Abstract

Vehicle detection and tracking is one challenge in collecting vehicle trajectories from Unmanned Aerial Vehicle (UAV). When the feature point cannot be well tracked because of the low quality of the video, many current algorithms do not work well. To solve this problem, a vehicle detection and tracking method was proposed with enhancements based on previous studies in several aspects: (1) an accurate vehicle boundary detection method using random walker segmentation to improve vehicle position accuracy, (2) a quick particle filter vehicle tracking method to handle point feature tracking low quality problem. To test the performance of the proposed method, two field experiments have been conducted. The detection and tracking rates of these experiments are 88% and 90% respectively, which is higher than the previous algorithm.

Paper Details