吕　敏　卿粼波　滕奇志　何小海

(四川大学电子信息学院　成都　610065)



基于DCT和SVD的岩心图像盲水印算法*

吕敏卿粼波滕奇志何小海

(四川大学电子信息学院成都610065)

摘要针对岩心图像的版权保护,提出一种基于离散余弦变换(DCT)和矩阵奇异值分解(SVD)的盲水印算法.首先根据人类视觉系统特征,将岩心图像变换到YCrbr空间,在亮度Y分量上,利用灰度共生矩阵(GLCM)选一块纹理复杂度最高的区域作为水印载体;再对该区域进行8×8分块DCT变换,在zig-zag变换后的频率系数中选取16个连续的系数进行SVD变换;最后利用U矩阵中系数的相关关系,嵌入一个二进制水印图像。实验结果表明,该算法能有效抵抗噪音干扰、裁剪和压缩编码等攻击,具有较好的鲁棒性、透明性。水印提取过程中不需要原始图像参与,实现水印图像盲提取。

关键词岩心图像; 灰度共生矩阵; SVD; DCT

An Algorithm of Blind Watermarking for Core Images Based on DCT and SVD

LV MinQING LinboTENG QizhiHE Xiaohai

(College of Electronic and Information, Sichuan University, Chengdu610065)

AbstractIn order to protect the core image copyright, a blind watermarking algorithm is proposed based on discrete cosine transform(DCT) and singular value decomposition(SVD). Firstly, according to the characters of human visual system, the core image is transformed to YCrbr space, and using the gray level co-occurrence matrix(GLCM) a texture complexity of the highest area is chosen as the watermark carrier on the luminance Y component. Secondly, the 8×8 block is transformed by DCT on the region, 16 consecutive coefficient are chosen in frequency coefficient after zig-zag transformation to be transformed by SVD. Finally, a binary watermark image is embeded by using U matrix of coefficient of correlation. The experimental results show that the algorithm can effectively resist some attacks, such as the noise interference, cutting and compression, with good robustness and transparency. The watermark image blind extraction is achieved, and the original image is not needed in watermark extraction process.

Key Wordscore image, gray level co-occurrence matrix, SVD, DCT

Class NumberTP391

1引言

在地质工作中,岩心是研究和了解地质和矿产情况的重要的实物资料,具有重要的学术和经济价值。为了解决其存放不便,以及存放过程中由于风化、腐蚀等造成的信息丢失,利用岩心扫描技术将岩心实物资料转换为数字图像资料[1]。数字水印技术已经成为岩心图像版权保护的重要手段之一。

目前,数字水印主要分为空域法和频域法。频域法的因其抗噪声攻击和抗压缩能力强,能将水印信号能量分布到所有像素上的特点而被广泛地使用。常见的用于数字水印的频域变换有傅里叶变换(DFT)、离散余弦变换(DCT)、小波变换(DWT)等。……