The Community for Technology Leaders
Green Image
Issue No. 01 - January (1980 vol. 2)
ISSN: 0162-8828
pp: 72-75
J. O. Eklundh , Defense Research Institute, Stockholm, Sweden.
A. Rosenfeld , Computer Science Center, University of Maryland, College Park, MD 20742.
H. Yamamoto , National Aerospace Laboratory, Tokyo, Japan.
ABSTRACT
Three approaches to reducing errors in multispectral pixel classification were compared: 1) postprocessing (iterated reclassification based on comparison with the neighbors' classes); 2) preprocessing (iterated smoothing, by averaging with selected neighbors, prior to classification); and 3) relaxation (probabilistic classification followed by iterative probability adjustment). In experiments using a color image of a house, the relaxation approach gave markedly superior performance; relaxation eliminated 4-8 times as many errors as the other methods did.
INDEX TERMS
CITATION
J. O. Eklundh, A. Rosenfeld, H. Yamamoto, "A Relaxation Method for Multispectral Pixel Classification", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 2, no. , pp. 72-75, January 1980, doi:10.1109/TPAMI.1980.4766973
99 ms
(Ver )