Research On SAR Image Segmentation Methods Besed On Fuzzy Clustering

Image segmentation is one of the key technologies in digital image processing,it is an essential area to study,and it is also a conventional research topic of computer vision.Cluster analysis is an important branch of image segmentation.It is also an important method in data mining to conduct data processing by analysis tools.In recent years,cluster analysis has been widely used and become a popular research topic for data analysis and information extraction.The Fuzzy C-Means Clustering is the most common one in clustering analysis.Since most image segmentation algorithms on SAR images use Gaussian probability function to describe the distribution of pixels in each area.However,the noises in each area follow Gamma distribution.So,in this paper,based on the theory of the FCM,authors study the problems of traditional algorithms,and practical difficulties that exist in the field of image segmentation.Several solutions for how to improve the algorithm are put forward,which focus on image processing efficiency under coherent noise.This paper consists of the.following parts:(1)The paper discusses and analyzes the research history of image segmentation,gives a complete literature review;(2)It introduces fuzzy clustering and several commonly used clustering algorithms;(3)Authors conduct fuzzy C-means algorithm with the integration of Gamma model for image segmentation;(4)Results of segmentation are analyzed using Matlab software.Performance of the segmentation algorithm is evaluated and is compared with the FCM algorithm.