Voxel-Based Spatial Filtering Method for Canopy Height Retrieval from Airborne Single-Photon Lidar

摘要: Airborne single-photon lidar (SPL) is a new technology that holds considerable potential for forest structure and carbon monitoring at large spatial scales because it acquires 3D measurements of vegetation faster more efficiently than conventional instruments. However, SPL instruments use green wavelength (532 nm) lasers, which are sensitive to background solar noise, therefore point clouds require elaborate noise filtering other determine canopy heights, particularly in daytime acquisitions. Histogram-based aggregation commonly used approach removing from photon counting data, but reduces the resolution dataset. Here we present an alternate voxel-based method filters points while largely preserving integrity data. We develop test our algorithms on experimental dataset acquired over Garrett County Maryland, USA. then compare attributes retrieved using algorithm with those obtained histogram binning approach. Our results show heights derived have strong agreement field-measured (r2 = 0.69, bias 0.42 m, RMSE 4.85 m) discrete return 0.94, 1.07 2.42 m). Results consistently better height accuracies (field data: r2 0.59, 0.00 6.25 m; DRL: 0.78, −0.06 m 4.88 Furthermore, find spatial-filtering retains fine-scale detail has lower errors steep slopes. believe automated such as one presented here can support large-scale, mapping airborne