Vegetation Indices (VIs) represent a useful method for extracting vegetation information from satellite images. Erosion models like the Revised Universal Soil Loss Equation (RUSLE), employ VIs as an input to determine the RUSLE soil Cover factor (C). From the standpoint of soil conservation planning, the C factor is one of the most important RUSLE parameters because it measures the combined effect of all interrelated cover and management variables. Despite its importance, the results are generally incomplete because most indices recognize healthy or green vegetation, but not senescent, dry or dead vegetation, which can also be an important contributor to C. The aim of this research is to propose a novel approach for calculating new VIs that are better correlated with C, using field and satellite information. The approach followed by this research is to state the generation of new VIs in terms of a computer optimization problem and then applying a machine learning technique, named Genetic Programming (GP), which builds new indices by iteratively recombining a set of numerical operators and spectral channels until the best composite operator is found. Experimental results illustrate the efficiency and reliability of this approach to estimate the C factor and the erosion rates for two watersheds in Baja California, Mexico, and Zaragoza, Spain. The synthetic indices calculated using this methodology produce better approximation to the C factor from field data, when compared with state-of-the-art indices, like NDVI and EVI.
Design of a computer tool for the optimization of solid waste collection circuits
