Remove close points OR specify minimum triangle size for surface?

[autoMick]

Hi, I have some survey data where the surveyor has gone nuts and recorded a lot more points than is necessary or useful. The points are often too close together along the tracks, and there are overlaps between different survey methods (i.e. the colours below). The goal is to create single TIN surface from the data that isn't full of 1m wide slithers. A nice way to deal with this would be to just set a minimum triangle length. Any ideas?

Thanks

Mick

I've attached the file which is PENZD and the coordinates are in MGA/20-56 (if you're using a recent version of Civil3d).

Thanks

Mick

[samir_rezk]

Hi @autoMick 

I am not sure if you have multiple surfaces overlapping or a single surface?! Use the Paste Surface command to generate a single surface if you have more than one surface objects and they are overlapping! 
To decrease the number of point a single surface, you may use the Simplify Surface command! Based on criteria you specify in the wizard it will decrease the number of overall points in the surface.

Hope this helps,

[autoMick]

Thanks Samir, the idea is to have a single surface. I have been using simplify surface as it's really the only option that I can see. I was wondering if there were other strategies?

Cheers

Mick

[Pointdump · Accepted solution]

G'Day Mick,
Thanks for posting your data. I was having some fun with your CSV this morning. While I doubt it's any better than Simplify Surface in C3D, in CloudCompare you can use Subsample to chop down your point file. I used a spacing of 25 Meters(I think the spacing is in Meters) which left 1/10 the points. Attached.

Dave

[autoMick]

Hi Dave, nice one... I'll definitely look into that. You seem to know all the cool tools.

Yes the coordinate system is based on metres.

Regards

Mick

[rl_jackson]

@autoMick

With hydro-graphic work, I've always found it best to treat it similar to a point cloud, just drop a bunch of data out. Hydro is very similar to Lidar, just the opposite direction. Simplify might help but the density is really going to be a issue just as Drone stuff is. Think of it this way, I probably only need every 5th or 10th point. Considering the motion in the ocean, its always going to be changing, one tide cycle to another.

[autoMick]

Thanks Rick,

Yes I agree, thinning the 'cloud' is entirely acceptable. It's a very interesting topic when considering survey methodologies and what analyses are appropriate, and what is not.

Regards

Mick

[autoMick]

Dave @Pointdump thanks for putting me on to cloudcompare - I've had a chance to have a good look at it properly now and it does what I was hoping might be achievable in Civil3d very quickly.

Thanks everyone else for contributing.

Regards

Mick