基于缩放不变矩的鲁棒DCT 域图像拷贝检测方法.docVIP

第 42 卷 第 4 期 2013年7月  电 子 科 技 大 学 学 报 Journal of University of Electronic Science and Technology of China  Vol.42 No.4 Jul. 2013  基于缩放不变矩的鲁棒DCT域图像拷贝检测方法 余艳玮1,2，周学海1,2，许华杰3  (1. 中国科学技术大学软件学院  合肥  230051;  2. 中国科学技术大学苏州研究院嵌入式系统实验室  江苏 苏州 215123; 3. 广西大学计算机与电子信息学院 南宁 530004)  【摘要】提出一种基于不变矩的DCT域图像拷贝检测方法，选取图像亮度分量的整体DCT变换后的中低频系数的符号作 为其DCT域特征，分别计算原始图像及其角度规范化后图像的DCT域特征，计算并比较查询图像与图像库中图像的两类特征 的相关性，取较大的相关值来判断图像是否存在拷贝。图像的整体DCT中低频系数已被验证对于信号处理攻击鲁棒性高，且 随着图像缩放攻击呈规律性变化，因而利用图像的整体DCT中低频系数构建的DCT域特征对于信号处理和图像缩放攻击鲁棒 性高；利用图像的缩放不变矩估算出图像特征角度，再依据特征角度对图像进行角度规范化可消除旋转失真的影响，因而利 用角度规范化后的图像的DCT域特征能抵抗旋转、以及旋转和等比例缩放的混合攻击。实验结果证明在抗旋转和包含旋转的 混合攻击方面，该算法的检测效果明显优于已有算法。 关 键 词 拷贝检测; 几何失真; 图像归一化; 查准率; 查全率 中图分类号 TP391.41 文献标志码 A doi:10.3969/j.issn.1001-0548.2013.04.014  Scale-Invariant Moment Based Robust Image Copy Detection in DCT Domain YU Yan-wei1,2, ZHOU Xue-hai1,2, XU Hua-jie3 (1. School of Software Engineering, University of Science and Technology of China Hefei 230051; 2. Embedded Systems Laboratory, University of Science and Technology of China Suzhou Jiangsu 215123; 3. School of Computer and Electronic Information, Guangxi University Nanning 530004)  Abstract  In this paper, an invariant moment based robust image copy detection algorithm in discrete cosine transformation (DCT) domain is proposed. Low and middle frequency coefficients which are calculated from full DCT transformation on luminance component of image are used as the image eigenvector in DCT domain. The query image and its angle normalized image eigenvector in DCT domain are computed, then correlation values of two eigenvectors of the query image and those of test image are respectively computed and the larger one is used to judge whether the query image is a copy or not. Full DCT coefficients are verified to have high robustness against common signal attacks and regularly vary with scaling attacks, thus the DCT eigenvector constructed by middle and low frequency full DCT coefficients can resist signal