LandCoverNet: A global benchmark land cover classification training dataset

TL;DR

LandCoverNet is a comprehensive global dataset derived from Sentinel-2 satellite imagery, designed to improve land cover classification models essential for monitoring sustainable development goals.

Contribution

It introduces a large, diverse, and verified land cover training dataset based on multispectral satellite data for global land classification tasks.

Findings

Provides a high-quality, verified dataset for land cover classification

Enables development of more accurate global land monitoring models

Supports sustainable development goal monitoring efforts

Abstract

Regularly updated and accurate land cover maps are essential for monitoring 14 of the 17 Sustainable Development Goals. Multispectral satellite imagery provide high-quality and valuable information at global scale that can be used to develop land cover classification models. However, such a global application requires a geographically diverse training dataset. Here, we present LandCoverNet, a global training dataset for land cover classification based on Sentinel-2 observations at 10m spatial resolution. Land cover class labels are defined based on annual time-series of Sentinel-2, and verified by consensus among three human annotators.