Copy Move Forgery (CMF) is a type of digital image forgery in which an image region is copied and pasted to another location within the same image with malicious intent to misrepresent its meaning. To prevent misinterpretation of an image content, several Copy Move Forgery Detection (CMFD) methods have been proposed in the past. However, the existing methods show limited robustness on images altered with post-processing attacks such as noise addition, compression, blurring etc. In this paper, we propose a robust method for detecting copy-move forgeries under different post-processing attacks. We use Discrete Cosine Transform (DCT) to extract features from each block. Next, Cellular Automata is employed to construct feature vectors based on the sign information of the DCT coefficients. Finally, feature vectors are matched using the kd-tree based nearest-neighbor searching method to find the duplicated areas in the image. Experimental results show that the proposed method performs exceptionally well relative to the other state-of-the-art methods from the literature even when an image is heavily affected by the post-processing attacks, in particular, JPEG compression and additive white Gaussian noise. Furthermore, experiments confirm the robustness of the proposed method against the range of combined attacks.

复制移动伪造（CMF）是一种数字图像伪造，其中将图像区域复制并粘贴到同一图像内的另一个位置，以恶意意图歪曲其含义。为了防止误解图像内容，过去已经提出了几种复制移动伪造检测（CMFD）方法。但是，现有方法对经过后处理攻击（如噪声添加，压缩，模糊等）更改的图像的鲁棒性有限。在本文中，我们提出了一种在不同后处理攻击下检测复制移动伪造的鲁棒方法。我们使用离散余弦变换（DCT）从每个块中提取特征。接下来，使用元胞自动机基于DCT系数的符号信息构造特征向量。最后，使用基于kd树的最近邻居搜索方法对特征向量进行匹配，以找到图像中的重复区域。实验结果表明，即使图像受到后处理攻击（尤其是JPEG压缩和加性高斯白噪声）的严重影响，所提出的方法相对于文献中的其他最新方法也具有出色的性能。此外，实验证实了所提出的方法针对组合攻击范围的鲁棒性。JPEG压缩和加性高斯白噪声。此外，实验证实了所提出的方法针对组合攻击范围的鲁棒性。JPEG压缩和加性高斯白噪声。此外，实验证实了所提出的方法针对组合攻击范围的鲁棒性。