This dissertation examines the methods and findings behind the most extensive dataset used as reference data for Machine Learning (ML)-based Land Cover Classification (LCC) with remote sensing imagery.The discovery of administrative data as training data for ML models is responsible for a significant leap in research, especially in the field of large-scale LCC from optical satellite imagery.EuroCrops has previously been presented as a solution, and in this work, it will be put into context, and its key findings and larger impact analysed.
«
This dissertation examines the methods and findings behind the most extensive dataset used as reference data for Machine Learning (ML)-based Land Cover Classification (LCC) with remote sensing imagery.The discovery of administrative data as training data for ML models is responsible for a significant leap in research, especially in the field of large-scale LCC from optical satellite imagery.EuroCrops has previously been presented as a solution, and in this work, it will be put into context, and...
»