# 25 May, 2018 23:02
I am having a difficult time processing data that Ive collected on NGC7023, The Iris Nebula. I have about 12 hours worth of data stacked and although I can bring the dust clouds out of the data the resulting image does not look very desirable.|
The data was captured through an ED80, imaging with a modified 70D. Data was stacked in DSS and am doing post processing in PS.
This is what I have so far. I would like to bring the dust clouds out without killing the rest of the image.
Thanks in advance Jer
# 26 May, 2018 10:59
Seems you have flat fielding problem.To bringing out faint details,Flat fielding is the most important part of process!
# 26 May, 2018 18:10
|what is the cause, or causes of flat fielding?|
# 27 May, 2018 00:31
Hi Jer, I would say that 12hrs on this object should show much more details, and also lighting up the nebula’s areas comparing to the dark ones. Maybe tell us what is your DSS stacking method. By the way, I have made some tutorials for both DSS and PS for dlsr image…but in French Let me know if it could help.|
# 27 May, 2018 12:33
I am imaging in about as bad of conditions as you could possibly get as far as light pollution goes so it takes allot more exposure time to gather enough light to work with.|
Ive actually tried stacking in few different methods. 1st was to stack the RAW images in DSS using HDR method and then also trying the Median method.
Then Instead of stacking RAW I also tried correcting each frame "white balance" "levels adjustment" ……. then saving each as .TIFF files and then stacking those frames in the same ways. Which actually seemed to work a little bit better than stacking RAW un-balanced frames.
For all stacking RAW Digital Development Process Settings Are brightness: 1.0, red scale 1.0, blue scale: 1.0 and then AHD Interpolation. With Black point set to 0
# 27 May, 2018 21:00
I actually apply those settings on DSS 4.1.1 when stacking RAW from my Canon 600D Astrodon :
- RAW settings : L=1.0, R=1.0, B=1.0, no white balance, Bayer transformation sets to "super-pixel", black point = 0.
- Stars detection : 800 - 1'000 stars for better alignement
- Result : mosaic method
- Image : kappa-sigma clipping (Kappa : 2.00 - Iteration : 5), RGB background calibration
- Flat, bias and dark : median
- No cosmetic adjustment
After stacking, I usually save the image unprocessed as a FITS (for PixInsight) or TIFF (for PS CS2). I guess that PS CC is able to read FITS format. I do not apply any curve or color adjustment in DSS.
Hope it will help a little
# 30 May, 2018 20:27
The one thing that I have not been doing, but will have to start taking the time to do is applying correction frames. Ive never really used darks, flats and biased frames. It's difficult when you have a limited time out to begin with. I barely have enough time to capture the light data that I do, though im going to have to find time to capture an equal amount of flat frames.|
I'm assuming flats have to be taken each and every time out if you move your optical equipment at all. Even turning the focal reducer, or indexing the camera by a few degrees would make any previously captured flats useless. Or am I missing something.
# 31 May, 2018 09:02
Correct. Flats have to be taken if you change anything in the optical way. But they can be taken very quickly if you have the possibility to give a lightfield into the scope. I am using the lacerta flatfieldbox and it takes me less than 2 minutes to capture 100 flats and the corresponding flatdarks.
# 31 May, 2018 13:20
I was thinking of deleting this post. I know now you do not have PI. I'm going to leave it though in case you decide you want to check it out with the demo version.
Why not just process these images in that program from end to end? My own experience is it does a far better job. You have to make a master bias, dark and flat, then you go through a structured approach to producing your image. The following is my basic approach and would work for your 12 hours of data.
1. Calibrate - clean up all frames with your master frames mentioned above.
2. Debayer - get color versions of the calibrated frames
3. Registration - prepare for stacking
4. Integration - stack frames
5. Linear Fit -align color channels - a form of white balance is the best I can describe it. It generates a new image at this point
6. Dynamic background extraction - get rid of your light pollution gradients and color casts - using the linear fit, DBE generates a new image - this is the one that you work with for the rest of your editing.
7. Multiscale linear transform - reduce noise
There is more, but the above gets your basic image. The Light Vortex and Harry's Astroshed videos are very informative to get your workflow going in PI.
# 31 May, 2018 14:00
"I'm assuming flats have to be taken each and every time out if you move your optical equipment at all. Even turning the focal reducer, or indexing the camera by a few degrees would make any previously captured flats useless. Or am I missing something."|
This is how I understand it and it has worked in practice for me.
Your flat is taken out of focus against your calibrated light source to profile the image path so you can correct your RAW images for those flaws during processing. You are mapping the optical path against your sensor for things like vignetting, sensor spots, etc. Turning your camera or adjusting focus does NOT require a new flat unless you have a radical issue in your optical path and that would need addressing from other perspectives. Also if you shoot a different part of the sky, that is not an issue either for your master flats, darks, bias files, they are the ones that get you back on track.
Now, if you removed your focal reducer to get a longer focal length, then yes, you need a new flat. Change telescopes, new flat, etc.
# 02 Jun, 2018 17:14
Knowing your imaging conditions, I must agree with the comments above. Calibration frames are almost as important as good lights. In you conditions the signal to noise ratio is pretty bad. Calibration frames will help clean it up. They can be taken at any time, not necessarily while imaging your lights. With the weather we have been having this winter and spring, any clear night is precious time for imaging lights and cannot be afforded to be hated on calibration frames.
Darks can be taken any time as long as there are taken at or near the same temperature. Same with bias. These are important to remove noise from the camera. On any long exposure, especially with an uncooled camera, the noise will increase with temp., and the longer the exposure the more noise will be generated from heat from the camera itself. Subtracting the noise will dramatically improve your lights.
Flats will remove any defects in your imaging train. Dust doughnuts, vignetting, etc. It will also help with pulling out faint details such as the dark lanes in you Iris image. When I started processing I was doing darks and bias . but not flats and I was asking the same question about my images. I found that flat fielding was the answer. Of all the gizmos and gadgets I have bought with this hobby, my Alnitak Flat Man was probably the best money I spent.
You can get good results with sky flats using a white tee-shirt over the objective. These grey skies we have had this spring are perfect to shoot flats. There are some good videos on you tube using this technique. These can be done anytime. no need to match the temp on these. Also if you try not to remove you camera from the train, or rotate it, you can reuse these over and over.
The same holds true for you darks and bias. You can use them over and over as long as they are at the same temp or close to your lights.
Clear Skies–MAYBE SOMEDAY!!!
# 05 Jun, 2018 21:33
Thanks for the info and advice guys.|
I definitely have to start making time for calibration frames. Ive never really had any issues before, though I think I was just so impressed with capturing anything at all that it was good enough at the time, if that makes sense. Id really like to improve my work though and from what I see there really is no point in collecting data if I can't process it correctly.
David I am seriously considering PixInsight. I have been working with Photoshop for well over a decade though. Ive tried other progs and nothing feels right except for PS. Though I am sure if I spent enough time working with it I would develop a workflow.
Yeah Bob I try not to let the weather get to me, though it's starting to become very irritating. We get a night, or two of clear skies then its completely overcast for 2 to 3 weeks. I honestly don't think we've had more than 2 weeks worth of clear skies since last October. Hopefully the pattern will break though. I think it was pretty crappy last year up until about June. June straight through September though it felt like I was imaging, or observing the night sky on almost a nightly basis.
Thanks again… Clear skies
|You have no new notifications.|