Classification of lawn

[beyon]

Hi everyone

I'm new here and this is my first post.
I recently started to use CloudCompare, before I worked with ArcGis and QGis.

What I'm trying to do is classify lawn. The result would possibly be quite close to the usual classification "low vegetation".
As i thought about it, roughness could be a good feature to differentiate those areas. So I thought Canupo could be a good choice, since it works with spacial distribution of the data ( If i understood that right).

So I was trying to train some classifier, which turned out quite differently. The first one didn't work so well, but it could classify my data. The seond and third try isn't working, because it uses so much memory 30GB+. That's way to much for my computer. I tried to split up my cloud, which actually has only 1,5 million points into eight parts, but that didn't changed anything on memory usage.
I'm using an apple silicon 2.

So my question is, what could be the issue here? And could my approach be fruitfull?


In regards
Beyon

[daniel]

Yeah, Canupo should work great. What parameters are you using? (maybe you could share a screenshot). Have you used very/too large scales maybe?