Hi There,
have a job coming up this week flying over mud flats in a tidal area around an island.
Making it hard to incorporate GCP’s which i always use.
Have purchased an rtk phantom but are trying to determine how pix4d process’s involving GCP and RTK images?
Which does it hold over the other or do i just need the GCP as check points if all conditions with the flight are good.
Will be utilising the boat as well for the hydro survey side of the job and will create overlap with the surface to check against each other.
Can i abandon the gcp as far as processing accurately? Will the rtk images be enough to brace the data or is it just recommended as “good practice” ie survey methods etc
Any help would be great as its going to be near impossible to place enough gcp around the tide / time limitations in a proper spread / format for the area.
Regards,
Andrew
Hello Andrew, I wouldn’t suggest completely abandoning the GCPs even if you are using the RTK system. When you are using the RTK system, you don’t need the same number of GCPs as you would need for the project captured by drone without RTK. So, place a few GCPs as a control point and the checkpoints to verify your accuracy.
Thanks Kapil,
flight went well today,
processing has been successful but do have some whitish banding on the flight lines edges of the mosiac, was flying high on a sunny day over water / mud flats/ reflective.
Any ideas to fix in the processing?
Hello @Andrew, Can you forward the quality report and the few screenshots showing the issue?
Hi Kapil,
Mosaic is in the report.
We still have Hydro Survey to go and have access to the land survey as far as georeferencing / checking the point cloud adding more gcp etc if needed, only had access to part of the project at low tide to place gcp.
And from reading other posts yes we did fly into / away from the sun, unfortunately was the only way to be close enough to see the uav and change batteries for the flight mission.
Surface was the point of the flight but would of been nice to provide a better image.
7703 Sandstone Island 220223_report.pdf (2.2 MB)
Hello @Andrew,
It’s weird why you are having that issue because the quality report looks good. The only thing I would suggest to see if that helps is as follows.
- Verify the raw image quality. If there is any exposure that might be causing it.
- Add a few more GCPs(Min of 3).
- Use different calibration methods. The Alternative calibration method uses the geolocation information so, you can use that.
- If the above three suggestions don’t solve the issue. Use the rolling shutter in place of the global shutter to see if that helps.
Hi Andrew,
Just saw your post and wondered if you figured out what the issue was, for my own future reference. The banding looked to consistent to be reflectivity. Maybe a unclean or bad lens?
Hi Andrea,
The data / surface came out well with good accuracy with the check points and overlapping another separate feature survey of the island.
Flight was flown at around 10am so sun wasn’t fully up which I’m assuming was the problem with the reflection from the water?
And possibly flight direction compared to the sun?
Lens was fine, haven’t had any other jobs look like this.
Couldn’t solve it unfortunately as was a good size project.
Hi Andrew,
I will have to watch for that. I am getting ready to fly 1 job tomorrow, FAA is limiting to 82 feet so it will take around 1,300 images. Another in a month with over 3,000 at 150’. Will that take long to process?
Hi. When i compare coordinates of points obtained:
- from the ortomosaic processed in pix4d 4.8.0
CL_5_maio_2023_RGB_report.pdf (1.5 MB)
whose images were obtained with a RTK drone and - from RTK ground control
i obtain large differences, of the order of meters. I wonder if the software parameters of the transformations WGS84 → ETRS89 → PT-TM=6 are correct (is it possible to confirm these parameters in the software?) Thank you