Research On Evenness Degree Of Digital Image Scrambling And Its Application In Image Information Security.

Digital image scrambling can be used not only as a kind of encryption schemes for digital images, but also as the pre-or post-processing for image information hiding schemes. For those methods of image scrambling in positional space, there's still lack of effective methods to judge the scrambling effect and to give them a reasonable classification. In this thesis, we point out the shortcomings of the existing methods in evaluating image scrambling effect, and propose the concept of"Evenness Degree"to describe the effect. The proposed method which is based on statistics and probability theory can describe the relationship between the scrambling iterations and the scrambling effect well. The experimental results show that the proposed method are quite consistent with visual observation. Several spatial scrambling methods are then introduced and their scrambling effect is analyzed in terms of a classification of the methods as point-wise, line-wise and region-wise. An application of the even scrambling in the image information security is given and implemented. Finally, some methods are given for restoring the missing pixels of a scrambled image during the transmission, and are evaluated as well.