Please select Into the mobile phone version | Continue to access the computer ver.
Strange artifacts in 12MP mode (Siemens chart)
937 5 2022-7-16
Uploading and Loding Picture ...(0/1)
o(^-^)o
SrMi
lvl.2
Flight distance : 3425 ft
  • >>>
United States
Offline

To better understand 48MP vs. 12MP, I ran some tests, including taking a picture of a Siemens Star Test Chart (investigate aliasing).
Strangely, the 12MP mode creates artifacts (strange dots) within the Siemens Star, while 48MP was clean.
To make the dot artifacts better visible, the 12MP image below was magnified to 400% and 48MP to 200%.

Does anyone know the reason? The dot artifacts appear only inside the Siemens star, I do not observe them anywhere else in the image. It looks as if the special pattern triggered some bugs in RAW preparation within the drone. I have verified that the artifacts are not caused by Adobe demosaicing.

48MP:

48MP

48MP

12 MP:

12MP

12MP




2022-7-16
Use props
Geebax
First Officer
Australia
Offline

What you are seeing is moire, it happens on all cameras. If you shoot the same pattern with a high priced DLSR you will get the same effect.
2022-7-16
Use props
SrMi
lvl.2
Flight distance : 3425 ft
  • >>>
United States
Offline

Geebax Posted at 7-16 19:18
What you are seeing is moire, it happens on all cameras. If you shoot the same pattern with a high priced DLSR you will get the same effect.

The aliasing that you talk about is a different issue.
I am talking about the colored dots spread all over the Siemens Star, but only in the 12MP version.
2022-7-16
Use props
Geebax
First Officer
Australia
Offline

SrMi Posted at 7-16 19:28
The aliasing that you talk about is a different issue.
I am talking about the colored dots spread all over the Siemens Star, but only in the 12MP version.

Sorry, I missed that in your initial post. The only thing I can guess at is that something has gone wrong with the dead pixel mapping, and it only shows up in RAW mode.
2022-7-16
Use props
SrMi
lvl.2
Flight distance : 3425 ft
  • >>>
United States
Offline

Geebax Posted at 7-16 20:17
Sorry, I missed that in your initial post. The only thing I can guess at is that something has gone wrong with the dead pixel mapping, and it only shows up in RAW mode.

Strangely it occurs only within the Siemens Star, which is a very small part of the picture, and nowhere outside the Siemens Star. To me it seems as the regular chart structure triggered a bug in the binning algorithm.
2022-7-16
Use props
Bussty
Second Officer
Flight distance : 320951 ft
New Zealand
Offline

This pops up on this test too..  

For 48MP I have found taking three shots and applying a median stack (affinity/photoshop) gets rid of it but also really good results using Topaz Denoise with Colour Noise Slider increased. Good luck!
2022-7-16
Use props
Advanced
You need to log in before you can reply Login | Register now

Credit Rules