A paper on counter-drone technology by Professor Xu Tingfa's research team at Beijing Institute of Technology (BIT) has become a highly cited work in the Essential Science Indicators (ESI) database of the Web of Science.

As of March/April 2025, it ranks in the top 1% of engineering papers from the same year, highlighting its significant academic impact and global recognition.

The paper, titled Anti-UAV410: A Thermal Infrared Benchmark and Customized Scheme for Tracking Drones in the Wild, was published in IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE TPAMI). This journal is issued by the Institute of Electrical and Electronics Engineers (IEEE) and focuses on areas such as artificial intelligence, pattern recognition, and image processing. It is recognized as one of the most authoritative international journals in the field. The journal has a five-year impact factor of 20.4. Additionally, IEEE TPAMI is recommended by the China Computer Federation (CCF) as an A-tier journal in artificial intelligence.

The research team addressed the issue of a lack of high-quality benchmark datasets and adaptive tracking algorithms for infrared small target drone tracking in complex outdoor environments by constructing, for the first time, the large-scale thermal infrared anti-drone video dataset, Anti-UAV410. This dataset encompasses typical application scenarios such as urban areas, forests, and mountainous regions, and includes real-world challenging factors like occlusion, interference, and low signal-to-noise ratio, significantly enhancing the practical representativeness and evaluation value of anti-drone research.

Additionally, to tackle the challenges of small target scale and contextual clutter in anti-drone locating and tracking, the team proposed a new method called the SiamDT.

The paper innovatively established a benchmark for tracking drones in the wild, effectively supporting the monitoring and defense against illegal drone intrusions. The newly proposed tracking method significantly outperforms existing deep learning models in performance and enhances the understanding of the internal decision-making mechanisms of Siamese networks, providing strong support for their application in anti-drone tracking systems.

This achievement also won the championship at the ICCV 2021 Anti-UAV Workshop & Challenge, highlighting its impact and effectiveness in the field.

Paper details: B. Huang, J. Li, J. Chen, G. Wang, J. Zhao and T. Xu, "Anti-UAV410: A Thermal Infrared Benchmark and Customized Scheme for Tracking Drones in the Wild," in IEEE Transactions on Pattern Analysis and Machine Intelligence, doi: 10.1109/TPAMI.2023.3335338.