MLS树物种分类数据集
收藏资源简介:
该数据集由芬兰地理空间研究所创建,包含1915个高密度移动激光扫描(MLS)树冠点云样本,涵盖7个树种,平均点密度达5000点/平方米。数据通过FARO Orbis激光扫描仪采集,经SLAM算法配准和手动标注验证,包含几何特征与多光谱强度信息。数据集旨在推动传感器无关的深度学习算法开发,解决复杂森林环境中树种自动分类问题,为生态监测和林业管理提供高精度数据支持。
This dataset was developed by the Finnish Geospatial Research Institute. It contains 1915 high-density mobile laser scanning (MLS) canopy point cloud samples covering 7 tree species, with an average point density of 5000 points per square meter. The data was collected using a FARO Orbis laser scanner, registered via SLAM algorithms and validated through manual annotation, and includes both geometric features and multispectral intensity information. This dataset is intended to advance the development of sensor-agnostic deep learning algorithms, resolve the challenge of automatic tree species classification in complex forest environments, and provide high-precision data support for ecological monitoring and forestry management.




