Hi!
I'm a PixInsight rookie, during my first attempts of preprocessing I noticed a high percentage of clipped pixels in dark and bias creation using Linear Fit Clipping.
I have sets of more than 30 frames so I decided to go with Linear Fit Clipping with defaults params and I've got 1-2% of rejected pixels; it seemed quite an high figure so I tried Winsorized sigma clipping (again: default params) and I got around 0.5% of rejected pixels.
Now, if you read here
https://pixinsight.com/forum/index.php?topic=2169.0, you know that you should expect less rejected pixels in Linear Fit Clipping and my conclusion is that the default params for Linar Fit Clipping seem too tight, moving lfit_high to 3.5 seemed the right choice because the percentage of rejected pixels dropped to values similar to Winsorized sigma clipping.
imaging details: camera is a modded uncooled Canon 700D, my BIAS and DARKS sets where taken at 800 and 400ISO, exposure time for DARKS are 4 to 6 minutes, sensor temperature is around 24 degrees (Celsius).
Questions:
1. does anybody else noticed this?
2. the real problem is that I have no idea about how to understand whether these params are good or not. If you have an image, you can see if the clipping succeeds in removing artifacts but how it is possible to understand if clipping is "right" in dark and bias frames?
cheers
Michele
