Algorithms for optical weak small targets detection and tracking: review

This paper provides an extensive review of algorithms currently used for detection and tracking in optical weak small targets environments where conventional tracking techniques are inadequate. Firstly, the constructions of point source target and detection model are introduced. Then, the algorithms of background clutter suppression (temporal filter, spatial filter, and three-dimensional matched filter algorithms) are reviewed. Finally, the concepts of track-before-detect (TbD) are reviewed and the major existing approaches are described, similarities and difference between the various approaches (velocity filter, three-structured multistage hypothesis testing, frequency domain, dynamic programming, and optimal nonlinear filter algorithms) are highlighted.