A Scalable Hybrid Track-Before-Detect Tracking System: Application to Coastal Maritime Radar Surveillance

Abstract

Despite their theoretical advantages, track-before-detect (TBD) methods remain largely absent from real-world multi-target tracking applications due to their computational complexity and limited scalability. This paper presents a scalable hybrid tracking framework that combines a TBD multi-target tracking algorithm with a detection-based multi-target tracking algorithm for coastal radar surveillance. In particular, the approach uses an integrated existence Poisson histogram-probabilistic multi-hypothesis tracking (IE-PHPMHT)-based TBD module with a conventional Poisson multi-Bernoulli Mixture (PMBM) point tracker. The system processes raw radar data through land clutter suppression, cell-wise detection, and clustering-based feature extraction. High-threshold detections are used to track strong targets via the point tracker, while low-threshold detections are employed for adaptive birth in the TBD module, enabling early initiation and sustained tracking of weak or ambiguous targets. Validated using real X-band radar data from the Trondheim Fjord, Norway, the approach demonstrates robust multi-target tracking performance in a full-scale application with a large observation area under resource constraints, highlighting its suitability for operational deployment in complex maritime environments needed for coastal surveillance and to support autonomy.