Removing spikes or dips

[Ron L]

Is there a lastool that can remove erroneous spikes or dips in data?
I'm getting several spikes that are hundreds to thousands of feet
above the surrounding features and would love to remove them, but not
exactly sure how.

Thanks,

Ronnie

[Martin Isenburg]

hi.

does the LIDAR data have ground points already classsified? then you
can do this with

lasheight -i lidar.las -o clean_lidar.las -drop_below -10 -drop_above
1000

otherwise you need to identify ground points first.

lasground -i lidar.las -o lidar_with_ground_points.laz -city
lasheight -i lidar_with_ground_points.laz -o clean_lidar.las -
drop_below -10 -drop_above 1000

this could probably be done more efficiently with a dedicated tool but
maybe this works for you already?

martin

[Veldman, Edgar]

Hi Martin,

Replying to an old post here.
Spikes and dips (birds, multi path, etc.) appear in most lidar point clouds.

These points need to be removed from the point cloud prior to ground filtering, and can't always be removed by drop_below or drop_above functions, because these functions use absolute values and most projects cover an area with quite a large variation in height. Although some of these points can be isolated by classification by low intensity, it's not ideal because when removing those low intensity noise points there is a chance I am also removing/classifying valid low intensity returns.

It would be nice if there was functionality in one of the lastools (or a new tool) which can remove/classify these spikes and dips, eg. by comparing each point with other points in a certain radius around the point to determine if the point is significantly lower or higher than the other points, or if a group of points is lower than it's neighbours, or if the elevation of one point or a group of points significantly contributes to a larger standard deviation of the z for a specific area or search radius? Just some thoughts on how to achieve this.

What's your thought on this? The next blue moon is years away ...

Edgar

--
You can download LAStools at
http://www.cs.unc.edu/~isenburg/lastools/
You can visit the LAStools group at
http://groups.google.com/group/lastools/
You can be social with LAStools at
http://www.facebook.com/LAStools
http://www.twitter.com/LAStools

[Ty Kennedy-Bowdoin]

Hi all,

Martin's suggestion of using lasheight above ground is definitely the first line of attack, however if you are a like me, you may have some cloudy areas that are too large to accurately classify ground. In these situations, I use lasheight above a reference dataset. I use lasmerge to merge my lidar point cloud with some SRTM data (in las format and classified with a unique class id) and then I use lasheight to measure points that are too high or two low from the SRTM class which gives me a rough idea where the real points should be.

I can also see the use of a future tool that might look at points that are isolated and far from others to remove them before ground classification, but this works pretty well in the meantime.

Cheers,

Ty

--
Download LAStools at
http://lastools.org/
Visit the LAStools group at
http://groups.google.com/group/lastools/
Be social with LAStools at
http://www.facebook.com/LAStools
http://www.twitter.com/LAStools

--
Ty Kennedy-Bowdoin
Carnegie Institution for Science
260 Panama St.
Stanford, CA 94305
USA
+1 831 227-3885
Skype: Ty.bowdoin