CHEN Nong-tian;ZHU Jian-gang(School of Computer,Civil Aviation Flight University of China,Guang Han,Sichuan 618307,China;School of Safety Engineering,Civil Aviation Flight University of China,Guang Han,Sichuan 618307,China)

机构地区：[1]中国民用航空飞行学院计算机学院,四川广汉618307 [2]中国民用航空飞行学院安全工程学院,四川广汉618307

出　　处：《光电子．激光》2020年第12期1270-1277,共8页Journal of Optoelectronics·Laser

基　　金：国家自然科学基金项目(60472127)资助项目。

摘　　要：选取YIQ域的颜色空间将共轭对称扩展傅立叶数字全息技术与Contourlet-SVD域进行结合,提出一种基于YIQ和Contourlet-SVD域的彩色图像全息水印算法。由于在RGB图像的任一分量嵌入水印对视觉影响较大,所以选择将RGB图像转换成YIQ域的颜色空间,选取亮度分量Y进行2层Contourlet变换,在得到的细节子带中选取低频系数进行SVD奇异值分解,将对水印图像进行共轭对称扩展傅立叶数字全息方法所生成得水印信息隐藏。最后的实验结果表明,该算法可以保证水印嵌入足量的基础上,确保了水印信息的较强抵抗能力。该算法对高斯噪声、JPEG压缩攻击的鲁棒性较强,NC值均达到0.91以上。Select the color space of the YIQ domain to combine the conjugate symmetric extended Fourier digital holography technology with the Contourlet-SVD domain.A color image holographic watermarking algorithm based on YIQ and Contourlet-SVD domain.Because the watermark is embedded in any component of the RGB image,it has a large visual impact.Choose a color space for converting RGB images to YIQ domains.A 2-layer Contourlet transform is conducted for the luminance component Y.Select low-frequency coefficients in the obtained detail subbands to perform SVD singular value decomposition.Conjugate symmetrically expanded fourier digital holography is conducted on the watermarked images to generate the watermarked images.The experimental results show that the algorithm ensures a strong resistance to the watermark information on the basis that enough watermark information can be embedded.The algorithm is robust to Gaussian noise and JPEG compression attacks,and the NC values are all above 0.91.

关 键 词：数字全息技术 YIQ颜色空间 CONTOURLET变换 SVD 

分 类 号：TN911.73[电子电信—通信与信息系统] TP391.41[电子电信—信息与通信工程]