The Generation And Evolvement Of Texture In Image Scrambling

At present, the scrambling transform of digital image has become one of important means of the secure transmission and cryptical storage. With the development of information hiding and digital watermarking, the scrambling transform has been also used in its pre- and post-processing and gets well effect. In this thesis, we analyze a number of spatial scrambling methods, use each method to scramble different simulated image several times, and then by observing and comparing the scrambling effects of the different simulated pictures, we sum up the rules of texture evolvement in the process of scrambling and explain the causes and conditions of the phenomenon of texture in image scrambling. At last, we use the "entropy" extracted by Gray Level Co-occurrence Matrices to evaluate the effect of image scrambling. The experimental results show that it is possible to evaluate the scrambling effect using the texture analysis where texture is available.