DBCF: Classification and Understory Terrain Estimation of High-Precision ICESat-2 Photon Point Clouds

Abstract

In understory terrain studies, high-precision photon point cloud classification is essential for the comprehensive analysis of forest ecosystems. However, ICESat-2 Advanced Topographic Laser Altimeter System photon data contain substantial noise photons and inherently lack terrain-adaptive properties, leading to frequent misclassification of noise photons as ground signals and thus severely reducing terrain extraction accuracy. To overcome this challenge, we propose a density-based clustering and filtering (DBCF) framework that integrates density-stratified adaptive denoising with geometry-driven cloth-surface classification, enabling the robust classification of ICESat-2 photon point clouds into ground, canopy, and noise categories without using airborne terrain products as classification references. Across four contrasting forest regions—South Carolina, Maryland, Massachusetts, and Georgia—the proposed framework consistently outperformed NASA’s official algorithm and conventional DBSCAN. With airborne Goddard’s LiDAR, Hyperspectral, and Thermal Imager digital terrain model/digital surface model reference data, DBCF achieved an average recall of 0.66, a precision of 0.72, and an <inline-formula><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula>-score of 0.69; with manual reference, it achieved an average recall of 0.79, a precision of 0.78, and an <inline-formula><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula>-score of 0.78. Notably, the extracted ground photon data points achieved an average <inline-formula><tex-math notation="LaTeX">$R^{2}$</tex-math></inline-formula> of approximately 0.99 and an average root-mean-square error (RMSE) of 2.21 m, with RMSE values ranging from 1.92 to 2.40 m in understory terrain estimation, substantially outperforming the forest and buildings removed copernicus digital elevation model across all study areas. Furthermore, slope-stratified evaluation showed that the <inline-formula><tex-math notation="LaTeX">$F_{1}$</tex-math></inline-formula>-score decreased from 0.61 on flat terrain to 0.57 on moderate slopes and 0.47 on steep slopes, indicating that increasing terrain slope remains an important constraint on ground-photon classification. These results demonstrate the framework’s adaptability across diverse forest and terrain conditions, offering a reliable basis for high-precision forest parameter retrieval and ecological assessment.