2 replies to this topic

[Niklas Gelda]

Hi everyone,

I’m currently working on setting parameters for my astrophotography data using Weighted Batch Preprocessing (WBPP) in PixInsight, and I’ve been recommended to use VarShape = 1.5 and DropShrink = 0.2 whenever I drizzle by the factor 2 and VarShape = 1.5 and DropShrink 0.8 whenever I drizzle by the factor 1. However, I’m not entirely sure what these parameters do or when it's appropriate to use them. I can't really find a source that explains the VarShape function but I’d really appreciate a deeper explanation of what they actually do and how VarShape and Dropshrink should be used in relation to each other and under what conditions.

 

Are these settings mainly useful in certain situations like poor seeing, differential FWHM, SNR or specific camera types (Mono/OSC)?

(I dither 10 pixels after each frame and my sampling is at 1.66 arcseconds/pixel. my pictures are usually undersampled)

I’m looking to understand why and when these values are recommended and how they affect the image/noise pattern, rather than just applying them blindly.

 

CS

Niklas

[WadeH237]

"Drop shrink" and "VarShape" are not specific to WBPP.  Those are both Drizzle parameters.

 

I don't know if your question even has a simple answer, but there is a good discussion of the algorithm and the effect of those parameters here.

[Peter Morrison]

The defaults for drizzle are "Square" and "0.90".  For 2x, I typically use the defaults for my ancient CCD (5.4micron pixels).  I found that using anything other than "Square" reduces the SNR and makes the background grain "chunkier".  The "chunkier" grain made noise reduction difficult.  

 

It's easy to manually redo the drizzle integration with different parameters - experiment with the settings.  Your scope and camera combination might require different settings than what someone recommends.  Drizzle might not be right for your images.

 

Peter