In light of the cumulative use of digital images in a wide range of apps, as well as the accessibility ofimage manipulation software, detecting image alteration has become a difficult task. The copy-movetechnique is the most commonly used sort of picture counterfeiting. A portion of an image is duplicatedand manipulated in various ways. Handcrafted qualities are commonly used in the identification ofpicture forgery and counterfeiting. It has always been a difficulty with earlier photo reproductiondetection systems that they would rather only detect a specific type of tampering if they are aware ofspecific image features. Deep learning is currently being used to detect image alteration, which is abreakthrough. These strategies were even more efficient than prior methods since they were able toextract complex images from them. The purpose of this work is to teach you about deep learning-basedpicture forgery detection methods, why they function, and how they can be improved upon. In addition,you will learn about publicly available image forgery datasets.