DefinePK

Abstract

With the development of the remote sensing imaging systems and hyperspectral sensors, the use of hyperspectral image is becoming more interesting. Remote sensing classification is a difficult procedure and requires thoughtfulness of lots of factors. The most important process of image classification may comprise resolve of a fitting classification system, selection of training samples, image pre-processing, feature extraction and accuracy assessment. Land cover information plays an important role in sustainable management, development and exploitation of resources, environmental protection, scientific analysis, modeling, monitoring and planning. Feature extraction recognizes and extracts remarkable features for a challenging task in order to decrease the complication of processing. Embedded Zero tree wavelet (EZW) are used for feature extraction process. Multi-layer perceptron (MLP) is recognized as the best ANN used in classification. The main aim of feature selection is to determine a minimal feature subset from a problem domain while retaining a suitably high accuracy in representing the original features. Particle Swarm Optimization (PSO) is a population based optimization technique used for feature selection. This paper focus on Land cover image classification using EZW algorithm. Here, MLP is an classifier and PSO used for Feature selection process.