An Improved Lexicographical Sort Algorithm of Copy-move Forgery Detection

As a technique for digital image tampering, copy-move forgery is frequently used. In this paper, an improved lexicographical sort algorithm based on discrete cosine transform (DCT) is developed to detect copy-move forgery. Firstly, the image is split into 8*8 blocks, and the image data undergoes a DCT. Then the DCT coefficients were grouped to reduce the dimension according to the frequency property. Finally, the distance of eigenvectors, instead of the DCT coefficients, was taken as the eigenvalue to fulfill the block matching. Experiments results showed that the false matching ratio of the proposed algorithm was reduced while the detecting ability was maintained.