Study Of Digital Image Scrambling Degree Evaluation Method Based On Pixel Value Relativity

With the rapid development of computer network and multimedia technology, a large number of image data information being transmitted by the Internet, the security of this information is gradually becoming the concerns of the people. Since image data with large quantity of information can not been encrypted by traditional encryption methods, many kinds of digital image encryption methods have been presented, image scrambling is one extensive method of them. Such as the scrambling method based on matrix transforming, the scrambling method based on pixel value changing, the scrambling methods based on chaotic map and so on. Despite more and more image scrambling methods have been proposed, the evaluation methods are rarely, and the existing methods also mainly focus on a specific scrambling algorithm.Because the digital image is a clear image containing a lot of information and the adjacent point pixel value is correlative, the relativity is large. After scrambling transformation, each pixel point of the digital image is altered, the corresponding pixel values are changed, thus the relativity is changed. When the relativity is the minimum after scrambling transformation, which is equal to the difference of the adjacent point pixel value is the maximum, scrambling effect is the best. According to this principle, an evaluation method of digital image scrambling degree based on pixel values relativity is presented in this paper. The material method is comparing the difference between scrambling image and an ideal scrambling image and practicing by Matlab software, the large amount of experimental data and results analysis show that the method is accurate and effective. For a digital image scrambling method, it can reflect the relativity between the scrambling times and scrambling degree well. For different digital image scrambling methods, it can reflect which method is better and which one is worse.At the same time, on the basis of analyzing the existing classical digital image scrambling methods, combining chaotic map, a new method based on image segmentation is presented in this paper. Through designing the key and pixel point rearranging, the method achieves position scrambling and achieves the purpose of image encryption. With Matlab software, the results demonstrate the practicality of the method. The method has the advantages of large key space and achieving easily. Meanwhile, the relation of the keys and the scrambling relativity has been analyzed.