A Large Crowdsourced Street View Dataset for Mapping Road Surface Types in Africa

This dataset includes Mapillary street view images and corresponding location data from Africa. The street view images are saved in JPG format, with each image assigned a unique ID number, totaling 200,000 images. The location data is represented as point vector data in Esri Shapefile format, where each point includes: a unique street view ID, longitude, latitude, and road surface type , with a total of 200,000 points. All data are projected using the World Geodetic System (WGS) 84 and the pseudo-Mercator coordinate system (EPSG: 3857) .

本数据集包含来自非洲的马普里（Mapillary）街景图像及其配套位置数据。该街景图像采用JPG格式存储，每张图像均分配有唯一标识符编号，总计200000张。位置数据采用Esri Shapefile格式的点矢量数据存储，每个点包含唯一街景ID、经度、纬度以及路面类型，总计200000个点。所有数据均基于世界大地坐标系（WGS 84）与伪墨卡托坐标系（EPSG: 3857）进行投影。