Semi-automated training approaches for spectral class definition

Abstract

A semi-automated approach to spectral training for a maximum likelihood classification is shown to maintain or improve classification accuracy while reducing analyst input five-fold. The semi-automated approach is based on spectral sampling via region growing and training set refinement via transformed-divergence based mergers and deletions. Classification accuracies at the Anderson level II/III in northern Wisconsin were higher for the semi-automated approach in five of six combinations of imagery, analyst, and study area, at significantly reduced training expense.