Density Based Fuzzy C Means (DBFCM) Image Segmentation

Abstract: The fuzzy c means image segmentation algorithm is mainly implemented taking the attribute as intensity. Often some of the relevant segments in different density are missed as they have same intensity which is not desired. But if the segmentation is done in density domain as well as intensity domain then far better result can be obtained. Here In this paper density based fuzzy c means (DBFCM) clustering is presented. It is divided into two steps first different dense region is found using kth nearest neighbour then fuzzy c means segmentation is done on each dense region. The DBFCM is implemented upon some of the satellite images to get the segments as experimental results. Then it is compared with the conventional fuzzy c means approach.

Keywords: Density based segmentation; Image segmentation; kth nearest neighbour distance; fuzzy c means; satellite image segmentation