Narrowband color cast after wipe

Questions and answers about processing in StarTools and how to accomplish certain tasks.
Post Reply
saublestarman
Posts: 18
Joined: Sun Aug 21, 2016 3:58 am
Location: Ontario Canada
Contact:

Narrowband color cast after wipe

Post by saublestarman »

I am just starting to use a LeNhance filter with my QHY294C astrocam. I calibrate with DARKS, BIAS and FLATs with roughly 3 hours of LIGHTs. I stack using intersection mode in DSS. I crop 5 pixels off of all edges to ensure good data. When I wipe I always get a reddish color along the Right Hand Edge and along the bottom. Does anyone have suggestions as to what might be causing this color cast. I also would like to know why Startools is not able to correct for it. The supplied image used narrowband mode for wipe but I have found that wipe has difficulties no matter what settings I use. I am using Startools v1.6.391

Peter
Attachments
Tulip Nebula LeNhance narrowband 20x 10 minute exposures GAIN=1600
Tulip Nebula LeNhance narrowband 20x 10 minute exposures GAIN=1600
Tulip Nebula Lightroom.jpg (357.91 KiB) Viewed 2086 times
User avatar
admin
Site Admin
Posts: 3367
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: Narrowband color cast after wipe

Post by admin »

Hi Peter,

This appears to be a problem with your calibration frames. Either your flats are not describing uneven lighting correctly, or your darks are not correcting for amp glow (if I had to bet, I would put my money on amp glow).

What's happening;

When constructing the gradient model, Wipe is extremely careful with removing faint nebulosity and real celestial detail (as opposed to less sophisticated manual sample-setting-based algorithms).
One of the criteria for modelling unwanted signal, is its undulation signature (e.g. the way and how fast potentially unwanted signal candidates change across the image). Celestial unwanted signal tends to have a slower undulation signature (e.g. it changes slowly in brightness over the image). Detail and faint nebulosity changes much faster in brightness from one part of the image to another.

In this case, the unwanted signal also changes very fast in brightness from one place to another, mimicking faint nebulosity. This leads me (and Wipe!) to believe the aberrant signal is not typical celestial gradient, and therefore by default, will/should not be removed by Wipe by default.

The real solution;

Figure out what has introduced this signal (e.g. my money is on a calibration frame issue)

A post-processing hack;
Use the Vignetting preset, and reduce the 'Corner Aggressiveness' parameter to the lowest value you can get away with before the unwanted signal starts showing again (as the name suggest, this parameter increases the aggressiveness by which faster undulating detail is added to the model to subtract).

Let me know how you get on!
Ivo Jager
StarTools creator and astronomy enthusiast
saublestarman
Posts: 18
Joined: Sun Aug 21, 2016 3:58 am
Location: Ontario Canada
Contact:

Re: Narrowband color cast after wipe

Post by saublestarman »

Ivo,
Fantastic answer, insights and suggestions. Looking forward to following up on your suggestions.

Many thanks,

Peter
P.S. All of the LIGHTs were 10 minute exposures with lots of Amp Glow. Most of it calibrated out but obviously some of it was left behind. I had a terrible time generating FLATs with narrowband because when I saw uneven illumination at the edges on the calibrated LIGHTs I immediately suspected my FLATs.
Post Reply