A REVIEW AND ANALYSIS OF BACKPROPAGATION NEURAL NETWORKS FOR CLASSIFICATION OF REMOTELY-SENSED MULTISPECTRAL IMAGERY

A literature survey and analysis of the use of neural networks for the classification of remotely-sensed multi-spectral imagery is presented . As part of a brief mathematical review, the backpropagation algorith m, which is the most common method of training multi-layer networks, i s discussed with an emphasis on its application to pattern recognition . The analysis is divided into five aspects of neural network classifi cation: (1) input data preprocessing, structure, and encoding, (2) out put encoding and extraction of classes, (3) network architecture, (4) training algorithms, and (5) comparisons to conventional classifiers. The advantages of the neural network method over traditional classifie rs are its nonparametric nature, arbitrary decision boundary capabilit ies, easy adaptation to different types of data and input structures, fuzzy output values that can enhance classification, and good generali zation for use with multiple images. The disadvantages of the method a re slow training time, inconsistent results due to random initial weig hts, and the requirement of obscure initialization values (e.g., learn ing rate and hidden layer size). Possible techniques for ameliorating these problems are discussed. It is concluded that, although the neura l network method has several unique capabilities, it will become a use ful tool in remote sensing only if it is made faster, more predictable , and easier to use.