Classification of IRS LISS-III Images Using PNN

Remote Sensing is widely used for mapping of land cover and land use. Classification of image satellites is also done by using these mapping. In this paper the classifier proposed is the Probabilistic based Neural Network developed using MATLAB. The data for image classification is acquired over various parts of Mumbai region which is LISS-III. Probabilistic based neural network is a supervised classification technique applied on the LISS-III satellite images. The use of artificial neural techniques was very efficient. Neural networks have shown great scope for image classification. Hence Probabilistic Neural Network has been applied. This algorithm also gave fast and accurate classification. The classification accuracy is calculated after applying the PNN artificial neural network using the Confusion matrix and Kappa co-efficient. The classification accuracy of the proposed classifier is 99.83% and the Kappa coefficient is 0.9975 respectively.