ROAD IDENTIFICATION IN LANDSAT THEMATIC MAPPER IMAGERY USING PULSE-COUPLED NEURAL NETWORKS: AN INITIAL ASSESSMENT

Classifying roads in remotely sensed imagery has been addressed by a number of research efforts. Detecting these features is important for a variety of endeavors such as agricultural assessment and urban planning. This study investigates the viability of using a pulse-coupled neural network to recognize roads in Landsat-4 Thematic Mapper multispectral imagery.