Delineating Smallholder Maize Farms from Sentinel-1 Coupled with Sentinel-2 Data Using Machine Learning

Rural communities rely on smallholder maize farms for subsistence agriculture, the main driver of local economic activity and food security. However, their planted area estimates are unknown in most developing countries. This study explores use Sentinel-1 Sentinel-2 data to map farms. The random forest (RF), support vector (SVM) machine learning algorithms model stacking (ST) were applied. Results show that classification combined improved RF, SVM ST by 24.2%, 8.7%, 9.1%, respectively, compared individually. Similarities estimated areas (7001.35 ± 1.2 ha 7926.03 0.7 7099.59 0.8 ST) can estimate with high accuracies. concludes single-date insufficient sufficient mapping These results be used generation validation national crop statistics, thus contributing