Remote sensing with AI maps small coffee plantations with high precision

Researchers have developed an unprecedented method for mapping coffee plantations via remote sensing with unmatched sensitivity and specificity. The technique achieved over 95% accuracy by combining time series of images from the Harmonized Landsat Sentinel-2 (HLS) program with artificial intelligence algorithms such as Random Forest and XGBoost. In addition to identifying coffee plantation areas, the study managed to distinguish four phenological stages of the crop—planting, production, pruning, and renewal—with accuracy between 77% and 95%, even in highly fragmented areas dominated by small properties. The technique is scalable and can be applied in any coffee-growing region. This paves the way for public policies, access to rural credit, and climate adaptation practices in producing regions. “The major challenge for remote sensing is mapping these highly productive regions with greater detail and accuracy, yet they have a small to medium-scale productive profile. Large-scale ...