Guided filter-driven kernel fuzzy clustering with local information for noise image segmentation

Fuzzy local information clustering is the most widely robust segmentation methods, but it is only suitable for image corrupted by certain intensity noise. Later, although fuzzy local information clustering integrated guided filter is improved the ability of suppressing noise, it still cannot meet the needs of image with high noise. This paper proposed a novel robust fuzzy local information clustering combined kernel metric with guided filter. Firstly, guided filter is introduced into fuzzy local information clustering with kernel metric (KWFLICM), and a novel multiple objective optimization model for fuzzy clustering is constructed. Secondly, the optimization model is solved by Lagrange multiplier method, and the iterative algorithm for image segmentation is presented. Experimental results show that the proposed algorithm has better segmentation performance and robustness than existing state of the art guided filter-driven fuzzy clustering with local information.