用卫星图像嵌入和图神经网络，提升非洲贫困地图的精度与覆盖范围。
Leveraging Compact Satellite Embeddings and Graph Neural Networks for Large-Scale Poverty Mapping
- 用低维卫星嵌入构建空间图，融合调查点与未调查点关系
- 在37个DHS数据集上，相比纯图像模型，预测准确率小幅提升
- 适合做大规模社会经济制图的研究者与政策制定者参考
全球南方地区精确、细粒度的贫困地图仍然稀缺。尽管人口与健康调查（DHS）提供高质量的社会经济数据，但其空间覆盖有限，且为保护隐私对坐标进行了随机偏移，进一步降低数据质量。本文提出一种基于图的方法，利用低维AlphaEarth卫星嵌入，在撒哈拉以南非洲地区预测聚类级别的财富指数。通过建模已调查点与未标注点之间的空间关系，并引入概率性“模糊标签”损失以应对坐标偏移问题，提升了财富预测的泛化能力。在37个DHS数据集（2017–2023）上的实验表明，引入图结构相较于仅使用图像的基线方法，准确率略有提升，验证了紧凑遥感嵌入在大规模社会经济制图中的潜力。
原文摘要 · Abstract (English)
Accurate, fine-grained poverty maps remain scarce across much of the Global South. While Demographic and Health Surveys (DHS) provide high-quality socioeconomic data, their spatial coverage is limited and reported coordinates are randomly displaced for privacy, further reducing their quality. We propose a graph-based approach leveraging low-dimensional AlphaEarth satellite embeddings to predict cluster-level wealth indices across Sub-Saharan Africa. By modeling spatial relations between surveyed and unlabeled locations, and by introducing a probabilistic "fuzzy label" loss to account for coordinate displacement, we improve the generalization of wealth predictions beyond existing surveys. Our experiments on 37 DHS datasets (2017-2023) show that incorporating graph structure slightly improves accuracy compared to "image-only" baselines, demonstrating the potential of compact EO embeddings for large-scale socioeconomic mapping.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。