ART neural networks for remote sensing: vegetation classification from Landsat TM and terrain data

Abstract

A new methodology for automatic mapping from Landsat Thematic Mapper (TM) and terrain data, based on the fuzzy ARTMAP neural network, is developed. System capabilities are tested on a challenging remote sensing classification problem, using spectral and terrain features for vegetation classification in the Cleveland National Forest. After training at the pixel level, system capabilities are tested at the stand level, using sites not seen during training. Results are compared to those of maximum likelihood classifiers, as well as back propagation neural networks and K Nearest Neighbor algorithms. ARTMAP dynamics are fast, stable, and scalable, overcoming common limitations of back propagation, which did not give satisfactory performance. Best results are obtained using a hybrid system based on a convex combination of fuzzy ARTMAP and maximum likelihood predictions. Fuzzy ARTMAP automatically constructs a minimal number of recognition categories to meet accuracy criteria. A voting strategy improves prediction by training the system several times on different orderings of an input set. Voting assigns confidence estimates to competing predictions.