Global-Scale
收藏资源简介:
Global-Scale数据集是由西安交通大学、中国科学院和西安电子科技大学联合创建的一个全球规模的卫星图像道路图提取数据集。该数据集覆盖了除南极洲以外的所有大陆,包含3468张2048×2048像素的卫星图像,涵盖了城市、乡村、山区等多种复杂环境。数据集的创建过程包括从Google Earth和OpenStreetMap收集数据,并通过人工筛选确保标注的完整性。Global-Scale数据集旨在解决现有道路图提取数据集在数据量和多样性上的不足,为自动驾驶和导航系统提供更全面的数据支持。
The Global-Scale dataset is a global-scale satellite image road extraction dataset jointly created by Xi'an Jiaotong University, Chinese Academy of Sciences, and Xidian University. It covers all continents except Antarctica, and contains 3468 satellite images with a resolution of 2048×2048 pixels, covering various complex environments such as urban, rural, and mountainous areas. The dataset was developed by collecting data from Google Earth and OpenStreetMap, followed by manual screening to ensure the completeness of annotations. The Global-Scale dataset aims to address the shortcomings of existing road extraction datasets in terms of data volume and diversity, providing more comprehensive data support for autonomous driving and navigation systems.




