Several industries, from earthworks to civil construction, mining to waste management, are harnessing the power of aerial drone surveys to improve efficiencies on the job site. After flying your site, drone data collection software can quickly and easily process all in-flight data into deliverables like textured models, digital elevation models, orthophotos, and more.
While the task of flying the job site and transforming your data into 3D models is simple, your survey processing workflow must be precise in order to deliver accurate data that you can use. This is where the Propeller Aero all-in-one processing platform can help. Here are some common workflow mishaps to avoid that can contribute to “bad” drone data.
Problems with ground control placement
Drone surveys offer a level of accuracy that can’t be achieved through traditional, ground-based survey methods—but only if you’re using ground control. Even if you’re using an RTK (real time kinematic)-enabled drone, you’ll never achieve survey-grade accuracy from geotagging aerial data alone.
Companies that adopt an aerial survey program and use Propeller’s all-in-one platform to process data collected in-flight achieve 1/10 of foot accuracy with every survey. Propeller’s proprietary AeroPoints, or “smart” ground control points, form the basis of a PPK (post-processing kinematic) workflow resulting in a dataset that’s accurate down to centimeter-level. It’s not enough to simply fly a site with an RTK drone—ground control points attach accurate coordinates to images collected in flight, providing a greater level of reliability for your data.
Survey teams need to place ground control points carefully to achieve the highest level of accuracy from their efforts. Good ground control placement covers the entire survey area, and considers both high and low elevations and the size of the site itself. This reduces the margin of error of each survey from a scale of meters to centimeters.
Rushing AeroPoints data capture
When using AeroPoints for ground control while surveying, you need to make sure the timestamps overlap. AeroPoints need to be on the ground collecting data for at least 45 minutes to achieve this overlap. Why is overlap important? It’s the only way to pinpoint each AeroPoint’s position to achieve that centimeter-level survey accuracy.
It’s difficult to process data sets if capture times are erratic, random, or rushed. At least one AeroPoint needs to collect data during the entirety of a survey, otherwise, you’ll likely run into accuracy problems.
Setting your ISO too high
Before each flight, you need to set the ISO on your drone sensor to match current light conditions. In some situations, like low-light conditions, a high ISO is ideal since it brightens up your images without decreasing shutter speed. But, a high ISO isn’t always recommended, and if you set your ISO above 400, you can expect accuracy issues to arise when the data is processed in the Propeller Platform, since there’s too much light “noise” in the source imagery.
Incorrect coordinates
The Propeller Platform uses the “known point method” to process AeroPoints data. After each survey, you need to upload AeroPoints data and input coordinates for the known point, a stationary point on site that serves as a base station for the rest of the AeroPoints. You’ll place an AeroPoint on top of this location, and you’ll enter the coordinates while Propeller’s platform processes the aerial data you’ve collected.
You need to enter the coordinates of this known point correctly, otherwise, the platform can’t accurately process your ground control data. If your aerial data is processed against the wrong points, survey accuracy is compromised. Make sure to double-check the coordinates of your known point and enter them correctly to make the most of your survey efforts.
Propeller’s all-in-one processing platform is helping construction, earthworks, mining, and waste management teams achieve unparalleled accuracy from aerial surveys. By avoiding some of these common workflow mishaps, you’ll gain precise, accurate data from every site survey.
