基于协方差矩阵的CFA插值盲检测方法  

Blind CFA Interpolation Detection Based on Covariance Matrix

作　　者：王波[1] 孔祥维[1] 尤新刚[1,2] 付海燕[1] 

机构地区：[1]大连理工大学信息安全研究中心,大连116024 [2]北京电子技术应用研究所,北京100091

出　　处：《电子与信息学报》2009年第5期1175-1179,共5页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(60572111)资助课题

摘　　要：从数字图像中盲检测数码相机采用的CFA插值算法,可以为数字图像取证提供重要的技术手段。该文基于线性插值模型,利用协方差矩阵构建插值系数方程组,并将估计的插值系数构成特征向量空间,采用支持向量机作为分类工具,提出了一种对不同的CFA插值算法进行准确分类的盲检测方法。实验表明,该文方法对于不同的CFA插值算法均能达到较高检测正确率。同时,相比现有的CFA插值检测方法,该文算法对加性高斯白噪声和有损JPEG压缩具有更好的鲁棒性。Blind CFA interpolation detection, which identifies the demosaicing method used in digital camera by analyzing output images, provides an efficient tool for digital image forensics. This paper proposes an approach of blind CFA interpolation detection based on interpolation coefficients estimation. By solving the covariance matrix equation, a vector of the interpolation coefficients is obtained, which is further fed to SVM classifier. The experimental results show a high accuracy on blind CFA interpolation detection. Compared with existing ones, the proposed method in this paper indicates a better performance on the robustness against additive Gaussian white noises and lossy JPEG compression.

关 键 词：数字图像取证 色彩滤波阵列插值 协方差矩阵 支持向量机 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]