Hyper Spectral Image Analysisinagriculture

Abstract

Satellite image classification process involves grouping the image pixel values into meaningful categories. Several
satellite image classification methods and techniques are available. In existing k-means
clusteringtechniqueisusedforclusteringthesatellitedata, with this method not able to cluster accurately all the
classes.Inourproposed methodself-organising mapsasa clusteringtechniqueisused.Self-organizingmapslearnto
cluster data based on similarity, topology, with a preference of assigning the same number of instances to each class.
Self-organizing maps are used both to cluster data and to reduce the dimensionality of data. They are inspired by the
sensory and motor mappings in the mammal brain, which also appear to automatically organizing information
topologically. CNN Classifiers meld results from many weak learners into one high- quality ensemble model. Our
proposed technique is K- Medoid clustering with CNN algorithm for classification
ofsatellitedataintowater,Agriculture,Barrenland,Green Land. The proposed method of self-organising map
clusteringandCNNclassifierisgivenbestresultcompared to existing ones.