Analysis of Global Kernels Using Fuzzy C Means Algorithm

Today the field of remote sensing has become exciting and glamorous with rapidly expanding opportunities. Many organizations spend large amount of money on these fields. Two main reasons why these fields are so important in recent years are: 1) Now-a-days scientists, researchers, students, and even common people are showing great interest for better understanding of our environment i.e., the geographic space of their study area and the events that take place there. 2) With the development in sophisticated space technology (which can provide large volume of spatial data), along with declining costs of computer hardware and software has made Remote Sensing affordable to not only complex environmental/ spatial situation but also affordable to an increasingly wider audience. Now a day it is a tough task to handle large amount of remote sensing data and to obtain information relating land cover mapping. Classification of images is one of the important process in this area. Effective use of various features of remotely sensed data and the selection of suitable classification method are especially important for improving classification accuracy. This paper explains how remote sensing data with uncertainty are dealt with fuzzy based classification using kernel approach. Here we mainly study FCM with global kernels.