Is sub-metre resolution necessary for cocoa mapping? A landscape-stratified evaluation of very high resolution imagery, decametric Earth Observation inputs, and operational products in Cote d'Ivoire

📄 arXiv: 2607.08945v1 📥 PDF

作者: Kasimir Orlowski, Filip Sabo, Michele Meroni, Astrid Verhegghen, Mariana Belgiu, Felix Rembold

分类: cs.CV

发布日期: 2026-07-09


💡 一句话要点

通过高分辨率影像提升可可种植区映射精度

🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)

关键词: 可可映射 高分辨率影像 地球观测 基础模型嵌入 农业监测 环境保护 森林砍伐监测

📋 核心要点

  1. 现有的中等分辨率地球观测影像在异质性小农景观中限制了可可的检测精度。
  2. 论文提出使用超高分辨率影像和基础模型嵌入来提升可可映射的准确性。
  3. 实验结果显示,超高分辨率模型在复杂景观中表现最佳,F1得分超过0.90,显著优于其他方法。

📝 摘要(中文)

准确的可可映射对于监测森林砍伐、供应链透明度和监管应用愈发重要。传统中等分辨率的地球观测影像在异质性小农景观中可能限制可可的检测。本文评估了在象牙海岸不同景观条件下,映射性能的变化,探讨了超高分辨率影像是否提供显著优势,以及基础模型嵌入是否改善了十米级可可映射。研究表明,超高分辨率模型在所有景观层次中均表现优异,F1得分达到0.92,而十米级输入中,TESSERA表现最佳,F1得分为0.86。

🔬 方法详解

问题定义:本文旨在解决在异质性小农景观中可可映射精度不足的问题。现有的中等分辨率影像在复杂地形和树木覆盖密度变化大的情况下,难以准确识别可可种植区。

核心思路:研究通过引入超高分辨率影像(0.5米)和基础模型嵌入,来提升可可映射的准确性,尤其是在复杂的景观条件下。

技术框架:整体方法包括数据采集、模型训练和性能评估三个主要阶段。数据采集使用Pleiades VHR影像和Sentinel-2年复合影像,模型训练则结合了基础模型嵌入。性能评估通过在不同景观条件下的准确性评估进行。

关键创新:最重要的创新在于使用超高分辨率影像与基础模型嵌入的结合,显著提高了在复杂景观中的映射精度。这与传统的中等分辨率方法形成鲜明对比。

关键设计:在模型训练中,采用了特定的损失函数以优化F1得分,并对网络结构进行了调整,以适应不同的输入数据类型和特征提取需求。

🖼️ 关键图片

img_0
img_1
img_2

📊 实验亮点

实验结果显示,超高分辨率模型的F1得分达到0.92,显著高于十米级输入的最佳表现(F1 = 0.86)。在复杂景观条件下,超高分辨率影像的优势更加明显,尤其是在树木覆盖密度变化大的情况下。

🎯 应用场景

该研究的潜在应用领域包括农业监测、环境保护和供应链管理。通过提高可可种植区的映射精度,可以为政策制定者和农民提供更准确的信息,从而促进可持续农业发展和森林保护。未来,该方法还可扩展至其他农作物的监测与管理。

📄 摘要(原文)

Accurate cocoa mapping is increasingly important for deforestation monitoring, supply-chain transparency, and regulatory applications. Spatial aggregation in conventional medium-resolution Earth observation (EO) imagery may limit cocoa detection in heterogeneous smallholder landscapes. In Cote d'Ivoire, we therefore evaluated how mapping performance varies across landscape conditions, whether very high resolution (VHR) imagery provides a meaningful advantage, and whether foundation-model embeddings improve decametric cocoa mapping. We developed models using 0.5 m Pleiades VHR imagery, a 10 m Sentinel-2 annual composite, and embeddings from TESSERA and AlphaEarth Foundations (AEF), and additionally assessed four publicly available cocoa mapping products. Performance was evaluated through a landscape-stratified accuracy assessment using 2,821 independently interpreted reference points distributed across gradients of tree cover density and landscape fragmentation. The VHR model achieved the highest performance (F1 = 0.92) and maintained F1-scores above 0.90 across all strata. Among the decametric inputs, TESSERA performed best (F1 = 0.86), followed by AEF (F1 = 0.82) and Sentinel-2 (F1 = 0.76). Of the existing cocoa products, the Kalischek product performed best (F1 = 0.83), comparable to the internally trained AEF model. Performance differences between VHR and decametric approaches increased with fragmentation and under low and high tree cover density conditions. Targeted VHR acquisition may therefore be particularly beneficial in complex cocoa landscapes, while foundation-model embeddings offer a scalable alternative for large-area mapping.