Adaptive Fuzzy Connectedness-Based Medical Image Segmentation.
Amol S. Pednekar, Ioannis A. Kakadiaris, Uday Kurkure · 2002
In this paper, we present an enhancement of the fuzzy connectedness-based image segmentation method based on dynamic computation of adaptive weights for the homogeneity and the directional gradient energy functions. Adaptive weights enhance the performance and robustness of the conventional fuzzy connectedness-based segmentation while decreasing the degree of user interaction. The accuracy of the adaptive fuzzy connectedness-based segmentation of typical fuzzy medical images is assessed with respect to the conventional fuzzy connectedness.