An Improved Track-before-Detection Algorithm Based on Dynamic Neighborhood Search

Track-Before-Detection (TBD) technology can effectively improve the detection and track performance of weak targets by exchanging energy through time. However, the target with glint effect and angular noise is easy to disappear in more than one of frames. To solve this problem, this paper proposes an improved TBD algorithm based on dynamic neighborhood search. It uses the dynamic neighborhood to keep tracking targets, which needs less computational load than the traditional TBD algorithm. The algorithm is validated by simulation scenes with real target data. The results show that this proposed algorithm can improve the performance of tracking weak targets at a low Signal to Noise Ratio (SNR) about -5dB and it can effectively detect and track those targets with glint effect and angular noise.