An Effective Leaf Area Index Estimation Method For Wheat From Uav-Based Point Cloud Data

Currently, Unmanned Aerial Vehicle (UAV)-based remote sensing is a flexible and reliable approach to gather data for agricultural crop intra-field monitoring. This study proposes real-time and low-cost approaches for crop leaf area index (LAI) estimation using UAV-based 3D point cloud data at field-scale. Crop LAI is an indicator of crop growth variation within crop fields which is one of the most essential crop parameters in crop growth models to predict other crop parameters including chlorophyll, biomass and final yield. After converting a circle with a radius of 2 meters 3D point cloud data to spherical projection, the sampling 3D point cloud data will be converted to a hemispherical photograph. The crop canopy LAI is then calculated from this hemispherical photograph using the gap fraction method. From the experiments over a winter wheat field, the estimated LAI from the UAV-based 3D point cloud data is highly correlated with the LAI estimated from an in-situ fisheye camera, the R-2 are 0.8995 and 0.8658 for 4 rings and 5 rings view angles calculation, respectively.