基于卫星图像时间序列的牧场恢复监测：注意事项与机遇
Monitoring Pasture Restoration from Satellite Image Time Series: Caveats and Opportunities
浏览论文内容
中文总结 AI 辅助
本研究将牧场恢复设为二分类深度学习问题，评估两种SITS架构在1397个瑞典恢复牧场上的表现，最优模型准确率达0.88，同时指出可靠部署需注意时间平衡标签等关键因素。
中文摘要 AI 辅助
大规模监测自然恢复是一项重要但困难的生态学问题。用于分析卫星图像时间序列（SITS）的深度学习方法已被广泛应用于地表监测。本研究聚焦的生境类型为半天然草原，其恢复结果是逐步发展的，但卫星观测会受到天气、采集条件和处理伪影的影响，导致难以将真实的恢复信号与无关的时间变异区分开来。据我们所知，本研究首次将牧场恢复设定为二分类深度学习问题，探究是否能直接从卫星图像时间序列中检测恢复状态。我们在1397个瑞典恢复牧场样本上，针对不同的Sentinel-2图像组合评估了两种常见的SITS深度学习架构，发现明确建模年内变异以及每个牧场的归一化操作可提升可分性，最优模型的准确率达到0.88。我们进一步对结果进行调查并开展针对性偏差分析，发现可靠部署需要时间平衡的标签以及明确测试年份相关混淆的评估协议。因此，我们的贡献并非一个已解决的恢复监测系统，而是一个现实的案例研究，明确了哪些方法有效、哪些方法失效，以及未来研究应控制的因素。代码和模型可在该https网址获取。
英文摘要
Monitoring nature restoration at scale is an important but difficult ecological problem. Deep learning methods to analyze satellite image time series (SITS) have been widely used for land surface monitoring. In semi-natural grasslands - the habitat type in focus in this work - restoration outcomes develop gradually, yet satellite observations are influenced by weather, acquisition conditions, and processing artefacts, making it difficult to distinguish genuine restoration signals from unrelated temporal variation. In this work, we examine - to the best of our knowledge, for the first time - whether restoration status can be detected directly from satellite image time series by formulating pasture restoration as a binary deep learning classification problem. We evaluate two common SITS deep learning architectures on different Sentinel-2 image combinations, across 1,397 restored Swedish pastures and find that explicitly modeling intra-year variability and per-pasture normalization increases separability, reaching 0.88 accuracy for the best model. We further investigate our results and perform a targeted bias analysis finding that reliable deployment requires temporally balanced labels and evaluation protocols that explicitly test for year-related confounding. We therefore frame our contribution not as a solved restoration-monitoring system, but as a realistic case study of what works, what fails, and what future studies should control for. Code and models are available at https://github.com/aleksispi/ml-nature-resto.
发表机构
- Lund University（隆德大学）
- RISE Research Institutes of Sweden（瑞典RISE研究院）
- Climate AI Nordics（北欧气候人工智能中心）
- Swedish Centre for Impacts of Climate Extremes (CLIMES)（瑞典气候极端影响中心（CLIMES））
机构由 AI 辅助整理，请以论文原文为准。
