Copy-move attack is one of the most popular digital image forgery attacks. The main problem is that existing studies do not provide high detection accuracy with low computational complexity. High complexity of existing feature based solutions makes impossible to use them for large remote sensing snapshots analysis. In this paper there is proposed a copy-move detection algorithm based on perceptual hash value calculation. Hash values are evaluated using the result of binary gradient contours computation. The proposed solution showed high detection accuracy and low computational complexity for copy-move detection in remote sensing data.
展开▼
