Ph.D. Tezi Görüntüleme

      Digital images, importance of which is increasing gradually in daily life, have a convincing effect on people and are even used as evidence in courts. The widespread use of software that enables tampering digital images has led to the spread of digital image forgeries. As results of these, an obligation has arisen to verify the authenticity of digital images. In the literature, there are many forgery detection techniques about detection of the copy move image forgeries, performed on digital images. However, the main deficiencies of these methods are; they are easily affected by the factors that make the forgery detection difficult, they cannot make discrimination between the source region and the destination region in forgery detection and they are usually tested against amateur forgeries that are understandable even with the human eye. Within the scope of this thesis; forgery detection methods which can focus on the whole of the forgery on the image, which are affected as little as possible by the post-processes that make it difficult to detect forgery, which can make discrimination between the source region and the destination region in the detection of forgery, which are about the detection of professional forgeries that cannot be understood by the human vision system, and which have high success rates have been developed.