
Accession Number : ADA300331
Title : A Comparison of the Performance of NonParametric Classifiers with Gaussian Maximum Likelihood for the Classification of Multispectral Remotely Sensed Data.
Descriptive Note : Master's thesis,
Corporate Author : AIR FORCE ACADEMY COLORADO SPRINGS CO
Personal Author(s) : Nessmiller, Steven W.
PDF Url : ADA300331
Report Date : 20 OCT 1995
Pagination or Media Count : 256
Abstract : This study compares the performance of two nonparametric classifiers and Gaussian Maximum Likelihood (GML) for the classification of LANDSAT TM 30meter resolution sixband data. The mathematical assumptions made in developing GML are valid if the pixels that constitute the training classes are normally distributed. Since it requires a model of the data, GML is termed a "parametric" classifier. Of current interest are new classification methodologies that make no assumptions about the statistical distribution of the pixels in the training class; these approaches are termed nonparametric' classifiers. This study will compare the nDimensional Probability Density Function (nPDF) essentially a projection technique that reduces data dimensionality, and an advanced neural network that utilizes fuzzyset mathematics, the Fuzzy ARTMAP, to the traditional GML approach to image classification. The different approaches will be compared for statistical classification accuracy and computational efficiency. (AN)
Descriptors : *IMAGE PROCESSING, *NEURAL NETS, *NONPARAMETRIC STATISTICS, MATHEMATICAL MODELS, ALGORITHMS, OPTIMIZATION, COMPUTATIONS, DATA MANAGEMENT, MAXIMUM LIKELIHOOD ESTIMATION, MATRICES(MATHEMATICS), COMPARISON, RESOLUTION, ANALYSIS OF VARIANCE, ACCURACY, LEARNING MACHINES, THESES, INPUT OUTPUT PROCESSING, PARALLEL PROCESSING, PROBABILITY DENSITY FUNCTIONS, CLASSIFICATION, PATTERN RECOGNITION, COMPUTER VISION, PIXELS, REMOTE DETECTION, MULTISPECTRAL, IMAGE REGISTRATION, SYNCHRONOUS SATELLITES, NORMAL DISTRIBUTION, STATISTICAL DECISION THEORY.
Subject Categories : Cybernetics
Distribution Statement : APPROVED FOR PUBLIC RELEASE