Research On The New Technology Of Region Copy-Move Passive Forensics Of Digital Image

Image forensic. as one of the most important branches of image analysis, have been widely researched by more and more scholars, especially in recent years, These could be result from the facts that the image have been become the official documents, merely while tampering with digital images without leaving any obvious traces becomes cheaper and cheaper with the development of the digital devices. In the other hand, Using the existing methods and systems can not give the perfect judgment. "The seeing does not believe". The problems and challenges of Digital image content are becoming more and more outstanding.This paper firstly briefly give an overview if the recent literatures about image forensics, and divide those methods into two categories. One is the identification of image source and the other is Image content forensic. As the most popular forgery, Copy-move forgery is the main research including the operations in one image and two or more images, this paper mainly introduced the first forgery. Comparing to the traditional literatures, the paper proposes the methods have some advantages as follows.The traditional literatures can not work well for the geometry deformation; we propose the new method based on the circle structures which is robustness for rotation. Through calculating the seven characterizes in space domain. The discrete wave decompression is applied to reduce the size of image. The experiments show that this method improves the accuracy and save time-consuming.Meanwhile this paper also introduces another method based on the singular value decompression and radial moments. Because singular value decompression is robustness for the geometry deformation and the radial moments is applied to compute the parameter of the rotation. The main rotate angle estimation is used to remove the mismatch blocks instead of the main vector shift. The pyramid decompression is also applied to reduce the complexity of algorithm. The seven characterizes from the SVD are computed by the circle structure.This paper also tries to use the geometry active contour for the image bind detection. its main idea is to chose the incredible area. Then with the help of the geometry moments and the Similarity function, we finally give the right answer whether the image is modified or not.In the last part, we give a simply method to detect the copy-move forgery between two images. This uses the standard quantitative matrix rather than estimating the exact matrix. Because the only fact must be indentified that whether the image contains the two or more quantitative factors. The different quantitative losses are produced between the nature image and forgery image. The experiment shows that this method can reach a good result in gray and RGB images.