Unmanned Aerial Vehicle Detection and Tracking Using Image Segmentation and Bayesian Filtering

Vehicle detection is an imperative topic to be discussed among researchers. Systems used to detect vehicles frequently rely on cameras mounted on bridges or roadsides. Whereas, aerial images provide the flexibility to use mobile platforms to detect the location and motion of the vehicle over a larger area. To this end, different models have shown the ability to recognize and track vehicles. However, these methods are not mature enough to produce accurate results in complex road scenes. Therefore, this research paper proposes a novel approach for vehicle detection and tracking in UAV (Unmanned Aerial Vehicle) imagery using pixel labeling and particle filter techniques. The proposed method first segments the input UAV imagery into meaningful regions using a pixel labeling algorithm and then extracts features from these regions to classify them as either vehicles or non-vehicles. A particle filter is then employed to track the detected vehicles across successive frames of the UAV imagery. The effectiveness of the proposed approach is evaluated on a publicly available dataset and compared with existing state-of-the-art methods. The experimental results demonstrate that the proposed approach achieves superior performance in terms of both vehicle detection and tracking accuracy. The proposed method has potential applications in traffic monitoring, surveillance, and search and rescue operations.