Detection of Copy-Move Forgery in Flat Region Based on Feature Enhancement

摘要

A new Feature Enhancement method based on SURF is proposed for Copy-Move Forgery Detection. The main difference from the traditional methods is that Contrast Limited Adaptive Histogram Equalization is proposed as a preprocessing stage in images. SURF is used to extract keypoints from the preprocessed image. Even in flat regions, the method can also extract enough keypoints. In the matching stage, g2NN matching skill is used which can also detect multiple forgeries. The experimental results show that the proposed method performs better than the state-of-the-art algorithms on the public database.