Novel DWT-based zero-watermarking for image with fake verification inhibited
Jiyin Sun · Jisuanji gongcheng yu sheji · 2011
According to analysis,a new security problem of fake verification in watermark algorithms is discovered.To avoid such security problem which may exist in the new presented image zero-watermark algorithm based on multilayer decomposition in DWT domain,a new method which can inhibits the fake verification is presented.On the basis of hypothesis testing theory,this method analyzes the conditions that the watermark algorithm robustness and fake verification are related to the image local wavelet coefficient difference and variance change.Then,the algorithm parameter's best alternative conditions and its improved watermark testing method are presented.Experimental samples show that the new algorithm not only has good robustness to such operations as compression,shearing,noise addition,median filtering and scaling,but also has good inhibition to possible watermark fake verification.