This work uses the canopy height model (CHM) based workflow for individual tree crown delineation from LiDAR point cloud data in an urban environment and evaluates its accuracy by using very high-resolution PAN (spati...This work uses the canopy height model (CHM) based workflow for individual tree crown delineation from LiDAR point cloud data in an urban environment and evaluates its accuracy by using very high-resolution PAN (spatial) and 8-band WorldView-2 imagery. LiDAR point cloud data were used to detect tree features by classifying point elevation values. The workflow includes resampling of LiDAR point cloud to generate a raster surface or digital terrain model, generation of hill-shade image and intensity image, extraction of digital surface model, generation of bare earth digital elevation model and extraction of tree features. Scene dependent extraction criteria were employed to improve the tree feature extraction. LiDAR-based refining/filtering techniques used for bare earth layer extraction were crucial for improving the subsequent tree feature extraction. The PAN-sharpened WV-2 image (with 0.5 m spatial resolution) used to assess the accuracy of LiDAR-based tree features provided an accuracy of 98%. Based on these inferences, we conclude that the LiDAR-based tree feature extraction is a potential application which can be used for understanding vegetation characterization in urban setup.展开更多
文摘This work uses the canopy height model (CHM) based workflow for individual tree crown delineation from LiDAR point cloud data in an urban environment and evaluates its accuracy by using very high-resolution PAN (spatial) and 8-band WorldView-2 imagery. LiDAR point cloud data were used to detect tree features by classifying point elevation values. The workflow includes resampling of LiDAR point cloud to generate a raster surface or digital terrain model, generation of hill-shade image and intensity image, extraction of digital surface model, generation of bare earth digital elevation model and extraction of tree features. Scene dependent extraction criteria were employed to improve the tree feature extraction. LiDAR-based refining/filtering techniques used for bare earth layer extraction were crucial for improving the subsequent tree feature extraction. The PAN-sharpened WV-2 image (with 0.5 m spatial resolution) used to assess the accuracy of LiDAR-based tree features provided an accuracy of 98%. Based on these inferences, we conclude that the LiDAR-based tree feature extraction is a potential application which can be used for understanding vegetation characterization in urban setup.
