Tree Species Classification Using Airborne LiDAR Data Based on Individual Tree Segmentation and Shape Fitting

Individual tree species classification is of strategic importance for forest monitoring, analysis, and management, which are critical sustainable forestry development. In this regard, the paper proposes a method based on profile segmented individual laser scanning points to identify species. The proposed methodology mainly takes advantage three-dimensional geometric features crown captured by point cloud Firstly, Digital Terrain Model (DTM) Surface (DSM) used Crown Height (CHM) generation. Then, local maximum algorithms improved rotating profile-based delineations segment trees from CHM data. next step, parallel-line shape fitting fit shape. particular, three basic shapes, namely, triangle, rectangle, arc shapes different If belongs same or combination, parameter used, such as ratio width height apex angle range triangles. was tested two real datasets were acquired sites located at Tiger Leopard National Park in Northeast China. experimental results indicate that average accuracy 90.9% optimal reached 95.9%, meets requirements rapid surveying.