An Algorithm of Detecting Moving Target with Parabolic Track Using Coordinate Transform and Randomized Hough Transform

This paper proposes a new algorithm of detecting target with constant acceleration along parabolic track on the radar coordinate plane. This algorithm combines coordinate transformation of the parabolic equation and three-point detection algorithm based on randomized hough transform to detect parabolic tracks in any orientation, and employs the velocity information of detected tracks to delete false alarms. Monte Carlo simulation shows that higher detection probability goes with higher false alarm probability in the background of heavy clutters; however in the background of sparse clutters, the detection performance is better.