Minimization of boundary curvature is a classic regularization technique for image segmentation in the presence of noisy image data. Gradient descent methods have been used for deriving the techniques for minimizing curvature which could be trapped by a local minimum and, therefore, required a good initialization. Recently, this barrier has been overcome by the combinatorial optimization techniques by providing solutions that can achieve a global optimum. However, when the true object has high curvature regularization methods can fail. In these circumstances, existing methods depend on a data term to overcome the high curvature of the object. Unfortunately, these methods fail when the data term is unambiguous in some images. So in order to eliminate these problems, we propose a contrast driven elastic model (including curvature), which can accommodate high curvature objects and an ambiguous data model. We demonstrate that we can accurately segment extremely challenging synthetic and real images with ambiguous data discrimination, poor boundary contrast, and sharp corners. By using this method we have provided a quantitative evaluation of our segmentation approach when these are applied to a standard image segmentation data set.
Euler elastic, weighted curvature, combinatorial optimization, primal formulation, image segmentation.
MADHAVI LATHA PARSA , SINDHU MARUMUDI , NAVYA KODURU , AKHILA PATIBANDLA , M. R. N. TAGORE "CONTRAST DRIVEN ELASTICA FOR IMAGE SEGMENTATION" Iconic Research And Engineering Journals Volume 1 Issue 8 2018 Page 95-102
MADHAVI LATHA PARSA , SINDHU MARUMUDI , NAVYA KODURU , AKHILA PATIBANDLA , M. R. N. TAGORE "CONTRAST DRIVEN ELASTICA FOR IMAGE SEGMENTATION" Iconic Research And Engineering Journals, 1(8)