Survey: Neural Network Authentication and Tampering Detection

Abstract

Neural networks have become quite the buzzword in a decade, resulting in extensive research and extensive integration of neural networks in application development. From self-driving vehicles to IoT devices, each such area has seen some form of integration of a neural network(s). Image and video content have found application in medical, forensic, etc. Due to the excessive use of digital content, there has also been a rise in various advanced image editing applications such as Photoshop, making it easier for people to tamper with images. Therefore, coming up with techniques to validate or authenticate images has gained much interest in recent times. Current neural network-based methods can see all kinds of tampering because neural network capability extracts complex features from the images, making them more effective. Thus, in this study, we review some image forgery techniques and look over how neural networks find their application to detect forgery and authenticate images. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.