Cigar and Bodes with Flux

B

Ben Eisenstein

Member
I'm a newbie to Pixinsight and trying to process the Cigar and Bodes galaxies. During processing I see some of the flux nebula between the two and do want to include that as a delicate feature. But my lack of skills fail me. If I bring the nebula out the galaxy details get overblown and the background is no longer black. So I think there might be some way to capture both. Do I need to go out and take separate exposures and then combine them somehow? Or can I do it with what I have?

Processing details:
Celestron Edge HD 8
Hyperstar
AVX mount non guided
ASI294MC with L-Pro Filter
100 subs at 60 seconds
Darks and Flats

I have processed thru integration and here is the file before DBE.
\https://www.icloud.com/iclouddrive/0Aeb5bk-IwdPR-cB_ZPylhVHw#integration_w_Cal

Here is what I finally produced. No flux :(

iCloud Drive - Apple iCloud

Store, organize and collaborate on files and folders with iCloud Drive. Easily upload, delete or recover files. Changes will sync across your devices with iCloud.
www.icloud.com www.icloud.com

Thank you,
Ben
 
Last edited:
i think it should be possible by processing normally, just making sure to never clip the blacks. near the end of processing you can try the BackgroundEnhance script (under Scripts->Utilities). this script has a pretty nice interface that lets you play with the various parameters and see the before/after dynamically. it uses multiscale techniques to isolate the large scale structures in the background and let you enhance them.

now, having said all that, the image has problems - it is definitely out of focus. also there is a pretty pronounced gradient just in the red channel. the other channels have gradients but the red channel seems to have a discontinuity diagonally across the center of the frame. i wonder if your flats might be messed up since if this were moonlight i'd expect it to affect all channels equally. i feel like before investing a bunch of time processing you should try to figure out what's gone wrong there. the focus you can't do anything about but there's at least a chance that the red gradient was caused by the calibration process.

rob
 
I knew about the out of focus problem. I should have mentioned I am new to astrophotography and the Hyperstar as well. Focusing has been a real problem with the Edge. I plan on buying an auto focuser to help with that. But I did not notice the red gradient. I took sky flats and maybe thats how it got introduced?
 
The flats and the master derived from them are from a ASI294MC, a one shot color camera. So no separate reds.
Here is the flat produced from the 20 flat subs.

iCloud Drive - Apple iCloud

Store, organize and collaborate on files and folders with iCloud Drive. Easily upload, delete or recover files. Changes will sync across your devices with iCloud.
www.icloud.com www.icloud.com

Ben
 
is there a CLS or other LP filter involved here? the flat has a very strong blue cast. the cast isn't really a problem but the SNR of the red channel of the master is about 11% lower than one of the green channels.

can you post a few uncalibrated subs? i don't see the artifact in the flat so i guess it must be in the subexposure(s).

did you blink thru all of the subs and try to eliminate any subs with clouds or other problems?

rob
 
oh - one more comment i forgot to add - there was no dithering during acquisition, right? there is 'walking noise' in the integration that is probably due to incomplete hot pixel removal during calibration and then incomplete rejection during integration (owing to the hot pixel intensity being reduced but not to 0.) it might help to run the subs thru CosmeticCorrection to try to eliminate the hot pixels before registration and integration.

rob
 
Yes, I used an L-Pro broadband filter to eliminate light pollution. There was no dithering when the lights were acquired.
Here are a couple of uncalibrated subs. They were taken at twilight. After I took them I noticed a couple of stars start to come through. But they don't appear in the master that was created so I figured they must have been removed. I have taken with the t-shirt method before when I used my dslr, but with a front mounted camera, it seemed easier to use the sky at twilight.

iCloud Drive - Apple iCloud

Store, organize and collaborate on files and folders with iCloud Drive. Easily upload, delete or recover files. Changes will sync across your devices with iCloud.
www.icloud.com www.icloud.com

iCloud Drive - Apple iCloud

Store, organize and collaborate on files and folders with iCloud Drive. Easily upload, delete or recover files. Changes will sync across your devices with iCloud.
www.icloud.com www.icloud.com

One more question for you, how did you see the red gradient?

Thank you for being so helpful!
 
sorry, i meant the light subexposures.

as for the gradient i first ran DBE on your image and then noticed a funny red streak running across the image where IFN should be (which is generally white in this area of the sky.) however, looking at this more in depth it could be just incomplete gradient removal on my part. i don't have an image as widefield as yours but i do have a pretty deep image of just M81 and M82. i registered that to my image and you can see the red stripe where the sky should be dark. there's also a blue gradient above it so again, i may have just not done DBE carefully enough. but it was pronounced enough to make me wonder if there was something wrong with calibration (or if the gradient exists in the subs, or some subset of the subs, which is why i was trying to ask for some uncalibrated lights.)

one trick is to find an image similar to yours out there on the web that's already processed and of similar depth. you can then register that image to yours with StarAlignment and then place DBE samples on the processed image. this will help you know ahead of time which parts of the sky have signal and which should be black - you want to put the samples where there's no signal. then you save a DBE process icon, close DBE, and then click on your image and double-click the DBE process icon. you can then tune the sample tolerance and run. in theory then you shouldn't remove any "real" signal from your image.

i'll see if i can't get rid of the remaining gradient in the registered version of your image with this technique.

rob

Screen Shot 2021-03-10 at 4.28.48 PM.jpg
 
You must log in or register to reply here.
Back
Top