A Thresholding Algorithm Based on Probabilistic Rough Set for Image Segmentation

Abstract

A new thresholding algorithm and its multi-threshold extension are presented to improve the performance of image segmentation. The probabilistic Rough set is used to portray the objects and backgrounds in image. The optimal threshold is given through a minimum boundary criterion. A new rough membership function is defined in the light of requirements of image thresholding.2008 International Symposium on Computational Intelligence and Design 978-0-7695-3311-7/08 $25.00