M.Sc. Tezi Görüntüleme | |||||||||||||||||||||
|
|
||||||||||||||||||||
| Summary: The accuracy and reliability of digital images are very important. Copy move forgery is one of the most common digital image forgeries. As such an action is easy to accomplish, the image authentication systems should carry out the detection of this kind of forgery in its best. In this thesis study, current key point based forgery detection methods are analysed and improved. Moreover, a novel keypoint based copy move detection method is proposed. The proposed detection method is found to be effective in detecting copy move forgeries without requiring pre treatment which is a clear advantage as to compared to the literature. Our method is especially effective in hidden object forgeries. All of the results of the suggested methods are compared with the current methods in literature and experimental results are reported.Keywords: Digital Image Security, Image Authentication, Copy Move Forgery, Keypoint Based Methods, SIFT, SURF, ORB, AKAZE, RANSAC, LPQ, Gabor Filter, ROC Analysis
| |||||||||||||||||||||