数字图像模糊滤波操作常用于美化润饰“伪造”图像。针对常用的均值滤波、空域高斯低通滤波与中值滤波, 提出了一种能同时检测上述3种操作的盲取证算法。首先将高频残差作为特征提取域, 然后分别基于二值局部模式LBP和自回归模型提取特征, 最后使用支持向量机构造模糊滤波检测器。实验结果表明, 所提算法能有效地检测模糊滤波操作, 对抗JPEG压缩的鲁棒性能较好。

The digital image fuzzy filtering operation is always used to beautify the forged image. In this paper, we propose a blind forensics algorithm which can detect the average filter, Gaussian low-pass filter and median filter. Firstly, the high frequency residual is used as the feature extraction domain, and then the feature extraction is carried out based on the local binary pattern (LBP) and the autoregressive model. Finally, A fuzzy filtering detector is set up by using the support vector machine. The experimental results show that the proposed method can detect the fuzzy filtering operation effectively and has good robustness against JPEG compression.