@article{Chen2026, 
author = {Li Chen and Hongbin Deng and Qizhi Xu and Xiaolin Han},
title = {Hierarchical Supervised Network for Multi-Object Tracking in Reconnaissance UAV Video under Foggy Weather},
year = {2026},
journal = {Tsinghua Science and Technology},
keywords = {hierarchical supervised architecture, reconnaissance unmanned aerial vehicle (UAV), multi-object tracking (MOT), salient foreground perception (SFP), multi-scale object detection (MSOD)},
url = {https://www.sciopen.com/article/10.26599/TST.2025.9010055},
doi = {10.26599/TST.2025.9010055},
abstract = {Multi-object tracking in reconnaissance unmanned aerial vehicle (UAV) videos under foggy weather presents several challenges: (1) The boundaries between the target and the background become blurred; (2) targets have limited target pixels and few available features; (3) reconnaissance UAV movement causes variations in the shape and size of targets. We tackl these challenges by developing a hierarchical supervised network for multi-object tracking (HSTrack). First, we employ the salient foreground perception (SFP) module to extract foreground targets from blurry backgrounds. SFP is incorporated to enhance the limited pixel objects and increase foreground-background contrast. Second, we utilize multi-scale object detection (MSOD) to predict the objects with scale variations. Meanwhile, to implement hierarchical supervision, we devise a unified loss function to compute the loss of the prediction between MSOD and SFP. Finally, we use mixed metrics to connect detections to tracklets, achieving multi-object tracking. We validate our approach through experiments on a self-assembled dataset collected from a reconnaissance UAV under foggy weather, named UAV-Fog. The experimental results demonstrate that our proposed method outperforms other state-of-the-art methods in reconnaissance UAV videos under foggy weather. Furthermore, we validate the effectiveness of HSTrack on the vision meets drone multi-object tracking (VisDrone2019-MOT) and UAV detection and tracking (UAVDT) datasets, where it also achieved outstanding performance.}
}