2018/02/09 by Tajuddin Manhar Mohammed, Mohammed, Tajuddin Manhar, Jason Bunk +15
Biochemistry, Genetics and Molecular Biology · Computer Science · #Cell Image Analysis Techniques #Computer Vision and Pattern Recognition (cs.CV) #Digital Media Forensic Detection #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis
paper · pdf · doi:10.48550/arxiv.1802.03154
openalex publication_date 2018/02/09 · openalex created_date 2022/09/30 · openalex updated_date 2026/07/28
Realistic image forgeries involve a combination of splicing, resampling,\ncloning, region removal and other methods. While resampling detection\nalgorithms are effective in detecting splicing and resampling, copy-move\ndetection algorithms excel in detecting cloning and region removal. In this\npaper, we combine these complementary approaches in a way that boosts the\noverall accuracy of image manipulation detection. We use the copy-move\ndetection method as a pre-filtering step and pass those images that are\nclassified as untampered to a deep learning based resampling detection\nframework. Experimental results on various datasets including the 2017 NIST\nNimble Challenge Evaluation dataset comprising nearly 10,000 pristine and\ntampered images shows that there is a consistent increase of 8%-10% in\ndetection rates, when copy-move algorithm is combined with different resampling\ndetection algorithms.\n