CropNet
收藏资源简介:
CropNet数据集是由特拉华大学和路易斯安那大学拉斐特分校联合创建的,是首个针对美国大陆县级气候变化感知作物产量预测的公开大型多模态数据集。该数据集包含三种模态数据:Sentinel-2影像、WRF-HRRR计算数据集和USDA作物数据集,覆盖2291个美国县,时间跨度为2017至2022年。数据集通过整合卫星图像、气象参数和县级作物产量信息,旨在帮助研究人员开发能够考虑短期生长季节天气变化和长期气候变化影响的作物产量预测模型。此外,CropNet数据集还提供了灵活的API,便于研究人员根据特定时间和区域需求下载数据,并构建精确的作物产量预测深度学习模型。
The CropNet dataset, jointly developed by the University of Delaware and the University of Louisiana at Lafayette, is the first publicly available large-scale multimodal dataset for county-level climate-aware crop yield prediction across the contiguous United States. This dataset includes three modalities of data: Sentinel-2 imagery, WRF-HRRR computational dataset, and USDA crop dataset, covering 2291 U.S. counties and spanning the period from 2017 to 2022. By integrating satellite imagery, meteorological parameters and county-level crop yield data, this dataset aims to help researchers develop crop yield prediction models that take into account both short-term growing-season weather fluctuations and the impacts of long-term climate change. Furthermore, the CropNet dataset offers a flexible API, allowing researchers to download data tailored to specific temporal and spatial needs and construct accurate deep learning models for crop yield prediction.




