Analyzing digital image manipulation detection algorithms

In recent years tampering in digital image is becoming very widespread and more interesting. With powerful Image software, editing tools make a tampered image easy, thus the field of image forensic has become very important and crucial. In this thesis, investigation is made on the image by depending on the traces that occur when the soft correlation between rows and columns are divested in altered images, in term of linear dependency and exploitation of these traces in image forensic, depending on the properties of singular value decomposition and derived features which can be helpful and accurate sufficiently to reveal the tampering that is done on the genuine digital image.