interpolation help using LiDAR .LAS data

[JosephCarter]

I am working on comparing different methods of interpolation using LiDAR data. I have several .las files and am wondering if it is better to convert the data to a terrain dataset or point feature class before performing the interpolations, and what are the steps for converting the data? The methods of interpolation being performed are kriging, IDW, and TIN.

[DanPatterson_Retired]

There are a variety of tools already built-in, including   las-dataset-to-raster and this The whole las toolset and tools on the scripts site that might help

[DarrenWiens2]

I'm going around in circles trying to decide what you're trying to do. Are you trying to improve upon the existing methods of interpolating LAS point cloud points to DEM/DTM surfaces, or just end up with such a surface? The input for the interpolations you mention are points, so I would suggest converting your LAS points (or a subsample) to points, although doing so directly may be tricky using only ArcGIS, if it's possible at all. The beauty of the LAS dataset is that you don't have to deal with the entire point cloud, since it's been interpolated already (or at least it will be once you export it to a raster).

[NeilAyres]

Yes, dozens of questions. LAS, has it already been classified, ground, non-ground, etc etc. Have you looked at the LAS to see if there are some unwanted spikes. This happens quite a lot. Classifying LAS can be a bit of a black art.

If LAS to Raster, which also offers IDW, NN interp, and options on how to fill holes.

If you want to specifically test Kriging (why), IDW etc, then load the LAS into a point feature and go from there.

But also use the LAS dataset tools to do a bit of QC on the LAS first.