Segmenting natural and outdoor images are challenging for most of the latest variational segmentation models. For this purpose we employ derived image data (DID) and propose a robust variational model. The DID rely on three images by utilizing image local and global statistics as well as filter image which is obtained through our design high pass filtering techniques. Then these derived image data are incorporated into our proposed energy functional which can robustly segment images having inhomogeneity, mix backgrounds and multi-regions. Furthermore, the results of DID are compared with other well known methods with finding Jaccard similarity index to proof the efficient and qualitative performance of proposed model over the traditional methods. Finally, the proposed
DID based model is tested on real world 3D images to ensure that it also
preserve its performance in vector valued images as well.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.