›

Drizzle and star "holes"

Dieter Lubbe
in PixInsight

Hi Adam, having issue with dark “holes” in stars after 2x drizzle in Pixinsight. Imaging system is RedCat 51 with ZWO 2600MC so undersampled. Similar issue with WO FLT91/2600 combo. 65 subs with good guiding, focus etc. Not affecting non-drizzled images. It is not because of over saturation as only affecting medium brightness, medium size stars. Same issue with manual drizzle with drizzle integration outside of WBPP. Have changed some parameters in ImageIntegration including using Winsor Sigma Clipping instead of auto but no difference. There were a few posts in 2021 on CN and Pixinsight forum regarding similar problem by some users but no definitive answer I could find. Juan from Pixinsight also commented with explanation I could not really follow but suggested doing it outside WBPP but not clear what change to make to settings outside WBPP. Multiple images affected by this, but only when drizzled. Drizzle settings pretty standard 2x, 0.9 and square. Any help appreciated. Thank you. PS, really enjoying your content.

Comments

  • Dieter Lubbe
    zoomed in stretched in ScreenTransfer, unprocessed
    Star holes.png
    1914 x 1338 - 679K
  • BlockHeadBlockHead
    Someone just asked this...
    Turn off high range rejection in ImageIntegration. 
    There are a number of threads on this (official forum)...and this is the answer in general.

    -the Blockhead
  • Dieter Lubbe
    So I tried unchecking "clip high range" and separately tried unchecking "clip high pixels" in ImageIntegration, also tried unchecking "enable pixel rejection" in DrizzleIntegration, but no change with any of these with multiple separate runs. Any confirmation that this worked in prior cases you refer to? I went through your videos on ImageIntegration and Drizzle again to make sure no other settings off but nothing stands out. Very happy with the undrizzled masters but keep getting these abnormal stars with drizzle, quite frustrating.
  • BlockHeadBlockHead
    There are a couple of different areas on this topic. My initial guess is based on this:
    Find Juan's answer on the first page.

    There was a suggestion this is an issue with CFA maybe?
    There are different threads that address this.

    -the Blockhead
  • Dieter Lubbe
    Adam,

    Thanks for following up. Unchecking CosmeticCorrection in WBPP did it. I had seen that post regarding CC but had not tried it as Juan subsequently commented and steered to rangecliphigh as culprit. Thank you.

    PS:  Can I make a plug for continued recorded image processing sessions as you have been doing on the Horizons platform, incl OSC personally, that is really incredibly helpful as supplemental to the Fundamentals theory in the ever changing PixInsight workflows as processes and scripts evolve.

    Thank you again
Sign In or Register to comment.