Bandwagon's Drones Thread

  • Guest, it's time once again for the massively important and exciting FoH Asshat Tournament!



    Go here and give us your nominations!
    Who's been the biggest Asshat in the last year? Give us your worst ones!

Vepil

Gamja
<Bronze Donator>
5,841
24,502
Basically you want to use the field calculator to only keep for example "Elevation <= 900" if you know that is about what your ground is and it will filter out all the higher. Take that output to a new layer and use the Interpolation plugin on that output layer and create your new TIN.

The engineers probably want the triangles that make up the DTM that is usually what is called a TIN as such:

rrr_img_136470.gif



AutoCad Civil 3D and Land desktop will use those triangles to recreate the DTM and contours to work off of. The contours can be used to make a surface but the surface isnt always the best as when using the triangles.
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
I'm assuming I can filter by any attribute I want with the calculator?

And I didn't know that a TIN actually contained triangles....I thought it was just points and the software generated the triangles on import. Cool to know, though. I guess it makes sense....you wouldn't need positional info for every point that way.
rrr_img_136486.jpg
 

Vepil

Gamja
<Bronze Donator>
5,841
24,502
Yeah the field calculator is on each column so you can pick and choose what you want to filter the data by.

The software using pts and breaklines will draw a TIN and interpolated pts along those triangles. The real challenge comes when you have to start changing the direction those triangles by swapping the face to get the contours to depict what is really happening on the ground.
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
Yeah the field calculator is on each column so you can pick and choose what you want to filter the data by.

The software using pts and breaklines will draw a TIN and interpolated pts along those triangles. The real challenge comes when you have to start changing the direction those triangles by swapping the face to get the contours to depict what is really happening on the ground.
The goal is really to avoid the "real challenge", then. My opinion is that the less hands-on manipulation required, the more accurate the results could *potentially* be...not to mention labor saved.

Do you think that the density of points generated with this tech (and Lidar as well) make it less likely that we need to manipulate the triangles to accurately reflect real conditions? If I can find a way to automatically filter points that aren't bare earth, I'm hoping that I can maintain enough density in the point cloud. Even then, my idea is always going to be at the mercy of the terrain we're flying.
 

Vepil

Gamja
<Bronze Donator>
5,841
24,502
I do Bandwagon, Lidar is able to generate so many pts that the contours really do a good job when generated following the ground features.

When I used to do 3D Laser Scanning we would take a point cloud and flip it to a side view and just window and remove all the points that were generated by reflecting off the trees, brush, buildings, ect... till I was happy with this. A cad program should work for this, or since most of your work is field/agriculture if you have an open area that you can get the elevation of just field calculate the elevation field to select only those in the elevation range you want to eliminate. Remove those pts and then calculate the TIN from that remaining data and just tweak till you get what you want.
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
I'm hoping to eliminate that extra hands on work...but I'm also going off zero experience with editing Lidar point clouds and little experience with doing it in photogrammetry point clouds for anything other than making them visually appealing. We'll see how it pans out.
wink.png


Your advice on the field calculator helps a lot though, that's exactly what I was looking for.
I still haven't been able to try it though. We've been going non-stop at the fire department and I had to spend my other time getting two quick flights finished.
I'm going to try tomorrow afternoon. I'll let you know how it goes.
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
So the field calculator is only usable on vector layers?

Both of the files I'm trying to work off of are rasters. I think I need to use one to mask the other before I turn it into a vector layer, so I still need to figure out how to do this part without the field calculator.
 

Lenardo

Vyemm Raider
3,626
2,534
Leica geosystems is having/sponsoring a webminar on the new FAA UAV certification testing that is beginning next month (wanting you to buy their study guide for the test most likely)

July 21 @ 11am EASTERN standard time

presenter is bryan baker of hexagon geosystems

link to register:

Registration
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
Skyward.IO is hosting a free webinar on the same thing, July 20th at 10am. Since it's regulatory requirements specifically related to aviation, I would probably check into the skyward one instead.

Vepil- haven't had a chance to play with your last link yet.
 

Lenardo

Vyemm Raider
3,626
2,534
I Hate general contractors who always ALWAYS asks...can we change this- from the designed -approved- plan. this site i actually did a flyover the other day to have pictures of the construction progress and the contractor goes, lenny, can we change to an all gravity sewer...and i looked, checked with the town for minimum cover they would allow for sewer main- and went yeah if we do the main at half a percent(which is about the design minimum for self cleaning for 8" sdr35 pvc)... then the city goes, no no we want a min of 1% slope..which means i have to change the profile of the road by a foot and change(at the low point), change the height of the massive retaining wall by a foot, change the proposed topography, etc...all while they are building the bloody ROAD. 3 hrs later and its done. wall got longer by 70 feet(to 365' long-had to change the wall slightly for conservation commission to not be within 30' of the wetlands) and 1 foot higher for 250' (bottom wall is 149 more or less- top wall around the culdesac is 156) changed the road slopes to -2.5% for most of the road to the vertical curve, then a +0.75% slope for the culdesac.

gah. got a client that wants me to video/topo his site in quincy when i have a chance...decent waterfront property adjacent(about 200' from) the neponset river...if it gets cleaned up.
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
Tons of cloud shadows in the multispectral results, so I probably won't get to try out that soil identification idea on this one.

Here's a quick video of the low res processing results, though.
 

Bandwagon

Kolohe
<Silver Donator>
24,333
65,480
Jesus, these engineering guys are selling us like hotcakes, and it's really just two of them that are offering it to their customers right now.

We're all booked up next week. That's a good feeling.
smile.png