Dynamic Robust Suppressed Ifcm Image Segmentation Algorithm Based on Spatial Information and Texture Features

As the IFCM algorithm does not consider the texture information, spatial information of the image and the antinoise performance, this paper proposes a dynamic robust IFCM image segmentation algorithm based on spatial information and texture features. Firstly, the neighborhood pixels are updated by the deep neighborhood information which is defined in this paper, and then the image texture information is extracted by using the Intuitionistic Fuzzy Local Binary Pattern (IFLBP) operator algorithm with an adaptive hesitation threshold <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$h$</tex>. At the same time, the Jac-coefficient between two image blocks is proposed to introduce image texture information into the objective function of IFCM, and the segmentation accuracy can be improved by introducing the image texture information; Secondly, according to each pixel's membership and hesitancy, the suppressed factor is dynamically defined to make it more reasonable; Finally, in order to improve the robustness, the new intuitionistic fuzzy local information factor is defined by combining the spatial term and the gray scale term. With different noise levels, the natural images and the SAR images are utilized for comparison tests, and the experimental results show that the proposed algorithm has better segmentation results in both visual and evaluation indexes.