Curvelet transform based robust copyright protection scheme for color images using extended visual cryptography

摘要

A copyright protection scheme based on curvelet transform, k-means clustering and extended visual cryptography is proposed for color images. Unlike the existing schemes, the proposed scheme creates meaningful shares to provide better security and handles false positive cases efficiently. Curvelet Transform is applied on Y component of host image to generate six-scale layers. Out of these layers, non fine scale layers are selected to generate the master share, which is further processed by Henon map to enhance its security. Two watermarks are used in the proposed scheme. One of them is provided by the user and other one is constructed by using C_b, and C_r components of the host image. Second watermark is used to to handle false positives. Meaningful ownership share is constructed from the master share, both watermarks and cover image by using extended visual secret sharing. Master share is overlapped with its corresponding ownership share to reveal the watermark which is further used to prove the ownership. Experimental results show that the proposed scheme clearly verifies the copyright of the digital images, is robust to withstand several image processing attacks and handles false positive cases efficiently. Comparison with the existing copyright protection schemes shows that the proposed scheme gives better performance.