DeepGlobe Land Cover 2018 Dataset
收藏资源简介:
The authors introduce DeepGlobe Land Cover Classification Dataset 2018 Challenge, a comprehensive collection of high-resolution sub-meter satellite imagery focusing on rural areas. This dataset includes 1,146 satellite images of size 2,448x2,448 pixels, split into train, valid, and test sets. The images encompass RGB data with a pixel resolution of 50 cm, sourced from the DigitalGlobe Vivid+ dataset. Pixel-wise segmentation masks were created by professional annotators to represent different land cover classes. The total area size of the dataset is equivalent to 1716.9 km²
作者们推出了2018年DeepGlobe土地覆盖分类数据集挑战赛,该数据集是一套全面的高分辨率亚米级卫星影像集合,专注于农村地区。该数据集包含1,146幅尺寸为2,448x2,448像素的卫星图像,分为训练集、验证集和测试集。图像涵盖了50厘米像素分辨率的RGB数据,数据来源于DigitalGlobe Vivid+数据集。像素级分割掩码由专业标注员创建,以表示不同的土地覆盖类别。该数据集的总面积相当于1716.9平方公里。




