HOME / RESEARCH / PUBLICATIONS /

Developing a neuro-fuzzy system to classify drainage sub-basins according to erosion processes on the Island of Lefkas (Greece).

Citation:

Gournellos T, Vassilopoulos A, Evelpidou N. Developing a neuro-fuzzy system to classify drainage sub-basins according to erosion processes on the Island of Lefkas (Greece). In: International Conference ‘Reorganizing field and landscape structures in a context of building strategies for water and soil protection’. Poland; 2005.

Abstract:

In this paper we attempt to classify drainage sub-basins according to their erosion risk. We have adopted a multi-step procedure to face this problem. The input variables were introduced into a GIS – platform. These variables are the vulnerability of the surface rocks to erosion, topographic variations, vegetation cover and land use and drainage basin characteristics. We then constructed a fuzzy inference
mechanism to pre-process the input variables. Next we used neural-network technology to process the input variables. The above system was trained to ‘learn’ and classify the input data. The output of this procedure was a classification of the sub-drainage basins related to their risk of erosion. This neuro-fuzzy system was applied to the island of Lefkas (Greece).