Correlation Filters for Unmanned Aerial Vehicle-Based Aerial Tracking: A Review and Experimental Evaluation

Aerial tracking, which has received widespread attention and exhibited excellent performance, is one of the most active applications in the remote sensing field. In particular, an unmanned aerial vehicle (UAV)-based remote sensing system equipped with visual tracking has been widely used in aviation, navigation, agriculture, transportation, public security, and so on. The UAV-based aerial tracking platform has gradually developed from the research stage to the stage of practical application, establishing itself as one of the main aerial remote sensing technologies of the future. However, due to severe real-world situations, e.g., harsh external challenges, the vibration of a UAV’s mechanical structure (especially under strong wind conditions), its maneuvering flight in complex environments, and its limited onboard computational resources as well as its accuracy, robustness, and high efficiency are all crucial for onboard tracking methods.