Hi,
I have read an article for volume calculation for STOCKPILE http://support.dronedeploy.com/docs/volume-measurement
I want to know how the drone is capable to calculate the height of the stockpile, and how they are calculating the volume.
Thanks
~Naman
The same principle that is used to calculate the elevation of any given point in an image (assuming two images are supplied) is used to calculate the distance of the earth to a star!
Hi htm2112,
Can you please elaborate more.
Thanks
Hi @dragonflyAS
Initially drone will capture the images, then it will develop the 3D model and from that it will hep me in determining the volume.
Am I in right path ?
Thanks
@naman14
Kind of 
you will fly the site and take your pictures. DD will process using their algorithms and deliver you a stitched map. Depending on the plan you paid for you can do the following:
Hi @dragonflyAS,
Thanks for the explanation. I want to know that how it is calculation the height i.e z axis from the 2D image.
Even it is converted into 3D model, still I am not able to understand how the scaling of x,y,z axis is been performed and how the volume is been calculated.
I want to understand the whole procedure in low level.
Thanks
@naman14
I think that the high level explanation is explained above in the Wikipedia link.
For the middle level explanation I always think of the drone as a kind of base station instrument. It is continuously tying into satellites to determine the x, y and z location of the drone. When you take a picture the coordinate is recorded and attached to the image that you took. This is great, however you are not surveying the location of the drone, as you are actually trying to survey the ground below.
Drone Deploy comes in for the low level, detailed part of the process. Through the use of software and algorithms, they are able to use geometry and math to extrapolate the ground x, y, and z properties. The process involves looking at commonalities between images at different angles (hence the overlap setting) and calculating the locations. Kinda, sorta how we look up at a stars throughout the night to determine our location on the earth.
Disclaimer, I could be way off base on DroneDeploy’s process
Hey @naman14,
Piggybacking off of the fantastic “Middle level” explanation that @dragonflyAS gave… Since you have a collection of overlapping images, and each image stores its position (of where the photo is taken from GPS of the drone), rotation, and camera angle / settings in an EXIF file given with the photo. From this photogrammetry software projects rays (also known as a raycast) from the point of origin (where the photo was taken) through the picture and into an imaginary 3d space.
By using the overlap in the photos to create tiepoints (commonalities in the photos) and repeating the raycast for the entire set of photos the rays will begin to overlap in the 3D space (in x,y, and z) from all of the pictures. This would be the sparse point cloud… think of it as a low def version, from here the process repeats a few thousand to millions of times depending on the data in each picture, creating a dense point cloud. This can be used to fill in the edges and faces to create a 3d model. To get a better understanding of this i suggest looking at the picture data in your map, you can see where each photo is and with a bit of imagination see how casting a ray down from multiple images will create an overlap and give you a 3D point. This may also give you a bit more respect for your geometry teacher in high school as this is the answer to the eternal “but when am i ever going to use this!” question.
This is a snip from a map with the images toggle activated, i tried to get a 3d view of this as well but they don’t show up in the 3d view… ill suggest this to DroneDeploy. Hope this helps to complete your knowledge!Regards,
PS Take a look at the Dense Point Cloud on one of your 3d models and it may help too!
Scott Lashmit
