Volumetrics

Same map 3 different processing companies and all way different measurements. Took great care in making the measurement as close as possible but the numbers are so far off from each other it has made me say who do you trust? MME - Precision - DD. I am a commercial business and this is just not good for business. Some comparisons off by 10,000 yards!

Hi John,

Can you share both the input data and the output maps? Then we can investigate what’s up. We’ve seen 2% accuracy on stockpiles compared to laser scans.

James

Are you guys not using GCP’s?

Actually thata irrelevant because same data set and 3 different software processors should be close gcp or not.

Use of GCP’s affect global accuracy. Relative accuracy shouldn’t be nearly as effected.
Isn’t that correct?

Hi there,

Can you share the names of the map(s) you’re having issues with as per James’s request?

Thanks,
Christina

What are the average sizes of the stock piles in square feet? Gcps will make you difference if they are rather large. Running three different companies with three different algorithms without using gcps will not be as comparable as you think. I’m guessing by the fact that you got a 10k CY bust in the piles that they are pretty big. Are you taking any oblique images?
Our experience is at best you should not expect more than 2 to 5% accuracy between any two stockpile surveys. Just because of the algorithms. This includes traditional survey versus laser scan versus photogrammetry.

I have seen the same problem over and over again. We have done comparisons with and without GCP, different flights on same day, various processing methods (DD, Pix4D, TBC, Civil3D) and as a result lost confidence in DD. Results on same pile differ on every flight from 5% to 25%.

Hi @GRP1,

I’m sorry to hear this. Can you provide the info as requested by @Jamespipe’s post above so my team can investigate?

Cheers,
Christina

Thanks GRP1, we are seeing some pretty big differences and it has cost me some big money. So until I can find a company or software on my computers (much prefer) I am not going to push volumetrics. I am going to have a meet with a Civil Engineering / Surveyor co and pick there brains because all this back and forth is a big waste of my time. I am not in it for the hobby its all about making money. Best of luck folks.

Hi @candidskyproductions,

We’re sincerely trying to help you out and figure out what’s going on, but in order to do that, we need a bit more information. If you’re able to share the info that James requested above, my team would be happy to look into this.

Cheers,
Christina

Have you heard of Carlson Precision 3D? Great tool for editing point clouds and including outside sources of information like survey points or model content Civil3D/Carlson Civil. We collect checkpoints that can verify the point cloud and it has been the only way we can guarantee that our output from DD is good enough. The variance between the checkpoints and the point cloud can also help you make fine adjustments. One way we have experimented is to make two surfaces, one with and one without the checkpoints and staking to that surface in the field to see the cut/fill between actual ground and processed data. Checkpoints have been better every time.