Detecting Copy-Move Forgery in Images Using Convolutional Neural Networks (CNNs)
Alaa Bashir Zabiya, Fatima Baio Madi, Mustafa Ali Abuzaraida · 2024
Recently, digital images have become widely used in various fields, attracting the attention of researchers in the field of digital image processing. This research focuses on detecting a type of image forgery using Convolutional Neural Networks (CNN), specifically the copy-move forgery. In this forgery, a portion of the image is copied and pasted onto another part of the same image. The research proposes a CNN-based network structure for detecting copy-move forgery in images. The model is trained on different datasets previously used in this domain and then tested on a new dataset to reliably evaluate the efficiency of the proposed model. The model was trained on the (MICC-F2000) dataset and achieved an accuracy of up to 97.75%. It was also trained on the (CoMoFoD) dataset and achieved an accuracy of up to 92.85%. The results of testing the trained models on the new dataset indicate the superiority of the model trained on the (MICC-F2000) dataset. However, both models did not achieve high accuracy due to the fact that the forgery in the new dataset is unclear and difficult to detect with the naked eye.