Shoreline Extraction from Coastal Images Using Chebyshev Polynomials and RBF Neural Networks
Abstract
In this study, we use a specialized coastal monitoring system for the test case of Faro beach (Portugal), and generate a database consisted of variance coastal images. The images are elaborated in terms of an empirical image thresholding procedure and the Chebyshev polynomials. The resulting polynomial coefficients constitute the input data, while the resulting thresholds the output data. We, then, use the above data set to train a radial basis function network structure with the aid of input-output fuzzy clustering and a steepest descent approach. The implementation of the RBF network leads to an effective detection and extraction of the shoreline of the beach under consideration.
| Origin | Files produced by the author(s) |
|---|
Loading...
