A Task-Driven Evaluation of UAV Detection and Tracking under Synthetic Fog

2026-07-06

AI Summary

This work addresses the significant degradation in visibility and performance of small unmanned aerial vehicles (UAVs) in distant, sky-dominated scenes under foggy conditions, which severely impairs object detection and tracking. The authors propose a task-driven joint evaluation framework that synthesizes realistic foggy images by integrating monocular depth estimation with an atmospheric scattering model, and systematically couples image restoration, detection, and tracking modules. Their analysis reveals a nonlinear relationship between image restoration quality and downstream task performance: when models are trained solely on clear-weather data, applying restoration at test time yields the best results, whereas training with foggy data substantially enhances model robustness. Crucially, conventional image restoration metrics prove unreliable predictors of task performance, underscoring the necessity of task-oriented evaluation protocols.

Abstract

Fog severely degrades the visibility of small unmanned aerial vehicles (UAVs) in skydominant, long-range imagery, reducing the reliability of downstream detection and tracking. This paper presents a task-driven evaluation framework that links depth-aware synthetic fog generation, image restoration, object detection, and tracking within a unified pipeline. Given the practical difficulty of collecting and annotating foggy UAV scenes, synthetic fog is generated from real clear-weather outdoor images containing UAV targets using monocular depth estimation and the atmospheric scattering model. Representative restoration methods from classical, convolutional neural network (CNN)-based, and transformer-based families are first compared, after which the selected restoration model is integrated into the downstream perception pipeline. Detection is evaluated under both clean-only and fog-inclusive training regimes using multiple detector variants, while tracking-by-detection is assessed on clean, foggy, and restored video sequences. Beyond image-level restoration metrics, the study evaluates how fog and restoration affect detection robustness and tracking performance. The results show that fog substantially degrades both detection and tracking, primarily through increased missed detections. Fog-inclusive training provides the most consistent improvement in robustness, whereas test-time restoration is most beneficial when the detector has been trained only on clean imagery. These findings show that restoration quality does not necessarily translate into proportional gains in downstream perception and therefore should be evaluated jointly with detection and tracking performance.

Problem

• UAV detection
• UAV tracking
• synthetic fog
• visibility degradation
• task-driven evaluation

Innovation

• task-driven evaluation
• synthetic fog generation
• UAV detection and tracking
• image restoration
• depth-aware atmospheric scattering