Stationary and Small Target Detection for Millimeter-Wave Radar

Abstract

Using millimeter-wave radar to scan and detect stationary and small foreign object debris (FOD) on an airport runway surface is a popular solution in civil aviation safety. In this paper, we proposed a novel FOD detection method based on pattern classification theory using bi-spectral features. Firstly, a non-parameter weighted generalized matched filtering (WGMF) is utilized to accomplish clutter suppression with low false alarm rate. Then low dimensional bi-spectral features are extracted from radar returns which are utilized to form the feature vector. Finally, support vector data description (SVDD) is used to accomplish FOD detection. Real airport data measured by 77GHz radar are used to validate the proposed method. Experimental results using a golf ball with a diameter of 43mm show that the proposed method can effectively detect the target with low false alarm rate.