A fuzzy partitioning method of spectral space for remote sensing image classification

Abstract

The aim of this study is to propose an efficient method for partition of spectral space into fuzzy subspace for multi-spectral remote sensing image. The suggested method predicates on sequential subdivision of the fuzzy subspace, and the size of constructed fuzzy space is variable. Under this procedure, n-dimensional pattern space, after considering the distributional characteristic patterns, is partitioned into two different fuzzy subspaces. From the two fuzzy subspaces, the pattern space for further subdivision is chosen; then, this subdivision procedure recursively repeats itself until the stopping condition is fulfilled. The result of this study is applied to 2, 4, 7 band of satellite Landsat TM and satisfactory result is acquired.<>