Encountering false alarms for detection of point targets in highly cluttered background

Detection of point targets becomes increasingly more difficult as targets become weak and engagement takes place in highly dense, varying and complex background like clouds. To detect weak point targets in this scenario, detection threshold should be sufficiently low. And this leads to high false alarm rate. In order to make detection system robust to dense clutter (we mean 'clouds') and noise, post processing algorithms are required. Almost all detection/tracking systems use post processing techniques, but less has been reported in this area. In this paper, we propose a simple and computationally efficient post processing algorithm to encounter false alarms due to dense and varying clouds. Models for target and cloud edges are presented. Results demonstrate that proposed algorithm is able to reduce false alarms to a large extent.