COMPARISON OF CONVOLUTIONAL NEURAL NETWORKS FOR CLOUDY OPTICAL IMAGES RECONSTRUCTION FROM SINGLE OR MULTITEMPORAL JOINT SAR AND OPTICAL IMAGES

This paper presents nine standardized crop type reference datasets collected between 2013 and 2020 in seven tropical countries. It aims at participating in the difficult exercise of mapping agricultural land use through satellite image classification in those complex areas where few ground truth or census data are available. These quality-controlled datasets were collected in the framework of the international JECAM initiative and contain 27 074 polygons documented by detailed keywords.