A Novel Method of Image Forgery Detection using Hybrid Feature Extraction and SVM

  • J. Gomathi, Rajasekararan, Laavanya mohan, Thirumalraj karthikeyan


            Presently a day’s Image fabrications makes numerous issues in the general public. To distinguish the phonies in the pictures Modified Super Pixels Segmentation and Modified Scale-Invariant Feature Transform based Support Vector Machine (MSPS-MSIFT-IFD-SVM) has presented. The fabrication area extraction calculation replaces the highlights point with little MSPS as highlight squares and neighboring squares have comparable nearby shading highlights that is distinguished by the MSIFT system. Bolster Vector Machine (SVM) is utilized for well discovery of fraud slices in the first pictures. It given to the morphological activity to the blended areas to create the identified phony locales. The exploratory outcomes demonstrate that the MSPS-MSIFT-IFD-SVM technique can accomplish worthy location results even under different testing circumstances contrasted with the current falsification identification strategies. MSPS-MSIFT-IFD-SVM technique give in expanding distinctive execution, for example, True positive (TP), True Negative (TN), False Positive (FP), Precision (P), False Negative (FN) Recall (R), Similarity (S), False measure (FM), False Positive Rate (FPR), True Positive Rate (TPR) and more exactness.

How to Cite
Thirumalraj karthikeyan, J. G. R. L. mohan,. (2020). A Novel Method of Image Forgery Detection using Hybrid Feature Extraction and SVM. International Journal of Advanced Science and Technology, 29(3), 2971- 2982. Retrieved from http://sersc.org/journals/index.php/IJAST/article/view/4515