Drone mapping has been around for about a decade and in the grand scheme of things, that makes it a relatively new industry. But already it’s filled with myths and misconceptions. Here we look at 5 of the most common fallacies.
Myth 1. "I need GCPs for accurate measurements"
We’ve done a fair few videos about the importance of GCPs, or Ground Control Points for drone mapping. But are they the be-all and end-all when it comes to getting accurate measurements? It depends on what the end goal of the project is. If you’re looking to measure a roof, facade or stockpile then usually the relative accuracy will be the most important thing. By that, we mean the accuracy of the measurements in relation to one another. GCPs assist with absolute accuracy - the accuracy of where your map is located in the world. Using GCPs, you will end up with a better geo-referenced data set which can be anchored to a specific place on Earth. For many drone mapping applications, you’ll only want to know that your measurements are accurate and relative to one another, so there is actually no need for GCPs, especially when it comes to measurements!
If you want to learn more about using GCPs, check out this video.
Myth 2. "I need a high-megapixel camera to get great drone mapping results"
Fallacy number 2. Again, it’s important to consider the end goal of your drone mapping project. What does your client or stakeholder truly want or need? A high-megapixel camera doesn’t necessarily yield better results. What really matters is how the camera is used to achieve the deliverable you’ve set out to achieve. If you’ve got a high megapixel camera, some people believe that you can fly further away from a target and still get better results. That’s not necessarily down to the high-megapixel nature of the camera. A higher megapixel camera will generally be a higher quality sensor which impacts all manner of things.
To prove the point, in a recent case study, we compared a Mavic 2 Pro Camera with the 50-megapixel camera on the DJI M300. The Mavic 2 Pro comes equipped with the Hasselblad L1D-20c 20MP camera. We looked at what the two datasets looked like from different distances by simulating a scenario where the Mavic 2 Pro was closer to the target than the M300. Yes, the data quality was better with the M300, even from a greater distance, but it wasn’t because of the megapixels. It was because of the better sensor size and overall make. The M300 is designed for enterprise or high-level projects where only the absolute best quality imagery and data will do, and therefore is designed to carry cameras with bigger sensor sizes such as a P1 camera, however, not every project requires that level data quality and in fact focusing on GSD (ground sampling distance) with a smaller drone can still deliver great results!
If you're curious to get our take on larger drones M300 with the P1 payload, vs the smaller Mavic 3 Enterprise drones, please visit this post.
With more megapixels also comes more data. Handling additional drone mapping data can become problematic for both you and the client. Sometimes less is more and a point can be perfectly well illustrated without the granular detail. So, will more megapixels really make a difference on that next job?
Myth 3. "I need to capture as many images as possible to make a great map or model"
This could not be further from the truth. Again, back to the headache of handling masses of data. Yes, you will have issues creating detailed maps and models if you don’t have enough images but what’s enough? Again, consider your required outcomes. You want to capture the right amount of images that have enough overlap between them and adequately cover the site or asset that you are trying to map or model. That number could be 500, or on a larger site, 2000. You really don’t need 10s of 1000s of images for the vast majority of projects. Less images = less data = less time and hassle for the same result!
To learn more about optimal number of images and aligning images for drone photogrammetry, check out our video on captured vs aligned images.
Myth 4. "Using RTK will help me avoid signal loss during a flight"
Myth number 4 surrounds the capabilities of an RTK drone. RTK systems are real-time kinematic systems. They correct positioning errors that normal drone GPS systems suffer from. They do not boost signals to your drone. If you’re flying in areas with low satellite signal strength you’ll still have those problems mapping with an RTK drone. Remember Myth Number 1? RTK systems are designed to improve your ‘real-world’ positioning, so if absolute accuracy is important in your project, an RTK system will help, but other than that, it will have little effect on your drone's signal strength.
Myth 5. "My take-off location will have no impact on my data quality"
Wrong! Say you’re doing a roof survey and you take off from the roof as opposed to the ground. When you do your overlap calculations you may find that, by default, your calculations are relative to the ground, skewing all of your data. For a roof inspection, you’ll want your calculations to be relative (remember relative accuracy?) to the roof, therefore you will need to have things like ground offset to be able to help you with that workflow.
For example, when it comes to DJI drones, most take-off applications are relative to the take-off points so your altitude is very much a function of where you took off from. Learn more about ground offsets from this short video.
Conclusion
Drone mapping isn’t rocket science but requires a right combination of patience, experience and skills to get right. The myths that surround it often make it sound harder and misdirect you on your learning journey. The key to successful drone mapping and modeling lies in identifying your expected outcomes before you start. What do you or your client really need to see? How much detail do you really need? The accuracy and level of detail required will probably not be the same for all projects so be wary of over-engineering your solution. At the end of the day, it's the simple, consistent and efficient workflows that win.
If you want to learn more about drone mapping and modeling, our YouTube channel is packed full of tutorials and example scenarios to help you approach your next project.