Zero or insignificant signal detected (empty image?) - lights integration

Deg77

New member
Sep 10, 2020
4
1
I am wondering if anybody can help me troubleshoot a nasty issue with my files, which I have been fighting for weeks now with very little or no success.
I use a modded Canon 6D DSLR and Voyager as automation software for astrophotography (https://software.starkeeper.it/). When I feed FITS for my flat, dark and light frames to the BatchPreproccessing script, the process fails with the error in subject at the step of the first light integration: Zero or insignificant signal detected (empty image?). Already reviewed some past similar issues on this forum, but I could not find much related to my case.
I figured out that the process succeeds if I remove the flats from the workflow, and just keep Dark and Light frames alone (have no Bias).
Also, if I use some lights from a friend of mine the integration process goes through; on the other hand, replacing the flats while keeping my own lights does not work. So I suspect an issue with my light configuration, even though they seem to contain good data.
Worth saying for those who are familiar with Voyager, I realized that I set the software camera output file as "Raw color" instead of "Raw monochrome", but I am not sure if this makes any difference.
I uploaded some examples of my file to my Google Drive storage, https://drive.google.com/drive/folders/1FmQIKQdKu1fgiimeLAbkHhdbqACdGvM0?usp=sharing
Many thanks in advance for everyone who might help or just suggest a direction for investigating!

Kind Regards
Daniele
 

pfile

PTeam Member
Nov 23, 2009
5,710
117
are you mixing CR2 and fits files?

or are all files fits files from voyager?
 

JPetruzzi

Active member
Jun 18, 2020
40
9
FWIW, I was just experimenting with Adaptive Normalization, and when I enabled it on a data set that successfully integrates without, I got that error. If I use "Additive with scaling" works just fine. Curious as to what this means.
 

dld

Well-known member
Sep 24, 2017
173
20
Hi! Your flat is grayscale while the light and dark are debayered (RGB). This can't be right.
 
  • Like
Reactions: pfile

Deg77

New member
Sep 10, 2020
4
1
Thanks for the hint ! This might explain why the BPP works fine if I take out the flats. I will try taking a new set of flats as RGB and then using them to calibrate. Finger crossed!
 

Deg77

New member
Sep 10, 2020
4
1
Hi all, in the end mixing up grayscale and RGB turned out to be the cause of the failure. Thanks to everyone who helped, I am definitely a newbie in PI and have a very long learning path to go!
 
  • Like
Reactions: dld