These are chat archives for PDAL/PDAL

24th
May 2017
Andrew Bell
@abellgithub
May 24 2017 12:09
What is the LAS version of the file?
Point record type?
Andrew Bell
@abellgithub
May 24 2017 12:19
It looks to me like your source file has strange classifications. PDAL simply reports the value in the classification field. It is 8 bits wide, but only the first five bits are actual classifications. The high order three bits contain other information about the kind of classification. Your data shows that these bits are set in many cases. PDAL doesn't attempt to interpret the classification values. lasinfo does.
This assumes a point record type of 0-5.
chambbj
@chambbj
May 24 2017 14:16
@ulimaps I know this is about a month old at this point - I'm just getting caught up on the discussion. The paper you reference is precisely what we've implemented in filters.pmf and pdal ground, so probably no need to put together the pipeline you are working through, unless I'm missing something. That being said, the step you describe of first finding minimum points per cell is quite common. We could expose this as its own filter, if enough people found it useful. It begs the question of "How are you going to use it in your pipeline?" For example, do you really want to remove all but the minimum (or central, or whatever) points per cell? Or do you want to label them in some way, such that this information can be used downstream in the pipeline without being a destructive operation.