Comparison of flight height and overlap % for mapping resolution


#1

I have a 50 ha property that I wish to map.
(125 acres)

I’ve inserted old school units for those who need them.

I’ve made a decision to only use 6 batteries for the job and a flight time of about 90 minutes.

I’m new to DD/drone mapping and would appreciate some advice from experienced users.

I need to make a choice between low flight height at 60 metres (200 ft) with higher side/front overlap ratios or 100 metre (300 ft) flight height with lower overlap ratios.

In order to achieve good quality mapping which should I go for?

What is the minimum side overlap one should use?
Does the front overlap only affect flight speed? - and therefore it should be as high as possible?
Or do I just keep both overlaps on 75%

…and why does DD not provide an total camera image tally parameter during flight planning? Surely this is a most useful variable to be given as an outcome of one’s flight planning?

Similarly does the total time parameter take into account the reduced flight speed? - as the flight speed parameter is not real time anyway - don’t even know why it’s there in DD - it’s about as useful as a missing prop on my P4 pro!

Appreciate all feedback.


#2

Hi Daniel,

Thanks for all your feedback. I hope I can help.

To briefly answer your questions - increasing frontlap will reduce flight speed at lower altitudes. This is because the camera can only capture images so fast.

Max flight speed should not be changed unless flying in very low light to reduce motion blur.

Sidelap should generally be increased after frontlap as it also increases flight time, but having very high frontlap and low sidelap will not produce as accurate stitch as more even numbers as you will have image density banding.

We have a couple of apps that might help you, while we are redesigning our interface to improve it.

First image count estimates:
https://www.dronedeploy.com/app-market/flight/tpuhhjdwihwstroyjguv.html

Second, overlap optimiser, uses data from over 10,000 maps to make recommendations:
https://www.dronedeploy.com/app-market/flight/wmxdbydyrrofceqaoxzq.html

Assuming the buildings and trees you are flying over are not more than 20m tall (or not more than 20m elevation gain overall), and you are flying 60m (200ft), You should get excellent results (2cm/pixel Ground Sampling Distance) with 73% sidelap, 76% frontlap. That will be about 1500 images, 5 batteries and sufficient image density to get good coverage.

If you fly at 100m you can do the same job at 3cm/pixel GSD and 66% sidelap with 75% frontlap and only 2 batteries and 441 photos. The processing will take less than 1/3 the time, but quality of 3D objects slightly lower.

Thanks,

James
DroneDeploy


#3

Thank you very much James!


#4

Hi Daniel,

Quick follow up. I’ve now also released another app called Auto Flight Modes which automatically optimises your flight plan based on the height of the objects in your map.

You’ll find it in our App Store!

James


#5

Have you looked at the difference in height as it relates to generating proper orthomosiac and 3D models? I had one of my pilots fly a stadium recently and the parts of the structure that are straight are showing up significantly skewed. I’m thinking it might have been in part because of the height of the flight. He flew fairly low over the structure then added a large number of angled shots but it didn’t help. When I take the point cloud model into ReCap or Navisworks it is also rotated at an odd angle. Not sure if it’s related but figured I’d put it out there. Any suggestions?


#6

Hi @RAGlover,

We recommend adding Ground Control Points to generate accurate orthomosaics and 3D Models. Without GCPs, you can expect to see skewed results that may have discrepancies. That could also affect the point cloud model when imported into another program. However, you might need to re-project (change the map projection) of your LAS file to correctly display your point cloud. Please refer to this Recap forum thread for more details. I hope it helps!

Best,
Stephanie