关键词: 数字图像取证, 图像区域复制粘贴篡改, 灰度共生矩阵

Abstract: With regard to the copy-move forgery of image region, this paper proposes a detection algorithm based on gray level co-occurrence matrix. Firstly, we divided the detected image into multiple overlapping blocks with same sizes, represented the textural features of each block with the statistics of its gray level co-occurrence matrix, and got the feature vector of the image. Secondly, we sorted the feature vector by dictionary, and located the tampered region by utilizing the displacement vectors of image blocks. Lastly, experimental results show that our algorithm performs better than the classical detection algorithm based on Principal Component Analysis (PCA) in terms of the processing against rotate operation and of efficiency.

Key words: digital image forensics, image region copy-move tampering, gray level co-occurrence matrix