Machine learning fusion of lidar and optical data for high-precision rice canopy height estimation: A multi-method comparison

Rice canopy height is critical for biomass estimation and crop monitoring. LiDAR provides accurate structure but limited temporal coverage,
while multispectral sensors offer frequent monitoring but lack direct height information.
Challenge: Can data fusion leverage complementary strengths of both sensors?

This study evaluates four fusion approaches combining LiDAR percentiles with optical vegetation indices to generate improved Canopy Height Models (CHM) for rice phenology monitoring.