AN EFFECTIVE CONTENT-BASED COPY DETECTION SCHEME FOR IMAGE AUTHENTICATION

In this paper, a content-based copy detection scheme for image authentication is proposed. The proposed scheme is based on DCT coefficients and Radon transform. It combines the local feature with the global feature to detect suspicious images. The local feature vector is extracted by dividing blocks and selecting the largest DC coefficients. If the similarity is less than a threshold, the global feature vector will be extracted. Radon transform (RT) and discrete Fourier transform (DFT) are then exploited to extract the global feature vector. Experimental results demonstrate that the proposed scheme outperforms other existing techniques for image authentication.