发表机构
Technological University of Uruguay（乌拉圭技术大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究以轻量YOLO模型为对象，探究数据集组成对嵌入式实时无人机野火检测的影响，发现真实未增强数据集的检测性能最优，合成数据混合与图像增强均未提升部署效果，数据集真实性和领域对齐更具价值。
AI 中文摘要
基于视觉的无人机野火检测系统的开发受限于多样化真实世界训练图像的有限可用性。本文以轻量YOLO模型作为受控验证系列，研究了数据集组成对嵌入式实时无人机野火检测的影响。评估了四种训练配置：真实未增强、真实增强、混合未增强、混合增强，其中混合数据集将真实野火图像与AI生成样本相结合。研究目标是确定合成数据混合和图像增强是否能在资源受限的部署条件下提高实际检测性能。实验结果表明，真实未增强数据集获得了最佳整体操作点，在无人机野火检测的召回率和平均精度之间实现了最强平衡。结果还显示，与合成数据混合或图像增强均未产生更好的最终部署选择。这些发现表明，对于嵌入式无人机野火检测，数据集的真实性和领域对齐比通过合成扩展增加训练集规模更有价值。
英文摘要
The development of vision-based wildfire detection systems for unmanned aerial vehicles is constrained by the limited availability of diverse real-world training images. This paper investigates the impact of dataset composition on embedded real-time UAV wildfire detection using compact YOLO models as a controlled validation family. Four training configurations were evaluated: real non-augmented, real augmented, hybrid non-augmented, and hybrid augmented, where the hybrid sets combine real wildfire images with AI-generated samples. The objective is to determine whether synthetic data mixing and image augmentation improve practical detection performance under resource-constrained deployment conditions. Experimental results show that the best overall operating point was obtained with the real non-augmented dataset, which achieved the strongest balance between recall and mean average precision for UAV-based wildfire detection. The results also show that neither hybridization with synthetic data nor augmentation produced a better final deployment choice. These findings suggest that, for embedded UAV wildfire detection, dataset realism and domain alignment are more valuable than increasing training set size through synthetic expansion.
CommentsPaper accepted at the ICCAS 2026
