Deep learning for conifer/deciduous classification of airborne LiDAR 3D point clouds representing individual trees.

Hamid Hamraz,Nathan B. Jacobs,Marco A. Contreras, Chase H. Clark

ISPRS Journal of Photogrammetry and Remote Sensing（2019）

引用 81|浏览11

暂无评分

摘要

•Designed discrete representations for a tree crown point cloud to input to a CNN.•Compared with shallow learning, classification accuracies were improved up to 14%.•Leaf-off LiDAR data provided the most useful information for the classification.

查看译文

关键词

Remote sensing,Convolutional neural network,Representation engineering,Unbalanced training data,Mislabel correction

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要