Global land use / land cover with Sentinel 2 and deep learning

2021/07/11 by Krishna Karra, Caitlin Kontgis, Zoe Statman-Weil +3 · 36 citations
Computer Science · Engineering · Environmental Science · #Geochemistry and Geologic Mapping #Remote Sensing in Agriculture #Remote-Sensing Image Classification

paper · doi:10.1109/igarss47720.2021.9553499

openalex publication_date 2021/07/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/31

Abstract

Land use/land cover (LULC) maps are foundational geospatial data products needed by analysts and decision makers across governments, civil society, industry, and finance to monitor global environmental change and measure risk to sustainable livelihoods and development. There is a strong need for high-level, automated geospatial analysis products that turn these pixels into actionable insights for non-geospatial experts. The Sentinel 2 satellites, first launched in mid-2015, are excellent candidates for LULC mapping due to their high spatial, spectral, and temporal resolution. Advances in deep learning and scalable cloud-based compute now provide the analysis capability required to unlock the value in global satellite imagery observations. Based on a novel, very large dataset of over 5 billion human-labeled Sentinel-2 pixels, we developed and deployed a deep learning segmentation model on Sentinel-2 data to create a global LULC map at 10m resolution that achieves state-of-the-art accuracy and enables automated LULC mapping from time series observations.

Citations

• An Ecoregion-Based Approach to Protecting Half the Terrestrial Realm

Cited by

• Automated identification of hedgerows and hedgerow gaps using deep learning
• The Global 2000-2020 Land Cover and Land Use Change Dataset Derived From the Landsat Archive: First Results
• Inequities in spatial access to childbirth care in the Grand Conakry conurbation, Guinea
• Testing Soil Moisture Performance Measures in the Conceptual‐Functional Equivalent to the WRF‐Hydro National Water Model
• Leveraging Open‐Source Geographic Databases to Enhance the Representation of Landscape Heterogeneity in Ecological Models
• Challenges in data-driven geospatial modeling for environmental research and practice
• Large-scale resource assessments for solar photovoltaics: A review of potential definitions, methodologies and future research needs
• FrogDeepSDM: Improving Frog Counting and Occurrence Prediction Using Multimodal Data and Pseudo-Absence Imputation
• Detection and Simulation of Urban Heat Islands Using a Fine-Tuned Geospatial Foundation Model for Microclimate Impact Prediction
• Large disagreements in estimates of urban land across scales and their implications
• Geospatial Machine Learning Libraries
• Geospatial data and workflows for environmental and sustainability compliance reporting: including the private sector
• Assessing the applicability of binary land-cover variables to species distribution models across multiple grains
• Movement patterns of three species of flying foxes across human-modified landscapes in the Philippines
• Local prediction of temperate forest structure in Eastern North America using LiDAR, radar, and optical data
• Advances on the Global Human Settlement Layer by joint assessment of Earth Observation and population survey data
• How accurate are existing land cover maps for agriculture in Sub-Saharan Africa?
• Detection and Simulation of Urban Heat Islands Using a Fine-Tuned Geospatial Foundation Model
• Geospatial Diffusion for Land Cover Imperviousness Change Forecasting
• AGI for the Earth, the path, possibilities and how to evaluate intelligence of models that work with Earth Observation Data?
• Evaluating Global and National Datasets in an Ensemble Approach to Estimating Carbon Emissions as Part of SERVIR’s CArbon Pilot
• Can tree cover extent be estimated directly from ICESat-2 spaceborne lidar data?
• LIGHTHOUSE: Fast and precise distance to shoreline calculations from anywhere on earth
• AgriPotential: A Novel Multi-Spectral and Multi-Temporal Remote Sensing Dataset for Agricultural Potentials
• Forecasting Residential Heating and Electricity Demand with Scalable, High-Resolution, Open-Source Models
• Impacts of climate change on the distribution pattern of Himalayan Rhubarb (Rheum australe D. Don) in Nepal Himalaya
• Holy yet Hindered: Habitat use patterns of a large herbivore reveal challenges to Farmers in India
• Análisis exploratorio de la amplitud ambiental y requerimientos ecológicos para identificar áreas potenciales de muestreo y cultivo de Elionurus muticus (Poaceae) en América del Sur
• A Contrastive Learning Foundation Model Based on Perfectly Aligned Sample Pairs for Remote Sensing Images
• Andean bears (Tremarctos ornatus) display selective behaviors while foraging bromeliads (Puya spp.) in high elevation puna grasslands
• Leopard occupancy correlates with tiger and prey occurrences in the Parsa‐Koshi Complex, Nepal
• Joint Super-Resolution and Segmentation for 1-m Impervious Surface Area Mapping in China's Yangtze River Economic Belt
• Land Use and Land Cover Mapping in the Era of Big Data
• Urban Flood Observations: A hand-labeled training and validation dataset of post-flood inundation
• When Proxy Prediction Becomes Equation Reconstruction: Diagnostics and Residual Learning for Factor-Derived Proxy Supervision
• VoxCity: A Seamless Framework for Open Geospatial Data Integration, Grid-Based Semantic 3D City Model Generation, and Urban Environment Simulation

Related

• Mapping Land Naturalness from Sentinel-2 using Deep Contextual and Geographical Priors
• EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification
• LandCoverNet: A global benchmark land cover classification training dataset
• Land Use and Land Cover Mapping in the Era of Big Data
• SEN12MS -- A Curated Dataset of Georeferenced Multi-Spectral Sentinel-1/2 Imagery for Deep Learning and Data Fusion