A Theory of Networks for Approximation and Learning
A Theory of Networks for Approximation and Learning
Context Information for Understanding Forest Fire Using Evolutionary Computation
IWINAC '07 Proceedings of the 2nd international work-conference on Nature Inspired Problem-Solving Methods in Knowledge Engineering: Interplay Between Natural and Artificial Computation, Part II
| Hi-index | 0.00 |
Fuel moisture content (FMC) is one of the variables that drive fire danger. Artificial Neural Networks (ANN) were tested to estimate FMC by calculating the two variables implicated, equivalent water thickness (EWT) and dry matter content (DM). DM was estimated for fresh and dry samples, since water masks the DM absorption features on fresh samples. We used the “Leaf Optical Properties Experiment” (LOPEX) database. 60% of the samples were used for the learning process in the network and the remaining ones for validation. EWT and DM on dry samples estimations were as good as other methods tested on the same dataset, such as inversion of radiative transfer models. DM estimations on fresh samples using ANN (r2 = 0.86) improved significantly the results using inversion of radiative transfer models (r2 = 0.38).