Design And Implementation Of Copy-move Forgery Image Detection System

In recent years, with the rapid rise of mobile Internet, the dissemination of information has become more and more frequent. Digital image as an important information carrier has become an integral part in people’s daily life communication. At the same time, the image editing software began to appear, and the image editing software is very easy to operate digital images and the important information content of the images is likely to change after faking images, so that the digital image is no longer credible. Copy-move forgery is a typical and common digital image tampering method. Therefore in the thesis, by combining copy-move forgery detection algorithm, the copy-move forgery detection system based on C# development language and.NET framework was designed. It is CMFD system. The main contents of the thesis are as follows:1. The thesis firstly introduces the research background and significance of the detection system, and describes the research status of the research on the copy-move forgery detection in China and abroad. Then, the CMFD method and related technology are analyzed, including the analysis of CMFD mathematical model and the introduction of the common CMFD method. The thesis gives the applicable conditions of the mathematical model in CMFD system. And five common copy-move forgery detection algorithms are introduced. At last, the.NET framework and C# technology are introduced.2. Each module of CMFD system is designed. Firstly, according to the function of the system, the overall design of the system is carried out. Each function module is divided, including image input module, detection method module and image output module. These methods including DCT detection, the main transfer vector detection, Hu moments detection, PCA detection and SVD detection method are designed. And each of these methods has its own characteristics and scope of application. The integration of multiple detection methods provides convenience for the user to select multiple choices.3. Lastly, the thesis describes the realization process of CMFD system, and some images from the copy-move forgery image library are tested. Firstly, according to the overall design content of CMFD system and the design content of each function module, the thesis introduces the implementation process of the main interface module, the image input module, the detection method module and the image output module in the system. Then, five kinds of detection methods for CMFD system are carried out system testing, and all kinds of test results are analyzed and compared.In the thesis, the digital image CMFD system designed on the system level solves part of the problem of copy-move forgery detection work, and this system can successfully detect various types of copy-move forgery image. But there are some deficiencies to need to be improved for further optimization of the system. For example, the lack of image input process image classification, and adaptive selection issues of detection methods, which require further study.