A Fuzzy Logic Classifier For The Detection Of Bounded Weak Echo Regions In Meteorological Images

V Lakshmanan¹
National Severe Storms Laboratory & University of Oklahoma
Norman, Oklahoma 73069
Arthur Witt²
National Severe Storms Laboratory, Norman, Oklahoma 73069

Abstract:

The presence or absence of a BWER within a storm is important for severe weather forecasting. The three dimensional structure of a BWER is often not clearly defined in radar data and a fuzzy logic approach where the various uncertainties associated with a BWER's radar profile (capping, three-dimensional structure, vertical height, reflectivity ranges, proximity to a mesocyclone, etc.) are taken into account leads to a reliable BWER detection scheme   [4].

The BWER detection scheme, however, provided crisp output in the form of BWERs and non-BWERs while meteorologists provide a fuzzy description, identifying some structures as ``marginal'' BWERs. The classifier described in this paper identifies BWERs in a way similar to that of a human observer of the same weather phenomena. We discuss the limitations of classification done by thresholding a weighted sum and introduce a fuzzy logic classifier, comparing its performance with that of the weighted-sum thresholder used in [4].

Keyword: Feature Identification And Classification. Also: fuzzy logic, remote sensing