2019/05/28 by Marc Rußwurm, Rußwurm, Marc, Pelletier, Charlotte +3 · 2 citations
Earth and Planetary Sciences · Environmental Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Isotope Analysis in Ecology #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Remote Sensing and Land Use #Remote Sensing in Agriculture
paper · pdf · doi:10.48550/arxiv.1905.11893
openalex publication_date 2019/05/28 · openalex created_date 2019/06/07 · openalex updated_date 2026/07/28
We present Breizhcrops, a novel benchmark dataset for the supervised\nclassification of field crops from satellite time series. We aggregated label\ndata and Sentinel-2 top-of-atmosphere as well as bottom-of-atmosphere time\nseries in the region of Brittany (Breizh in local language), north-east France.\nWe compare seven recently proposed deep neural networks along with a Random\nForest baseline. The dataset, model (re-)implementations and pre-trained model\nweights are available at the associated GitHub repository\n(https://github.com/dl4sits/BreizhCrops) that has been designed with\napplicability for practitioners in mind. We plan to maintain the repository\nwith additional data and welcome contributions of novel methods to build a\nstate-of-the-art benchmark on methods for crop type mapping.\n