Real Time Multiple Target Tracking Using Arrayed Sensors

This paper presents a discrete time-step optimization algorithm for the real time tracking of multiple targets using an array of sensors. The proposed method is based on a limited set of parameters characterizing the sensor-target system and introduces the concept of a covering graph to model the spatial relationships between sensors and detected targets. An iterative solution is first derived for the case of a single sensor, which is then extended to accommodate multiple sensors through the use of a flag matrix that coordinates the tagging and assignment operations across the sensor array. The algorithm has been implemented using the OpenCV library and experimental evaluation demonstrates that the proposed approach achieves superior performance compared to other notable methods in the literature, particularly in scenarios involving occlusion and target crossover.