FLAIR-one

FLAIR-one数据集是由法国国家地理和森林信息研究所创建，用于支持法国国家土地覆盖图的制作。该数据集包含77,412个512x512像素的图像块，总计覆盖约810平方公里，涉及50个时空域和13个语义类别。数据集的创建过程涉及使用高分辨率航空图像和数字表面模型等数据源，通过深度学习方法进行语义分割。该数据集主要应用于土地管理和规划，旨在通过人工智能技术提高土地覆盖信息的准确性和可靠性。

The FLAIR-one dataset was developed by the French National Institute of Geographic and Forestry Information to support the production of France's national land cover maps. This dataset contains 77,412 image patches with a resolution of 512×512 pixels, covering a total area of approximately 810 square kilometers, and involves 50 spatiotemporal domains and 13 semantic categories. The creation of the dataset utilized multiple data sources including high-resolution aerial imagery and digital surface models, and adopted deep learning-based approaches for semantic segmentation. This dataset is primarily applied in land management and planning, with the goal of enhancing the accuracy and reliability of land cover information through artificial intelligence technologies.