Line-shaped non-precipitation echo detection using fuzzy inference system

In weather forecasting process, it is important to analyze weather radar data for classifying precipitation and non-precipitation echoes due to increase forecasting accuracy. There are lots of ongoing researches about non-precipitation echoes for detecting and removing them. In this paper, we present an algorithm to detect and remove line-shaped non-precipitation echo. The line-shaped echo can be divided into sun strobe echo that occurred by solar energy at sunrise or sunset and radial interference echo that occurred by electromagnetic interference wave. In order to detect and remove these two echoes, Mamdani fuzzy inference system is applied to combine expert´s knowledge and characteristics. We tested the algorithm by applying the actual line-shaped echoes. And the result shows that the fuzzy inference system can distinguish not only sun strobe echo but also radial interference echo from actual radar data.