WBPP Processing fits Files of OIII SII results in Greyish can't reset to default Dark but Ha is Okay?

H

Hagar

Member
Hi All,
This is my first post here and need some guidance why stacking and processing SHO Fits files in WBPP my Ha comes out okay but my OIII and SII seem to come out Grey looks over stretched before I do anything not as the default Dark as the HA. (see below )?
I appreciate your thoughts
best regards Peter
1712619916325.jpeg
 
Hello @Hagar

It seems something went wrong in the calibration of your frames. Can you provide the WBPP log of your run (the one located in the log folder)?
 
Hi Nico,
thanks for your reply attached is the log file as requested(Note: I got these files and they where pre calibarated as Fits files).
let me know what you find
thanks
regards Peter
 

Attachments

  • 20240407031835.zip
    95.2 KB · Views: 19
Hagar said:
Hi Nico,
thanks for your reply attached is the log file as requested(Note: I got these files and they where pre calibarated as Fits files).
let me know what you find
thanks
regards Peter
Click to expand...

Ok, if the files you have are already calibrated then the log is not that useful. What happened if you open an OIII or S2 fit in Pix? Do you observe the same thing as with the master?
 
When I open the OIII and the SII they both open greyish but the HA opens dark when you say the master this is the masterLight.xisf files ?
the fits files are okay
 
Hagar said:
When I open the OIII and the SII they both open greyish but the HA opens dark when you say the master this is the masterLight.xisf files ?
the fits files are okay
Click to expand...

So to be clear, when you open an OIII or SII fit they are not "greyish" ?
Can you share one of them?
 
okay I just reloaded Fits file and yes it does load Grey do you still want the fits OIII if so whats the best way to upload fits file here?
 
This is indeed a strange fit file:
It is a 32 bit floating point fit and it is strangely calibrated: the normalized mean value of this file is 0.45 (the histogram pic is at the middle)

It is not a Pix problem in my opinion but I'm not exactly sure what to do at this point. I am not even sure if it really is a problem :I think you can just clip the shadows on the masterlight with the HistogramTransformation tool.
 
This is a pre-calibrated file in 32 bit float format.
nico1038 said:
I think you can just clip the shadows on the masterlight with the HistogramTransformation tool.
Click to expand...
... but it is hard to be exact. The image looks quite good if you load with float limits set to 0..65535. Lots of big telescope archive sites (HST; JWST) save images with rather odd floating point structures; I often pre-process them in MatLab, where I can histogram them precisely and work out what is happening.
1712665585957.png
 
nico1038 said:
This is indeed a strange fit file:
It is a 32 bit floating point fit and it is strangely calibrated: the normalized mean value of this file is 0.45 (the histogram pic is at the middle)

It is not a Pix problem in my opinion but I'm not exactly sure what to do at this point. I am not even sure if it really is a problem :I think you can just clip the shadows on the masterlight with the HistogramTransformation tool.
Click to expand...
Okay Nico I appreciate your time on this ..I will do some more homework
 
nico1038 said:
This is indeed a strange fit file:
It is a 32 bit floating point fit and it is strangely calibrated: the normalized mean value of this file is 0.45 (the histogram pic is at the middle)

It is not a Pix problem in my opinion but I'm not exactly sure what to do at this point. I am not even sure if it really is a problem :I think you can just clip the shadows on the masterlight with the HistogramTransformation tool.
Click to expand...
It's not strange, actually, but just the FITS format in its purest state: a bag of anything stored in any way. XISF exists to make things like this impossible, among many other reasons.
 
fredvanner said:
-15..65535 looks even more precise.
Click to expand...
Hi Fred thanks for your in put... this is interesting you mentioned above" if you load with float limits set to 0..65535" how did you do this? in Pixinsight or Matlab ? and is Matlab software you can download?
thanks Peter
 
My fits format preferences are set like this:
1712739986266.png

So when I try to open your image, I am prompted:
1712740137748.png

I then enter:
1712740217935.png

Leaving the "Truncate all out-of-range values" selected.
The image then loads, and looks like screenshot in post #12 (with an STF).
MatLab is a fairly expensive paid app, but I use it professionally, so for me it is worth it.
 
If you are (or aspire to be) Python-literate there are a number of free download maths visualisation packages buit on Python and the NumPy package (SageMath; Jupyter notebooks; matplotlib; etc).
 
fredvanner said:
If you are (or aspire to be) Python-literate there are a number of free download maths visualisation packages buit on Python and the NumPy package (SageMath; Jupyter notebooks; matplotlib; etc).
Click to expand...
Hi Fred,
first thanks for the Fits format explanation, I set this up In Pix and it works and behaves how it should on those OIII and SII files now ...this gives me control, I will have a bit of play around with this later and yes I am Python literate, I already have Jupyter notebooks; matplotlib setup I also will revisit this later and am looking at the SageMath as you suggested... I was using the program R for Fits file grading and ploting .. I need to revist as well.
thanks again Fred I ready appreciate your help
regards Peter
 
I should also have mentioned astropy.io.fits, which will handle all the nausea of the messy fits format for you.
 
You must log in or register to reply here.
Back
Top