Go Back   IceInSpace > Images > Deep Space

Reply
 
Thread Tools Rate Thread
  #1  
Old 04-06-2008, 10:19 AM
Alchemy (Clive)
Quietly watching

Alchemy is offline
 
Join Date: Jul 2007
Location: Yarra Junction
Posts: 3,044
Trying to get it right ..........

Have reprocessed M8, i think i figured out part of the problem with my last effort, it was stacked in deepskystacker with auto RGB balance... this has always worked for me with DSLR imaging but ithink it supressed the colors and by the time i stretched the image so it was bright enough i lost a lot of data.(hope that makes sense) thats the major difference in the initial processing between them and this has turned out better

I went looking on APOD and various places to see if i could find a definitive color balance ..... they are all sorts Red Blue Purple and everything in between.

This is the color i settled on (for now) i can easily slide it to the yellows to bring out more red, but the reflection nebulosity will goto a red as well.... im assuming this is reflection material. this image has given me more challanges than i expected, i had hoped being a bright object it would have been a walk in the park.... not so

i have tried to keep the hourglass brighter sections from blowing completely out but have lost all the colour in the glass itself, BUT i have got the fainter extensions of Ha .... Have put in a couple of 1:1 crops of some interesting bits. The whole image is a crop from the original at 40 %.

Depending on the comments i may tackle it some more , but i dont really know where to go from here. i did not do any contrast masking .... i can get away with it at 50% and it looks fine but at 100% it apears gritty .

i have put up the imaging details before so wont bother again.... im more looking for feedback on the processing.

thanks Guys

clive
Attached Thumbnails
Click for full-size image (M8-section-a.jpg)
162.9 KB21 views
Click for full-size image (m8-core.jpg)
196.3 KB21 views
Click for full-size image (M8-section-b.jpg)
135.8 KB25 views
Click for full-size image (m8-cropped.jpg)
197.4 KB33 views
Reply With Quote
  #2  
Old 04-06-2008, 10:32 AM
RB's Avatar
RB (Andrew)
Moderator

RB is offline
 
Join Date: Aug 2005
Posts: 26,635
Wow Clive, I love 'em.
Only had a quick look because I'm in a hurry.
I'll take a longer look later but for now they look great to me.

Reply With Quote
  #3  
Old 04-06-2008, 12:37 PM
jase (Jason)
Registered User

jase is offline
 
Join Date: Sep 2006
Location: Melbourne, Victoria
Posts: 3,916
Clive, pleased to see you've worked out the root cause of the issue being auto RGB balance. In the processing game - anything listed as "auto" I intentionally stay well away from. The only exception this is if I've masked the background to neutralise or de-saturate it. I feel the images shown present the best M8 you've done to date. Colour is always a subjective topic of imaging. If we base the reference of a G2V star (identical our Sun) for colour balance, typically pleasing results will be obtained. They are pleasing in that they are familiar to the world around us. If our Sun was like star Procyon with a spectral type of F5IV, our images would look subtly different. Despite what people may believe, you still need to work hard for your colour data, even with OSC cameras. A good G2V star white balance is the foundation. Take note of the objects elevation while shooting and make adjustments to the blue channel to counteract atmospheric extinction. Sure, you can shoot a piece of white paper and adjust your balance. It will probably get you close. Depends on what level of technical accuracy and perfection your striving for. Long subs exposures will ensure rich data and allows you do to more with it - i.e. stretch it really hard to push colours.

The processing appears good. Round stars that aren't "crunchy". A quick look in Maxim shows most stars have an intensity shift from edge to edge - a good thing. The loss of colour on and around the hourglass neb is a little concern, but I can't comment to possible causes based on the data you've presented. Usually any form of data stretching needs to be followed up with a boost in saturation to stop loosing the information. Keep in mind that heavy use of saturation is the equivlent of a linear stretch on your data. Its not always the best way to deal with certain situations. You many need to make small and methodical stretches and saturation boosts to you reach a happy medium. Alternatively, you can use the shadow/highlights tool which will perform a non-linear stretch and manage colour saturation in one go. You'll find that using this tool, you shouldn't need to stretch too far. As you raise the black point, your colour data will get richer. Again, small and methodical steps. If you want to try something different in colour management, the use of the soft light blend can deliver great results especially with star colours. First take your combined image and perform a hard digital development (DD) stretch. Make sure the stretch is pretty hard; avoid blowing out the white points. I like what DD does do the stars. In most cases (unless you've got really bad data) DD provides a near spherical look to stars. You get a gradient from edge to edge with a white point at center. Nice. Make sure your DD doesn't sharpen in the process. Once you're happy with the stretch, make a hard saturation adjustment. I typically go around the 170% mark depending on how good the RGB data is. You don't want to introduce noise; again, long subs will allow you to handle this. Once done, the image will look rather unnatural. Open this in PS and layer it over your originally stretched image as softlight blend. You may need to alter the opacity to keep things in check. In addition keep an eye on the histogram (including RGB split). Softlight blend is a darkening function so black clipping can occur if not careful. There are a few other colour control techniques I use such as similar to the process above, but create a mask and blend it as color. Depends on the target and the drastic measures I choose to take in the quest for aesthetics. Anyway, keep at it. Process the image over and over. Take a break and come back to it with fresh eyes. Add some more data, and then try it again. While people maybe keen to release/post their images almost immediately after they’re acquired, in most cases it is evident that more time processing the data would have delivered a much superior result. I'm no expert, but I do know that it takes dedication and effort to get great results...so take your time Above all, enjoy it.

Last edited by jase; 04-06-2008 at 12:57 PM.
Reply With Quote
  #4  
Old 04-06-2008, 01:45 PM
Alchemy (Clive)
Quietly watching

Alchemy is offline
 
Join Date: Jul 2007
Location: Yarra Junction
Posts: 3,044
Jase thanks for the reply and processing tips, will copy to a file your comments and have a play with a less processed version, will try the saturation between stretches, have not done so during initial steps to get it to a reasonable visible level... clearly a bit of experimentation to be done. Hadnt considered he G2V seriously before, ive seen it posted a number of times but hadnt put 2 and 2 together quite so simply. I havent used digital development yet, its in a program somewhere ...will have a try with it to see the effects.

rb thanks for looking
Reply With Quote
  #5  
Old 05-06-2008, 11:20 AM
Garyh's Avatar
Garyh
Amongst the stars

Garyh is offline
 
Join Date: Jun 2006
Location: Glen Innes, N.S.W.
Posts: 2,888
Very nice result Clive!
You see so many M8`s that are sooo red or magenta. You have chosen a nice natural color to follow.
Nice sharp and crisp!
Very enjoyable!
You have to put up a article on the QHY8!
cheers
Reply With Quote
  #6  
Old 05-06-2008, 11:31 AM
Alchemy (Clive)
Quietly watching

Alchemy is offline
 
Join Date: Jul 2007
Location: Yarra Junction
Posts: 3,044
Quote:
Originally Posted by Garyh View Post
Very nice result Clive!
You see so many M8`s that are sooo red or magenta. You have chosen a nice natural color to follow.
Nice sharp and crisp!
Very enjoyable!
You have to put up a article on the QHY8!
cheers
thanks gary... not sure if its completely correct, hopefully its close, i will have to mask the hourglass area and see if i can reintroduce some color from a less processed version.

RE the article... already written one, its in icemans hands, but he probably has a lot to do and one has to edit these things to make sure everyone is considered i guess. its written more from a first use point of view - how to set up the gain and offset, what sort of response and noise it has... its not definitive by any means and will need to be updated as i work things out.

cheers clive
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 07:41 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
Advertisement
Astrophotography Prize
Advertisement
Bintel
Advertisement