Some Information About ENVI LiDAR's Tree Extraction Algorithm

While the exact algorithms used to extract features from LiDAR data using ENVI LiDAR are considered proprietary, our developers are often happy to help answer questions that come up about how those algorithms work. Here are answers to a couple of common questions about ENVI LiDAR's tree extraction algorithm.

Question: How does the tree extractionalgorithm work? 

Answer: It essentially takes points that are not part of a planar surface (terrain, buildings etc.) and that lie between the min and max tree height parameters and creates a DSM raster from them. From this DSM it looks for high points where the surface slopes downward in all directions. The high point is the tree crown and the extent of the downward sloping surface is the tree canopy.

Question: Why are some trees not extracted?

Answer: There are many possible reasons, depending on the data. It could be that the trees are densely packed together in a forest. The algorithm can perform better in urban areas where the trees are more separate.

Note: If you find that ENVI LiDAR does not extract all of the trees from your LiDAR data, and if you are willing to share your data with Exelis VIS, please contact Tech Support. Our developers are happy to explore new data to try to improve the performance of ENVI LiDAR's algorithms.

Review on 11/18/14 PS