基于地基天空图像分类的标签高效学习：GCD上迁移学习、主动学习与伪标签的基准研究
Label-Efficient Learning for Ground-Based Sky-Image Classification: A Benchmark of Transfer Learning, Active Learning, and Pseudo-Labeling on GCD
- Université Paris Dauphine-PSL（巴黎第九大学）
- Anglia Ruskin University（安格利亚鲁斯金大学）
- Polish Academy of Sciences（波兰科学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究在GCD数据集上基准测试迁移学习、主动学习和伪标签三种策略，发现迁移学习在少量标签下即可接近全标签性能，而主动学习和伪标签增益有限。
AI中文摘要：
准确的地基云分类对于大气监测、太阳能预测、航空天气评估和气候观测系统至关重要。然而，可靠的天空图像标注非常耗时，尤其是当云类型在视觉上相似或混合时。我们利用地基云数据集（GCD）研究了深度学习在地基云分类中的标签效率。我们没有提出新的架构，而是在有限的标注预算下对三种实用策略进行了基准测试：监督迁移学习、基于不确定性的主动学习和高置信度伪标签。使用ImageNet预训练的ResNet50作为共享的冻结骨干网络，实验在五个随机种子上重复进行，标签预算从训练标签的1%到100%。监督迁移学习已经具有很高的标签效率：测试准确率从1%标签时的0.635±0.018提高到40%标签时的0.730±0.002，接近全标签结果0.735±0.003。主动学习和伪标签与监督采样相比具有竞争力，并在某些指标和预算下提供了小幅改进，但两者都没有带来显著或一致的总体增益。诊断分析表明，接受的伪标签是可靠的，准确率从0.946到0.977，但偏向于更容易的高置信度天空类型组。相比之下，不确定性采样优先查询视觉上具有挑战性的组，包括混合组以及易混淆的层积云和积雨云组，但这些有针对性的获取仅带来了适度的改进。总体而言，迁移学习大幅减少了GCD的标注需求，而简单的主动和半监督策略在强监督基线上提供的额外收益有限。
英文摘要：
Accurate ground-based cloud classification is important for atmospheric monitoring, solar-energy forecasting, aviation weather assessment, and climate observation systems. However, reliable sky-image annotation is time-consuming, especially when cloud types are visually similar or mixed. We study the label efficiency of deep learning for ground-based cloud classification using the Ground-based Cloud Dataset (GCD). Rather than proposing a new architecture, we benchmark three practical strategies under limited annotation budgets: supervised transfer learning, uncertainty-based active learning, and high-confidence pseudo-labeling. An ImageNet-pretrained ResNet50 is used as a common frozen backbone, with experiments repeated over five random seeds for label budgets from $1\%$ to $100\%$ of the training labels. Supervised transfer learning is already highly label-efficient: test accuracy increases from $0.635 \pm 0.018$ with $1\%$ labels to $0.730 \pm 0.002$ with $40\%$ labels, approaching the full-label result of $0.735 \pm 0.003$. Active learning and pseudo-labeling are competitive with supervised sampling and provide small improvements for some metrics and budgets, but neither gives a large or consistent aggregate gain. Diagnostic analyses show that accepted pseudo-labels are reliable, with accuracy from $0.946$ to $0.977$, but biased toward easier high-confidence sky-type groups. In contrast, uncertainty sampling preferentially queries visually challenging groups, including Mixed and the confusable Stratocumulus and Cumulonimbus groups, but these targeted acquisitions yield only modest gains. Overall, transfer learning substantially reduces annotation requirements for GCD, while simple active and semi-supervised strategies provide limited additional benefit over a strong supervised baseline.
