From Traditional to Deep: A Survey of Image Forgery Detection Techniques

The detection of image forgeries is a crucial area of study within the discipline of digital image analysis, with the objective of identifying and accurately determining the location of modified regions inside images. The importance of developing strong and dependable forgery detection methods has grown increasingly significant due to the widespread availability of advanced image editing tools and the growing complexity of forgery techniques. This survey paper presents a thorough examination of the comprehensive classification of methods used for detecting image forgeries. The paper considers various factors, including the type of forgery, the detection methodology applied, the evaluation metrics used for performance analysis, and the datasets utilized. The primary objective of this paper is to stimulate further progress in the field by offering valuable insights and up-to-date information on contemporary techniques in forgery detection to researchers in the field.