Digital images are foremost source for information transfer. Due to advancement of the technology, images are now not treated as reliable source of information. Digital images can be edited according to the need. Adding and deleting content from an image is most easiest and popular way of creating image forgery, which is known as copy-move forgery. Digital Image Forensics is the field that deals with the authenticity of the images. Digital image forensics checks the integrity of the images by detecting various forgeries. In order to hide the traces of copy-move forgery there are editing operations like rotation, scaling, JPEG compression, Gaussian noise called as attacks, which are performed on the copied part of the image before pasting. Till now these attacks are not detected by the single method. The novel approach is proposed to detect image forgery by copy-move under above attacks by combining block-based and keypoint-based method.
Keywords: Digital image forensics, copy-move forgery, passive blind approach, keypoint-based, block-based methods