Applying a U-Net Convolutional Neural Network for Mapping Banana Crops in the Atlantic Forest Region of Brazil Using CBERS-4A High Spatial Resolution Imagery
Abstract. Mapping banana crops in heterogeneous tropical landscapes remains challenging due to spectral similarity with surrounding vegetation, fragmented smallholder systems, and complex land-use mosaics. This study applies a deep learning approach, using a U-Net model, on high spatial resolution CBERS-4A imagery to m...