Copy and Move Forged Image Detection by Deep Learning

Copy-Move forgery detection (CMFD) is one of the most active study topics in blind image forensics. Most known algorithms are based on block and key-point approaches or a mix. Recently, various deep convolutional neural network approaches have been utilised in image classification, image forensics, image hashing retrieval, and other areas, and they have outperformed the previous way. The paper proposes a unique copy-move forgery detection system based on a convolutional neural network. The suggested technique takes a trained model from an extensive database such as ImageNet and minimally modifies the net structure using tiny training examples. The experimental findings indicate that the approach we presented outperforms the counterfeit picture created automatically by the computer using a simple image copy-move operation.