MARKABLE CONTROLLED WATERSHED SEGMENTATION BASED FORGERY DETECTION USING FEATURE POINT MATCHING ALGORITHM
D. Karun Kumar, Kishore Reddy, B. Sreenivasu · IJITR International Journal of Innovative Technology and Research - IJITR International Journal of Innovative Technology and Research · 2017
A novel duplicate copy move fraud identification scheme based on Markable Controlled Watershed Segmentation and feature point matching.The current project incorporates the novelty of marker controlled watershed segmentation and also feature point matching techniques.To begin with, the arranged marker controlled Over-Segmentation equation sections the host picture into non-covering and sporadic squares adaptively.At that point, the component focuses are separated from each piece as square alternatives, and furthermore the piece choices are coordinated with each other to locate the forgery based areas; this strategy will show better results compared to the adaptive over segmentation technique as it reduces the computational complexity of the process but selectively segmenting the image based on markers generated during the segmentation process.To watch the falsification areas extra precisely, we propose the marker controlled segmentation, that replaces the component focuses with minimal super pixels as highlight pieces at that point combines the neighboring hinders that have comparable local shading choices into the element squares to think of the coordinated districts; at last, It uses the morphological techniques to an incorporated area to concoct the recognized phony locales.