Community Forums

Share:

Drone work  

Page 2 / 2

thebionicman
Posts: 3019
Member
(@thebionicman)
2,500+ posts
Joined: 5 years ago

The Trimble software is free. Worth every penny...

Lee has a point. Many services require 20mp minimum (among other specs). 12mp sounds like a P3. Still a good bird but it is more limited.

Reply
2 Replies
LRDay
Member
(@ridge)
Joined: 9 years ago

2,500+ posts
Posts: 2656

I'll check into the cost of adding the module into TBC.  I don't think it is free!

Looks like the photogrammetry module added to my existing TBC would be about $5-6000.  Not happening at this time!

Reply
thebionicman
Member
(@thebionicman)
Joined: 5 years ago

2,500+ posts
Posts: 3019

It used to be a free add-on. If you have photo processing covered, I suggest virtual surveyor for data extraction. Powerful stuff for a reasonable price. Demo mode works well and they have great tutorials. Blue Marble also has software to convert and reproject ortho photos. Probaly not something you need for a one off..

Good luck 

Reply
leegreen
Posts: 1646
Member
(@leegreen)
1,000+ posts
Joined: 8 years ago

Your suggestion of "white plastic bucket lids to mark control points/panels" do not work well. Difficult to find the center and too much reflective surface in bright sun light.

These 24" aerial targets work great and last for many flights.

Reply
5 Replies
LRDay
Member
(@ridge)
Joined: 9 years ago

2,500+ posts
Posts: 2656

I planned to drill hole in lid center and place a 2 inch diameter fender washer painted black under the spike on top of white lid.  Should be able to see that.

Another associate told me some are using cardboard cake beds.  You can get them up to 16" dia, cheap, tack them to a hub stake and forget picking them up.

Those Sky High Bull's Eye look pretty good.  Might go for that.

Good to get all this experience on here.  No need to reinvent the wheel!

Reply
chris mills
Member
(@chris-mills)
Joined: 7 years ago

250+ posts
Posts: 403

Lee is correct. There is a high risk that the glare from the white lid will blot out any sign of the washer in the centre, or worse, you'll end up with the light offsetting the apparent washer position so you only see part of it on some images - resulting in your control accuracy being lower and inconsistent.

Cost of the job is far higher than the cost of ensuring you have good control points. I suggest you set up a selection of different types in a local field and then fly them under different lighting conditions. You can then select the best type for your set-up and software.

Reply
LRDay
Member
(@ridge)
Joined: 9 years ago

2,500+ posts
Posts: 2656

Maybe I could paint them with some paint that reduces the glare, put a butterfly in another color on there.  What colors are best to see in the pictures?

Reply
chris mills
Member
(@chris-mills)
Joined: 7 years ago

250+ posts
Posts: 403

Yellow is good. The real problem is using a small point to identify the centre. Depending on how the pixels fall (especially with a low resolution camera) the centre might appear as a group of 4 pixels, or it might get offset just onto a couple at one side of the centre point. A nice big black cross on the lid will work wonders.

Reply
thebionicman
Member
(@thebionicman)
Joined: 5 years ago

2,500+ posts
Posts: 3019

I use blue and white. Very little blue in nature...

Reply
Andy J
Posts: 2613
Member
(@andy-j)
2,500+ posts
Joined: 9 years ago

Interesting projects... I'm curious when all is said and done if using the drone actually saves any time/money/effort as opposed to just surveying the mile long gravel road.  

Reply
2 Replies
RADAR
Member
(@dougie)
Joined: 9 years ago

2,500+ posts
Posts: 4481

Me too...

But, I'm sure there's a learning curve. Once you've done a couple of dozen projects like this; things should be flowing smoothly and a better comparison to how long it would take; compared to how it would take TDD.

 

Reply
thebionicman
Member
(@thebionicman)
Joined: 5 years ago

2,500+ posts
Posts: 3019

If it doesn't make you money, you're doing it wrong. Of that may mean you're using the wrong tool. In my experience corridor surveys are perfect for SUAS mapping.

Reply
WA-ID Surveyor
Posts: 534
Member
(@wa-id-surveyor)
500+ posts
Joined: 7 years ago

A 1 mile long drone project is actually quite sizeable when flying at the low heights and way more then i would recommend for a first project.    You will have a significant amount of data to process.  You haven't mentioned how you plan to tie the ground control points.  The 13mp camera may not produce the required project accuracies.  We use the 20 and are always cross checking our data with on-the-ground checks, some good....some bad.  

What platform is the end product going to reside in?  You will need to work out the kinks in the workflow as well.  I would recommend a smaller open square site or two first before you jump into a 1 mile roadway topo.

Reply
2 Replies
LRDay
Member
(@ridge)
Joined: 9 years ago

2,500+ posts
Posts: 2656

What platform is the end product going to reside in? 

I do everything as far as design and drafting within TBC.  The road is very near to the house.  GCP's will be done with RTK GPS.  Once the surface is developed I could  load it into the TSC3 and check any point.  Any elevation within a couple tenths +- is probably good enough at this point.

Maybe Trimble will let me demo the photogrammetry module for a while.

I'll probably fly it all, then work on the smallest project first and go from there.  No one is going to be damaged here until I get it right.  Only loss will be my time. I can walk to this projects from my house.

Reply
D Bendell
Member
(@d-bendell)
Joined: 2 years ago

50+ posts
Posts: 73

My experience with dronedeploy is that they have a decent mission/flight app. Beyond that they're pretty ho-hum. No GCPs, or Coordinate System unless you pay extra. Pix4D is still the heavy-hitter for processing in the industry, most will agree. You can get that for two weeks free, full trial.

We're not using Bam-Tech or P4P-RTK like LeeG, but I set elevations on my GCPs from established benchmarks with the laser to get better resolution on Z. Make sure you have plenty of additional checkpoints with your ground control.

Fly when it's cloudy if you can. I prefer to paint targets and set mag nails with them in some cases. Don't skimp on your prep and planning. Take advantage of free advice/experience available. Plenty of folks have tried to pioneer this stuff and there's a pretty solid consensus on what is good and what isn't so great. "Garbage in, garbage out" applies to data collection with the drone.

Reply
LRDay
Posts: 2656
Member
(@ridge)
2,500+ posts
Joined: 9 years ago

Leaning towards painting the targets with a yellow and blue buttery fly.

I can get a 30 day trial of the TBC photogrammetry module from Trimble .  I can also use Drone deploy the next time my associate buys a month of Drone Deploy.

So that should dip my feet in the pond.

Thanks for all the advice here!

Reply
Page 2 / 2