qCANUPO (classifier files, etc.)

[Dimitri]

Hi,

indeed Canupo only uses geometric informations at various scales (our team is working on a more generic version using any kind of attribute, including intensity, but it's not ready yet). I've already used Canupo to roughly classify rock outcrop separating gravel beds from loess material (I've a background in geology), but if you're looking at classifying layers which are relatively fine with little difference in roughness, there's one limitation with Canupo right now that is it computes neighborhood within spheres....given that you're interested in layers, it would be probably much more efficient to use a neighborhood search that would be perpendicular to the layering so that you don't polute the classifiers with geometric features computing for neighboring layers (i.e. a kind of ellipsoidal neighborhood search or even a cylinder). Not necessarily too difficult to implement, but as always, it means time to implement and test that we don't have right now.

But it's good to know that Canupo is useful for you to remove vegetation from outcrops.

Dimitri Lague

[bluehorizon]

Daniel and Dimitri,

I've been able to improve my classifier so that now I over 90% of the ground points in the toe part (as mentioned in previous comment) get classified into bare earth. I should thank you guys again for creating something like this and making it opensource.

The next step for me is to fill the holes left by tress and buildings in my model after classification. Could you please point me towards any opensource software/plugin for that? I'll really appreciate an earliest response.

Regards,

Umair.

Hi Umair,

How were you able to improve your classifier so that more of the steep land gets correctly classified as land and not vegetation? I am in a similar situation where chunks of steep cliffs are being classified as vegetation. I've been trying various combinations of scales as well as different classifiers (cliffs vs. vegetation, cliffs vs. flat land, etc.), and nothing seems to be doing the trick.

Thank you!

[Lookia]

Hi Sir,

Thank you for your work and all the information. I am trying to test a classifier on my cloud but can't find any. The webpage you listed (http://www.geosciences.univ-rennes1.fr/ ... rticle1284) couldn't be found. Just wondering if there is any available classifier to test now?

[daniel]

Hum, it seems that Dimitri has updated his page and these files are no longer accessible...

I've linked to this post the files I saved on my computer (here the "Otira" example, for clouds with meter units, as well as the 'tidal vegetation' classifier).

Otherwise, with Canupo it's generally pretty easy to train your own classifier.

[Dimitri]

Ah yes...my dept has updated its webpages recently, and the Canupo page got lost (arrghhh). I'll need to fix that at some point. Thanks Daniel for putting the classifier files back online.

[Lookia]

Hum, it seems that Dimitri has updated his page and these files are no longer accessible...

I've linked to this post the files I saved on my computer (here the "Otira" example, for clouds with meter units, as well as the 'tidal vegetation' classifier).

Otherwise, with Canupo it's generally pretty easy to train your own classifier.

Thank you Daniel for the classifiers! The situation was that whenever I used my own classifiers, the software got no responding and dead... And I applied the classifier you provided, it would run. Then that should be the problem of classifiers. Do you have any suggestions on how to improve this issue? Thanks

[daniel]

Maybe it's a matter of scales? If the scales are too big (and there are a lot of them) then the computation can take a really long time.

Make sure the scales match the units of the cloud also.

[swiss_knight]

Hi,
just a small question, I'm working on Linux (Ubuntu 16.04), I have build CC from sources but I can't find any 'canpuo' tag within cmake-gui.
I thought there would be some shared lib with the canupo repository itself, which I have cloned as well locally.

So...How to compile CC with qCanupo?

Thanks

[daniel]

Canupo sources are not public yet. Therefore it is only available in the Windows version for now.

[bluehorizon]

For those who have used the Canupo command line suite- have you ever encountered classification results that are missing huge stripes of data? (see photo below)

When I used the plug-in version in CloudCompare, I never had results like this. I've tried creating and using several variations of the .msc file to see what effect is has but with no success. I'm scratching my head here- any ideas?

Thank you!