Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time

Build Status Build status codecov


Pure Julia reader of lasindex .lax files. Translates the .lax file to a quadtree from the RegionTrees package and provides a bounding box intersect method to retrieve a vector of ranges.

julia> using LASindex
julia> using FileIO

julia> laxfile = joinpath("sample", "Palm Beach Post Hurricane.lax");
julia> qt = load(laxfile)
INFO: Processed Palm Beach Post Hurricane.lax with 1924631 points.
Cell: RegionTrees.HyperRectangle{2,Float64}([955000.0, 885000.0], [32000.0, 32000.0])

julia> using StaticArrays
julia> bbox = SVector(955000.0, 985002.0, 885000.0, 988800.0);
julia> r = LASindex.intersect(qt, bbox)
1-element Array{UnitRange{Integer},1}:

The resulting ranges can be used in combination with LasIO or LazIO to stream a small subset of a (larger than memory) dataset.

using LasIO
header, points = load("Palm Beach Post Hurricane.las", mmap=true)

intersected_points = points[LASindex.intersect(qt, bbox)]

Note that all points inside the bounding box are given, but not all points given are inside the bounding box. In other words, because of how lasindex groups ranges together, some ranges will include points outside the bounding box.


LAX files are quadtree indexes used by the LASTools suite [1] if present. You can generate them with lasindex -i *.laz[2]. There's a good introduction to lasindex here [3].


I'd rather use Cxx to call the lasindex c++ code directly, but at the moment the Cxx package is harder to install on Windows than compiling the lastools shared library itself.

An example how to interface with lastools directly is provided in example/cxx.jl