A Hough transform track initiation algorithm for multiple passive sensors
Kiril M. Alexiev, Ljudmil Bojilov · 2000
This paper is concerned with the problem of associating measurements from several passive (angle-only) sensors. In contrast with the static problem when measurements at a given time are processed we present a technique, processing measurements from several consecutive scans. We use a particular feature of the Hough transform for batch data processing. The technique is derived assuming presence of clutter, missed detections and an unknown number of targets. An effective heuristic technique for ghost elimination is proposed which preliminary reduces considerably the total number of so called ghost intersections and thus alleviating consequent Hough transform algorithm implementation. This Hough transform algorithm, as it can be seen from numerical results presented, initiates tracks and successfully resolves the ambiguous measurement-target associations. The main advantage of the presented approach is avoiding the S-D matching problem which is known to be NP-hard.