Brazil - Crop Dataset

LEM Benchmark Database for Tropical Agricultural Remote Sensing Applications

The LEM Benchmark Database is a publicly available dataset designed for tropical agricultural remote sensing applications in Brazil. This project focuses on creating reference data for crop mapping in the western Bahia region, one of Brazil’s most important agricultural frontiers.

The dataset supports research on land use and land cover classification, crop type mapping, and agricultural monitoring using multi-sensor satellite data (optical and SAR).

  • Sanches et al. (2018, 2020) — LEM benchmark database publications
  • Prudente et al. (2022) — Multisensor approach to LULC mapping in the Brazilian Amazon (ISPRS Journal)
  • Santos et al. (2025) — LULC Products for Agricultural Mapping Applications in Brazil (Remote Sensing)

References