Comparing the MLC and JavaNNS Approaches in Classifying Multi-Temporal LANDSAT Satellite Imagery over an Ephemeral River Area

摘要: This paper analyzes two pixel-based classification approaches to support the analysis of land cover transformations based on multitemporal LANDSAT sensor data covering a time space about 24 years. The research activity presented in this was carried out using Lama San Giorgio (Bari, Italy) catchment area as study case, being prone flooding proved by its geological and hydrological characteristics significant number floods occurred past. Land classes were defined accordance with CN method aim characterizing use attitude generate runoff. Two different classifiers, i.e. Maximum Likelihood Classifier (MLC) Java Neural Network Simulator (JavaNNS) models, compared. Artificial Networks (ANN) approach found be most reliable efficient when lacking ground reference priori knowledge input distribution.