针对数字图像篡改一种最经常使用的复制粘贴篡改手段,提出了一种基于脊波变换的图像篡改检测取证方法。算法利用了脊波变换是Radon变换切片上应用小波变换这种特性,实现了复制一粘贴篡改的鲁棒识别。针对十一大类图像的仿真实验结果表明,算法对于旋转变换、JPEG压缩和噪声添加都具有良好的鲁棒性,对于压缩新标准JPEG2000也表现出了较好的鲁棒性。
While it is more and more popular for obtaining and manipulating digital image, digital image forensics is more and more important in the field of mulimedia information security. This task can be crucial when images are used as evidence to influence or decide judgment such as in a court of law. Ai ming at copy-move forgery, which is the most popular tampering method, many detecting methods have been proposed by researchers all over the world. But these methods seldom take the new compression standard-JPEG 2000 into account, which will be more and more popular. Based on this, a robust detecting method based on ridgelet transform is proposed in this paper. The method is according to the fact that ridgelet transform applies wavelet transform on the slices gained from Radon transform. Copy-move for gery detecting methods using Radon transform have already been proposed and proved to be robust to some kinds of attacks except JPEG 2000. The experiments done with eleven types of digital images show that the proposed method is a robust to attaekes, such as rotation,JPEG compression, adding noise. Es- pecially,the method is robust to the new compression standard of JPEG 2000.