The purpose of the Li tree filter is to segment individual trees from an input PointView. In the output PointView points that are deemed to be part of a tree are labeled with a ClusterID. Tree IDs start at 1, with non-tree points given a ClusterID of 0.


The filter differs only slightly from the paper in the addition of a few conditions on size of tree, minimum height above ground for tree seeding, and flexible radius for non-tree seed insertion.


In earlier PDAL releases (up to v2.2.0), ClusterID was stored in the TreeID Dimemsion.

Default Embedded Stage

This stage is enabled by default


The Li tree algorithm expects to visit points in descending order of HeightAboveGround, which is also used in determining the minimum tree height to consider. As such, the following pipeline precomputes HeightAboveGround using filters.hag_delaunay and subsequently sorts the PointView using this dimension.




Minimum number of points in a tree cluster. [Default: 10]


Minimum height above ground to start a tree cluster. [Default: 3.0]


The seed point for the non-tree cluster is the farthest point in a 2D Euclidean sense from the seed point for the current tree. [Default: 100.0]


An expression that limits points passed to a filter. Points that don’t pass the expression skip the stage but are available to subsequent stages in a pipeline. [Default: no filtering]


A strategy for merging points skipped by a ‘where’ option when running in standard mode. If true, the skipped points are added to the first point view returned by the skipped filter. If false, skipped points are placed in their own point view. If auto, skipped points are merged into the returned point view provided that only one point view is returned and it has the same point count as it did when the filter was run. [Default: auto]