ABSTRACT
Deep learning will be used, taking advantage of CNNs and RNNs to improve the temporal prediction of crop seasons with high resolution satellite imagery as input. Comprehensive spatial-temporal land use and land cover maps and data obtained through geo-informatics with transfer learning-based integration of different remote sensing data (multispectral and radar) demonstrated significant skill in enhancing the understanding of global changes while limited by computational complexities, suggesting improvements in datasets quality kind be meritoriously considered to comprehensively support agroeconomic studies worldwide.
