Fuzzy Categorization of Weather Conditions for Thermal Mapping

Abstract

Thermal mapping is a technique that uses a vehicle-mounted infrared radiometer to measure the variation of road surface temperature (RST). Conventionally, the technique is conducted under three qualitatively categorized weather conditions: extreme, intermediate, and damped. These three categories represent basic weather patterns and are widely used in thermal mapping. In real-time operation, however, determination of the weather category is hampered by the lack of systematic classification. Furthermore, certain skills and knowledge of both thermal mapping and meteorology are required. As the thermal mapping technique develops in the direction of providing a platform for automatic and dynamic forecasting of RST over an entire road network, it is necessary to have some kind of hands-off, quantitative, systematic, accurate, and fast categorization of weather conditions for thermal mapping. For this purpose, the relationship between the change of weather conditions and variation of RST was analyzed to define a time domain for application of a reliable categorization algorithm. Fuzzy membership functions were then established, based on cloud amount, cloud type, wind speed, and relative humidity, to compose a fuzzy function of weather categorization for thermal mapping. The results of validation for the fuzzy categorization show that the algorithm can become a useful tool for thermal mapping.