A Fuzzy Based Approach to Classify Remotely Sensed Images

Classification of Images is one of the challenging tasks in image analysis. Image classification is used in many fields such as Remote sensing, medical diagnosis, robotics, etc. Classification is to identify homogeneous groups of data points in a given dataset and assigning it to a class. In this paper classes of image objects are to be classified as region or area of interest for the land use/land cover types. Different classification techniques available categorize all pixels in a Multispectral image. An attempt had been made to analyse the performance of supervised classification and unsupervised classification methods for RSI images using Fuzzy. Experimental study revealed that the proposed supervised classification approach provides consistently better result than unsupervised classification.