Precision Volume Measurement in the Web Browser
All jobs processed by Maps Made Easy after February 15, 2016 have the ability to measure 3D volumes.
This sample was hand measured using RTK to be roughly 10,000 cubic meters. Try it yourself to see how the tool works. This calculated value is well within 1% of the hand measured and calculated value.
Maps Made Easy allows users to upload images taken from their drone to create high resolution orthophoto maps and 3D models.
These models can be used to measure the volume of dirt stock piles, the amount of dirt removed from a hole or check the flatness of an area.
The tool is completely web-based and there is no software to buy or download.
This popup explains a bit about how we calculated the volume:
Using the Polygon Select Tool (second tool down), the user selects the perimeter of the are area for which they would like to make a volume measurement. This triggers a few different things behind the scenes.
First, elevation lookup images that were generated at the time of processing are downloaded. Then we determine a dynamic base surface within the defined polygon. Each pixel in our lookup image has a size of 4in x 4in. For each 4in x 4in sample we get the elevation and subtract off the dynamic base surface's height to give us that sample's volume. This is repeated for every sample that falls within the polygon and the sum of them is the "Calculated Volume".
The same process is repeated for the airspace in this volume or the "Void Volume" and is useful for measuring the air volume of a hole or pit. The "Void Volume" is measured downwards from the "Maximum Height". The sum of the "Calculated Volume" and the "Void Volume" is roughly equal to the "Area of Measurement" multiplied by the difference between the "Maximum Height" and the "Base Plane".
When you are drawing your polygon, imagine a stretched surface that goes between all those surface points. This is our dynamic base calculation. The volume that is present above this imaginary surface is counted as the volume. For the purposes of calculating the void volume, or volume of a pit, we use an imaginary plane at the highest selected polygon point and calculate the volume of air down to the ground surface. This is intended for drawing around the perimeter of a level surfaced pit.
Yes very Prone to error. My 10,000m3 stockpile shows 15,000m3 in MME !!!
Unless you have a 100% flat area or a very small stockpile, the measurement tool is of little value at this stage, too bad if you paid for it in points... :/
Steven: Try it again. We are getting 10,000 m3 on your pile now. We added a dynamic base calculation to account for the slope.
Very well done :) I am impressed. (insert applause here)
I got almost exactly the same volumes using Pix4d.
Coal MME= 10,157 Pix4d=10,225
Wood MME= 10,775 Pix4d=10,770
Differences are certainly within any margin of error, and more likely my choice of boundry points. I am actually stunned that they are so close.
Pardon a nube question, but how does the volumetrics tool factor for an uneven base? Can it import base maps generated from other survey tools?
Depends on your definition of uneven.
If the ground curves or has different elevations then just select a lot of boundary points and it will create a baseplane which follows. But if you have uneven ground inside the pile which is not reflected by the sides.... No basic software can do this. You would need a full pro package (pix4d or CAD etc)
Thanks Steven - yes all large earth-based stockpiles will have such variations inside the pile as a consequence of weight and weather etc (unless they are sitting on a concrete pad or similar). We have Pix4D, AutoCAD etc. hence the question - I was curious if this online app could cater for these variations. I still think it sounds useful, however probably not for large stockpiles imho.
Thanks for the great answer, Steven. There is no reason why large stockpiles would be any different but as Steve mentioned, no software will be able to account for unknown base undulations under the surface of the pile.
Hi Ryan - Granted unknown variations cannot be factored however on large stockpiles base undulations are common and form part of the volumetrics survey procedure (usually the base is surveyed every time the stockpile is empty so the data can be imported into the volumetrics app) - We are moving towards UAV photogrammetry / LIDAR so all options are currently under consideration! - and this looks like a great online tool for smaller piles where the base is level or has a constant gradient.
This tool appears to be developed primarily for volume measurements, yet you also describe that it can be used to check the flatness of an area. Flatness (and volume to flatness...for fill/removal projects) may be just as powerful and useful. Two questions: (1) are the notations and colors relating to "contoured elevation" determined in relation to a fixed elevation surface across the shape (or trimmed photo) or is it in relation to the 'dynamic base surface' (which sounds like a surface that is based upon an estimated underlying contour using adjacent points, right?). If it is in relation to an estimated contoured base, then how can this tool be used to measure flatness? A smoothed, flat surface (e.g., a construction pad) will show up in the heat map and shape volume measures as the estimated contoured base using various points in the photo, right?
(2) Would you consider adding a feature that allows the user to specify a point in the photo that would be a fixed standard base level against which to measure the determined contour elevation (colors and volumetrics) in a selected polygon shape? This would provide a measure of flatness at 0% slope and a measure of removal and/or fill to achieve flatness. Very useful!
Hi I want to do a basic volumetric on a stockpile that is lying on a slight gradient. Do I need GCP's ? and what is the simplest method? Is it best to take a ground image at take off?
Volume measurements are basically just relative measurements so the XY accuracy which GCPs provide won't really make a difference. The slight gradient is accounted for in the calculation of the base of the stockpile. The simplest method is just to draw a polygon around the base of the stockpile with plenty of points.
I realize this is an oldish dormant thread by now, but for what it's worth, this darn volumetric calculation is working amazingly well for me. Even on quite small volumes, 49m^3 containers sitting on pavement, I'm measuring repeatedly with only a 0.3m^3 difference and just a 1-2% Error on expected volume. I bow down in awe!
Thanks Chris!
This is great. I'm new to this technology, think I'll try this with my Mavic Mini 2 . Wondering what sort of accuracy I should expect. The inevitable question from the aggregate company will be ... "yeah but how accurate is that number?" Would be nice to have an answer for him in my back pocket. Anyone have an idea?
Andrew this is noted in a few places. From the article:
"This calculated value is well within 1% of the hand measured and calculated value."
From the comment above:
"I'm measuring repeatedly with only a 0.3m^3 difference and just a 1-2% Error on expected volume"
As long as you are using 80% overlap and the stockpiles are on reasonably level ground this has been what people are getting.
I think in order to offer an accurate error figure for any would-be client customer, you would need to conduct a set of repeated measurements on piles or containers of known volumes with your equipment and crunch the numbers to see how close you can get to the actual known volume. Then calculate the %-error on those figures. I believe an acceptable industry standard is within a 3% error.