An Adaptive Technique for Regularized Level Set Evolution to Image Segmentation
Asst. Professor, Dept. of ECE, DIET, Anakapalle, A.P., Kusumanchi Avinashkumar, SatyaSwathi lapudi · IOSR Journal of Electronics and Communication Engineering · 2014
In this work numerical therapy, called reinitialization which is intrinsically maintained during the level set evolution.This is applied to periodically replace the degraded level set function with a signed distance function because due to the development of irregularities in level set functions.This yields a new type of level set evolution called distance regularized level set evolution (DRLSE).The distance regularization effect eliminates the dictate for reinitialization and thus avoids its induced statistical errors.In dissimilarity to knotty implementations of straight level set formulations, a simpler and more resourceful finite divergence design can be used to execute the DRLSE formulation.DRLSE also allows the employ of more broad and capable initialization of the level set function.In its numerical implementation, somewhat large time steps can be used in the finite difference scheme to reduce the number of iterations, while ensuring sufficient statistical precision.To reveal the efficacy of the DRLSE formulation, we apply it to an edge-based active contour model for image segmentation, and provide an easy narrowband realization to significantly diminish computational cost.