Statistical Features based Optimized Technique for Copy Move Forgery Detection
CMFD method is proposed using Steerable Pyramid Transform, Grey Level Co-occurrence Matrix and Optimized Naive Bayes Classifier and shows robustness over existing algorithms in the literature even the forged image has undergone many attacks.
Abstract
Copy Move Forgery Detection (CMFD) is helpful to detect copied and pasted areas in one image, it plays a crucial role in legal evidence, forensic investigation and in many more places. In this paper, CMFD method is proposed using Steerable Pyramid Transform (SPT), Grey Level Co-occurrence Matrix (GLCM) and Optimized Naive Bayes Classifier (ONBC). The suspected image is given to SPT to obtain different orientations, from all suspected image orientations GLCM features are extracted. These features are used to train ONBC as well as to classify ONBC. Wide range of tests conducted on CoMoFoD, MICC_F and CASIA v1.0 databases using proposed algorithm and performance is measured in terms TPR and FNR. It shows robustness over existing algorithms in the literature even the forged image has undergone many attacks.