Welcome to the 3DF Zephyr tutorial series.
This tutorial will show you how to use RTK data for scaling and georeferencing a point cloud or a mesh in 3DF Zephyr.
This tutorial cannot be completed with 3DF Zephy Free or Lite versions.
RTK (Real-Time Kinematic) positioning technique increases the accuracy of satellite navigation systems (GNSS) on the order of centimeters.
It is based on differential positioning, using satellites and two receivers simultaneously:
If you know the base position the onboard software of the receiver applies the same corrections to the rover, thus determining its position(s).
3DF Zephyr automatically detects those pictures taken by RTK drones leveraging their coordinates to scale and georeference the 3D model.
You can import RTK coordinates into 3DF Zephyr either while creating a new project or after completing the camera orientation phase.
Click Workflow > New Project, leaving the default options enabled, and click “Next”. If your images include RTK coordinates (1), 3DF Zephyr will automatically find them in the pictures’ Exif data, setting the camera position accuracy accordingly.
The software will also select the Target reference system corresponding to the input coordinates. However, you can click the drop-down menu and select the desired coordinate system. “Select Projection CS” option (2).
The Bundle Adjustment (3) option is enabled by default and allows 3DF Zephyr to adjust the camera positions according to the RTK coordinates.
Click the “Next” button (4) and choose the most suitable Category and Preset running the Camera orientation phase afterward.
Once the sparse point cloud has been generated, you can scale and georeference it using the camera’s RTK coordinates (preferably involving at least one GCP).
From the GCP panel, click the “Align model with 3D constraints” button (1). Please, notice all the pictures and related coordinates will be automatically listed in the Chart (2). Click the “Transform and check Errors” button (3) after selecting the Target Reference System. That will allow you to roughly scale and georeference your sparse point cloud.
NOTE: it is always worth using at least one Ground Control Point (GCP) even when dealing with RTK imagery in order to be aware of the ground accuracy.
If you want to import GCPs coordinates, simply click the “Import 3D” button in the GCP panel, and find them on the images by right-clicking on each GCP label and selecting the “Edit” option. Alternatively, you can click the “From images” button to start placing your GCPs on the pictures. Once you are done, click the “Align model with 3D constraints” button (1) and then the “Import constraints from file” button (4) to load your coordinates (.txt or .csv file formats).
(For further details about this workflow, take a look at the Manage control points and distances in 3DF Zephyr tutorial).
Once the coordinates are loaded, 3DF Zephyr will automatically set the Camera position Accuracy (5) to 30 centimeters (RTK). According to your device specs, you can also change the accuracy threshold e.g. selecting the Custom option from the drop-down menu.
The coordinates reference system can be changed from the input/target reference system (6) (as explained in Option 1) and and so can the Unit of Measurement. The Bundle Adjustment optimization is enabled by default and will affect both the cameras and the GCPs marked as Constraints. For more details, see the tutorial dedicated to the Bundle Adjustment.
Click the “Transform and check Error” button (3) to start the scaling process, and an Error report will be automatically displayed when done:
Congratulations, you scaled and georeferenced your model!
To see the workflow summary with RTK drone imagery in 3DF Zephyr, take a look at the 3Dflow Academy video: