A Literature Review on IMAGE FORGERY DETECTION

Abstract

Taking pictures has grown in popularity recently as cameras are so widely accessible. Since they are so rich in information, images are crucial to daily life.Pictures frequently need to be enhanced in order to gain more information due to their wealth of data. Although there are many technologies available to enhance picture quality, they are also regularly used to alter photos, which leads to the dissemination of false information. This makes picture forgeries more severe and frequent, which is now a major cause of worry. To identify fake images, several conventional methods have been developed over time. CNNs have drawn a lot of interest recently, and CNN has also had an impact on the area of picture forgery detection. In recent years, CNNs have gained great attention, and CNN
has also affected the field of picture fraud detection. The majority of CNN-based picture forgery detection methods, however, are
only capable of spotting one kind of fraud (either image splicing or copy-move). Hence, a novel method that can quickly and precisely identify any hidden forgeries in a picture is needed. In the context of double image compression, the suggested system is a strong deep learning-based system that is introduced for detecting picture forgeries. The suggested model is trained using the variation between the original and recompressed versions of a picture.