[Desktop] Processing DJI Phantom 4 RTK datasets with Pix4D

For datasets acquired with DJI Phantom 4 RTK (P4RTK) we recommend the following workflow:

Creating a project

1. Configuring the image properties:

  • (RTK) Image geolocation, horizontal and vertical accuracy is automatically recognized when importing the images.
  • (PPK) After PPK postprocessing in a third party software, accurate image geolocation can be imported using the Image geolocation file (.csv, .txt, .dat). If necessary, edit the Accuracy Vert/Horz column to define the actual accuracy values.
  • The optimized camera parameters for P4RTK (FC6310R_8.8_5472x3648) are already included and automatically recognized in the internal camera database.

2. Configuring the image coordinate system:

In the Image Properties Editor, change the vertical coordinate system to ellipsoidal heights instead of automatically detected geoid heights (EGM96). To do so:

Edited on January 24th 2019

With the latest Pix4D Desktop 4.4.9 Preview version, the image vertical coordinate system is correctly detected automatically (WGS 84 Ellipsoid). With the previous versions, the above mentioned workaround needs to be applied.

Processing options

There is no need to change the default processing templates, e.g. 3D Maps, and the following processing options can be kept:

FAQ

Can I process with the accurate geolocation pipeline?

  • Accurate geolocation and orientation pipeline can be used when both, image geolocation and orientation, are precise.
    We recommend using it for large datasets (faster processing) and when the dataset contains homogeneous or complex visual content (forest and dense vegetation, flat terrain with agriculture fields, snow, sand, and water surfaces). 
    More in the FAQ on the accurate geolocation pipeline.

Is it possible to use another vertical datum besides EGM84, EGM96,  EGM2008 or the ellipsoid of the selected horizontal system?

Do I need to enable the linear shutter optimization?

What could cause a big offset when comparing the results with checkpoints?

  • The horizontal offset could be caused by the horizontal grid corrections and transformations.
    In some cases, when flying fast, close to the ground or in high winds, enabling the linear shutter optimization could reduce the offset. 
  • The vertical offset could be present due to wrongly selected vertical coordinates system or custom geoid models
    In case the optimized camera parameters are far from the initial values, e.g. focal length, we recommend using the All Prior processing option. 

Further reading:

11 comments

Please sign in to leave a comment.