Digital Image Forgery Detection Using Pre-Trained Xception Model as Feature Extractor
Niyantha Maruthu Pandiyan · 2023
During this time of social network bloom, there has been a tremendous increase in the number of images shared over the internet. This along with the advancements in software, like photoshop, has made morphing or tampering with the original images very easy. This paper proposes a Deep Learning approach to tackle this problem and detect Fake or factored images. In this paper, we propose a method where a pre-trained Xception Model is used as the feature extractor to classify a certain image as Tampered or Real. The proposed model was trained on the CASIA-V2 dataset. The model was able to attain good Accuracy on the Test Split.