Dynamic World Test Tiles

From the Dynamic World dataset, DOI: <strong>PENDING DYNAMICWORLD DOI</strong> <strong>Description:</strong> These data comprise the Dynamic World estimated probabilities, Dynamic World Top-1 label, and expert consensus set of GeoTIFF test tiles used for validating Dynamic World and other LULC maps. A metadata CSV is included to enable cross-walking between these tiles and the Sentinel-2 L2A used to annotate the tile. The label_* images contain bands 'lulc' and 'label' corresponding to the annotated, and Top-1 predicted label respectively. The 'lulc' band is 1s-indexed, with 0 corresponding to "no markup." The probs_* images contain the estimated probabilities. <strong>The Dynamic World abstract:</strong> <em>We developed a new automated approach for globally consistent, high resolution, near real-time (NRT) land use land cover (LULC) mapping leveraging deep learning on 10m Sentinel-2 imagery. When compared to other global LULC datasets, our data exceeded the next best global product agreement with an expert consensus test set by 7.5%. We utilize a highly scalable cloud based system for generating LULC maps and provide an open, continuous feed of LULC in parallel with Sentinel-2 acquisitions. This NRT product accommodates a variety of user needs ranging from extremely up-to-date LULC data to annual global maps. Furthermore, the continuous nature of the product's outputs enables refinement, extension, and even redefinition of the LULC classification. In combination, these unique attributes enable unprecedented flexibility for a diverse community of users across a variety of disciplines.</em>来自Dynamic World数据集，DOI: <strong>待提交Dynamic World数据集DOI</strong> <strong>数据集说明：</strong> 本数据集包含Dynamic World估算概率、Dynamic World Top-1预测标签，以及用于验证Dynamic World与其他土地利用与土地覆盖（Land Use Land Cover, LULC）地图的附带专家共识标注的GeoTIFF测试瓦片集合。配套提供元数据CSV文件，用于实现这些测试瓦片与用于标注该瓦片的Sentinel-2 L2A数据之间的交叉关联。label_*图像包含"lulc"与"label"两个波段，分别对应已标注的土地覆盖标签与Top-1预测标签。其中"lulc"波段以1为起始索引，数值0代表"无标注"。probs_*图像则包含各分类的估算概率值。 <strong>Dynamic World数据集摘要：</strong> <em>本研究提出一种全新的自动化方法，可基于10米分辨率的Sentinel-2影像，借助深度学习技术生成全球一致、高分辨率的近实时（Near Real-Time, NRT）土地利用与土地覆盖（Land Use Land Cover, LULC）地图。与其他全球LULC数据集相比，本数据集与专家共识测试集的一致性较次优全球产品提升7.5%。本研究采用高度可扩展的云计算系统生成LULC地图，并提供与Sentinel-2卫星过境同步的开放、持续更新的LULC数据馈送。该近实时产品可满足多样化用户需求，覆盖从极新LULC数据到年度全球地图等各类场景。此外，产品输出的持续性特性支持对LULC分类体系进行优化、扩展甚至重新定义。综上，这些独特属性为跨多学科的广泛用户群体提供了前所未有的灵活性。</em>