Feature extraction and recognition of two-dimensional data by the method of moments
Date Issued
August 1976
Author(s)
Harris, James Micheal
Advisor(s)
R. C. Gonzalez
Additional Advisor(s)
R. E. Bodenheimer
Michael G. Thomason
Abstract
A pattern recognition system applicable to two dimensional data is presented, and training algorithms for generating pattern classifiers are surveyed. The method of moments is used by the system as a feature extractor. The Mahalanobis distance measure is presented as a criterion for the selection of moment pairs to be used as descriptors. Experiments conducted using high resolution radar images examine the effectiveness of the system using unstructured data. Classification results for the system are compared to those of human interpreters.
Degree
Master of Science
Major
Electrical Engineering
File(s)![Thumbnail Image]()
Name
Thesis76.H2766.pdf
Size
5.81 MB
Format
Adobe PDF
Checksum (MD5)
41edcaa887d1f86002c13be44ac430f0