Streaking ?

[Rezn8]

Eos 60Da, 8" Edge SCT, 80mm Guidescope, DSI, F6.3 Focal Reducer, 25C, Lumix Deep Sky Filter, 14 x180S , 800ISO, no calibration, dithering.

[tingting44]

lovely colors you captured there mate!!! what nebula is it? looks a bit like orion but i am a noob lol

[Rezn8]

Yip, Orion.

[tingting44]

Yip, Orion.

ohh im getting better lol (joke) lol

[Tim]

I used to get streaking like that in the red channel when I used a DSLR for imaging, and I never did find the proper solution for it, although I believe there are routines out there which help minimise it.

I wonder if the original RAW files were separated by channel then stacked, then re-combined, whether that would fix it.

At any rate, nice capture

Tim

[IanL]

I can see a hot pixel near the bottom left making a lovely wiggly line of red dots that matches the orientation of the streaking, thus suggesting that the streaks are a product of the stacking process. I'd hazard a guess that the streaks are due to fixed pattern noise.

1. You say 'no callibration' in your original post, and I'd suggest the first thing you need to do is actually callibrate your images with bias and dark frames (plus flats if you have them). This should noticeably reduce the streaking by eliminating the fixed pattern noise element, plus it will reduce the dark current noise in the image and probably make any residual streaking less noticeable.

2. Unfortunately Canon DSLRs suffer from 'Canon Banding' which looks like fixed pattern noise, but isn't fixed. Typically you will see noise patterns running orthogonally across the frame (either from left to right, or top to bottom). These seem to be a product some element of readout noise combining with on-camera processing of the frames. You cannot remove this by normal means (bias and dark frames have no effect), and stacking actually makes the banding more visible not less.

This could be the problem since the orthogonal lines of banding noise would align diagonally as that is the direction your dithering has cause things to stack. You could try some standard noise reduction processing in your preferred package but it will probably not deal with all of it. The only package that I know of that specifically deals with this issue is PixInsight, which has a 'Canon Banding Reduction' script specifically aimed at minimising these artefacts.

You could get a free trial of it and have a go; the learning curve is steep unfortunately, but ultimately those who become 'converts' wouldn't go back to other software!

[Rezn8]

Thanks for the info . I will try with calibration files. I used dithering for this pic. Also , I normally use long term noise reduction , I didn't this time , so I will see if that helps reduce it to. Need to buy a proper Ccd cam.

[IanL]

If you use darks to calibrate then there is no point in using long exposure noise reduction. What LENR does is take a dark frame of the same duration immediately after the normal exposure and subtract it from the image.

The advantage of LENR is that it is simple since the camera does the work, but the downside is that every image takes twice as long to expose (so for every five minutes of light exposure, you have to wait another five minutes for the dark to be taken). Also it is a single dark frame per light frame, which is going to be noisier than doing it manually.

If you turn of LENR and take dark frames separately (by covering the camera), the converse is true. Say you had six hours available to image and were doing five minute exposures; with LENR you would end up with three hours of light frames. With manual dark frames you could shoot (say) 20 dark frames at the end of the session, taking 1.7 hours, and end up with 4.3 hours of light frames. So it is a more efficient use of time. Better yet, provided that you usually image at a similar ambient temperature, same ISO, same duration, you can re-use those dark frames again in the future, saving even more time. (Generally people build up a large library of dark and bias frames over time with different temperatures and exposure lengths they re-use).

Also you can stack the dark frames together and make a 'master dark' which will be less noisy than the single dark used in LENR, thus creating a better end result.

[Rezn8]

Thanks for the info , I will have to avoid the lazy attempts, and start doing calibration files.