How to delete distant points from the scan's origin?

[Darius_bd]

.
Hi! I'm a newbie to CC but I can see that is very powerful. Regrettably I get lost in its powerfulness and I'm pretty sure there's an easy way to do what I'm asking but I have no clue how, also my googling skills are not the best....

Anyho, I have a project with, let's say, 20 structured scans and I want to be able to delete all points farther than 10m from the scan origin. Is it possible? The idea is to delete the points that I consider are not too reliable for registering against.

PS: Is there the opposite as well? Delete all points closer than?

Thanks!

[daniel]

I get lost in its powerfulness

Ah ah :D

You can compute distances between one cloud and another (Tools > Distances > Cloud to Cloud Distances - http://www.cloudcompare.org/doc/wiki/in ... d_Distance). Just select your central scan, and another one. Then call this tool. Set the central scan as 'reference' and the other as 'compared'. Compute the distances by pressing the red 'Compute' button (by the way you can set a maximum distance of 10.1m if you want to get a faster computation). Once done you can close the tool.

The select the compared cloud (the one with distances now). And call the 'Edit > Scalar fields > Filter by value'. Set the range of points you want to keep (e.g. [0 - 10]) and click on 'Split'. This will actually split the cloud in two (the points below 10m and the others).

Do that again for all the remaining clouds ;)

[Darius_bd]

.
Thanks for the quick reply Daniel!

But what I want is actually done on a single cloud. Supposedly on structured scans, the center of the scan is the origin of the laser. So I want to remove all points which are farther than this origin point. In essence, it would create a sphere out of the cloud.

Actually... maybe that's the way? Create a sphere of r=x at 0,0,0 (origin of laser?) and remove points outside that sphere?

[daniel]

Ah in this case yes!

Or you create a fake sensor ('Edit > Sensors > TLS/GBL > Create') in the right position (and you don't care about the other parameters). Then use the 'Edit > Sensors > Compute ranges' method. It should run much faster.