StarXterminator and small galaxies

Comments

• BlockHead

  July 2023

  Just out of curiosity... it does this on a linear image?

  -the Blockhead
• Adrian Dan Cioc

  July 2023

  Hello, yes I've tried it on the linear image. I was able to detect the gravitational lensing in the cluster, but I wondered if there is a way to remove the stars without the small galaxies. Given that they are so small I presume StarXterminator thinks they are stars.

  This is the image in question: https://www.astrobin.com/full/1g1vn0/0/#

  I am going through the PI Fundamentals, and I think there is much to improve in my processing, but for my first image, it looks ok.
• BlockHead

  July 2023

  Ah... it looks a bit undersampled (and perhaps you used BXT). So, yes, the galaxies will look like stars. 

  There is a small chance if you drizzled the date it *might* help... but no guarantees. 

  I do not know of a good (non-selective) way to do what you want. Yes, masking might help. You could blend in a GHS or masked stretch region of the cluster...but these things are very hackish. 

  (a bigger telescope would definitely help for this object :) ).

  
  

  -the Blockhead
• Adrian Dan Cioc

  July 2023

  The data was drizzled 2x. I dithered every frame just to be sure. I think I pushed BXT a little too far.

  I will experiment, since I have a ton of data on the object. For example if you look closely there are still some dust donuts remnants in the image. A couple of days ago I did experiment with a synthetic flat (tutorial found on YouTube), and to my surprise the synthetic flat did a wonderful job, also the background looked more even. Did you experiment with this?

  Another mistake I probably did was mixing darks from 2021,2022,2023 sessions into one folder, same for biases. Today I was renaming the darks and biases files with Y_2021,Y_2022,Y_2023 keywords to see if there is a difference.

  Yes a bigger telescope would help, but I love the plug and play capability of a nice refractor :)

  
  
• Adrian Dan Cioc

  July 2023

  I managed to apply the set of darks and biases for each session as it should be. I forgot that WBPP can use multiple keywords. The downside was that I had to rename about 3000 fits files. In the end the name scheme is like this:

  For the 2023 session:
  

  • For a luminance sub: Abell2218_Session_1072023_Y_2023_Lum_01.fits
    
  • For a coresponding dark or bias: Dark_Y_2023_01.fits, Bias_Y_2023_01.fits
    
  • For a flat: Abell2218_Session_1072023_Y_2023_Lum_Flat_01.fits
    

  
  

  Keywords for WBPP: Session, Y

  The resulting Luminance master was a much better.

  I discovered the GAME mask script, and I was able to apply an elliptical mask on the centre cluster, and then run SXT. The resulting star image did not contain the small galaxies, so it's a step forward