Question about Pre-Processing..

[JohnP]

Hi All,
I have a question about PreProcessing. I am currently using the preprocessing script & apply BIAS, DARKS, FLATS & Costmetic Correction to my lights – Obviously, the script also registers the subs. If I then use ImageIntegration to stack the ‘c_cc_r’ subs I thought that it wouldn’t be necessary to use Winsorized Sigma stacking as all the defects should have already been removed. However, if I do this I still get some rogue pixels in the finished stack. If however, I used Winsorized sigma & set the Sigma LOW & HIGH rejection to 10 all the rogue pixels get removed & I end up with a very clean image. There are NOT MANY pixels rejected when I use values of 10 but it does the job. I am just curious why there are still ‘rogue’ pixels after doing darks, bias, flats & cosmetic clean up…
Any ideas?

Thks,  John.

[jkmorse]

John,

Try running a manual calibration, star alignment and integratiuon and compare the results.  I personally never use the batch pre-processing since i like more control, especially in the integration stage where I test a whole series of different rejection settings, though with Winsoried usually end up with best results in the 4.0/4.0 range.  I use the SubFrameSelector script to test the various iterations to see which gives the best results before moving on.

If nothing else, running a manual set up will give you a good comparision to the batched result.

Best of luck,

Jim

[JohnP]

Thks Jim, Makes sense - as a 'new user' I am trying to get my head around all the various techniques for doing things.. Out of interest though do I assume that if I set 10/10 for the Hi/Low reject in Winsorized sigma it will reject the fewest pixels?

John

[jkmorse]

That will certainly reject few or no pixels, but there comes a point, easily identified when you use the SubFrameSelector, that shows continued to reduce the settings by making them higher does little or anything to the image.  For me, that usually occurs in the 6.0-6.0 range and I shhot for something in the 4-5 range for each as the best compromise on preserving SNR while reducing noise and cleaning up the integrated image.

Jim