After many hours and a few days of experimenting with DSS and Photoshop, finally produced a satisfactory (I hope) image of the Milky Way.
I can't seem to be able to get the colours the way I like them to be while trying to maintain a black background (due to light pollution?) and the purple at the bottom of the image ended up bieng empahsised (from light pollution also? Just under the camera's view, there is a brightly lit tree and a neighbour's spotlight)
Also noticed that the "colour" Milky Way goes missing past the centre right of the image. Any ideas?
And yes, the stars aren't so round.
Comments, critique and advice all welcome.
Thanks
The facts:
EOS450D
74x25s subs
ISO1600
F3.5
PS: added a reworked image. More of the Milky Way seem visible (but colour doesn't stand out as much) and the purple at the bottom is less noticable but background isn't that great. Would an image between the two versions be better?
Last edited by austinstkong; 18-08-2010 at 08:34 PM.
Reason: Added a reworked image
Not sure how you guys will see this as an achievement or anything but I nearly fell of my chair when I found Lagoon Nebula in my photo.
Since knowing that the cluster of stars near the centre of the image is M7, I decided to play "join the dots' with my photo. I did Scoprius' tail then moved on to Sagittarius. As I started joining the dots there, noticed a merged cluster of stars with a colourful glow around it which I initally dismissed as a very badly deformed stars and artefact from my camera. However, it later came to my realisation that it lined up nicely with the position of the Lagoon Nebula so I decided to zoom right in and queue falling off chair.
There it was, in the centre of my screen, though heavly pixilated, was the Lagoon Nebula!
However on second thoughts, I reaised that it wasn't that spectular (it doesn't actually look like much, and some of its colour is actually off).
Call me names but I'm not particularly in love with these images.
The focus seems to be off and the shape of the stars is not right. The black seems to be a little clipped.
The color seems to be the least problematic area to my eyes.
With the 18-200 I think you are very limited in optical quality: aberrations, coma and CAs will be a problem.
Not sure at what focal length you shot but maybe exposure was too long for round stars without tracking.
With light pollution, no tracking and the 18-200 you did a great job, but the result is not very effective.
I think you've done great given the conditions.
Hope you don't mind, I had a little play with the image in PhotoShop.
I raised the Curves and adjust the Levels. Upped the Saturation and applied a filter to reduce the stars. Also I cropped out the worst of the gradient caused by the light pollution.
Not much more I can do with a jpeg, it compresses our images so much when we save them for the web that the artifacts become atrocious.
One suggestion though, you might want to resize your image to about 45% before posting. Not only does it make it easier to view on the forum, but it hides a multitude of sins. LOL
Compare the difference to the two, the second I resized 50%.
I'd say it's definitely that pesky neighbor's spotlight and funky tree that's hampering your imaging efforts. It's probably a good thing to try & find out what it was exactly that caused the stray light, so you can take it into account (or better: block it!) next time you go out.
There's also a fair bit of vignetting happening (dark corners) - you might want to take some flats, so you can correct for it during processing.
jjjnettie gave some good tips (curves!) on how to get some more detail from your images. You might already know this, but keep your eye on the histogram as you play around with the curves and other tools. Make sure the levels in the histogram "hug" the left and the right (e.g. drop off just before they hit the far right and left), but never "crash into" the far left (=clipping in dark areas) or right (= clipping in light areas/overexposure).
Another great tip from jjnettie, which not just applies to forums, is scaling down your image. If your optics, tracking or seeing conditions are introducing errors/distortions in your image, scaling down will make them less noticeable. It sounds like cheating, but it isn't!
It's hard making anything meaningful from a heavily compressed JPEG, but I attached my two cents (I used nothing but StarWipe and StarBright + The GIMP only for cropping the usable part).
Shorter subs will reduce the star trailing Austin.
Sometimes it can take only a couple of bad subs to ruin the whole batch. Pick and choose carefully and you'll see a big improvement.
I think Austin was just excited he found M8 and wanted to share his excitement with us.
I fondly remember myself being excited and staring for a good few hours at the 8x8 patch of pixels that was the Carina Nebula!
Widefield stuff like what Austin is doing is a great way to get to know the sky. I learned heaps from exactly the stuff Austin is doing.
Besides, excitement like this demonstrates a raw passion for astronomy that's here to stay. It's good to know you have this, before the aperture fever and software envy sets in... I say good on him and I'm looking forward to more of his (no doubt much improved) images!
Call me names but I'm not particularly in love with these images.
The focus seems to be off and the shape of the stars is not right. The black seems to be a little clipped.
The color seems to be the least problematic area to my eyes.
With the 18-200 I think you are very limited in optical quality: aberrations, coma and CAs will be a problem.
Not sure at what focal length you shot but maybe exposure was too long for round stars without tracking.
With light pollution, no tracking and the 18-200 you did a great job, but the result is not very effective.
No, I won't be calling you names . I did ask for some criticism, which I got and I thank you for it.
Quote:
Originally Posted by jjjnettie
I think you've done great given the conditions.
Hope you don't mind, I had a little play with the image in PhotoShop.
I raised the Curves and adjust the Levels. Upped the Saturation and applied a filter to reduce the stars. Also I cropped out the worst of the gradient caused by the light pollution.
Not much more I can do with a jpeg, it compresses our images so much when we save them for the web that the artifacts become atrocious.
One suggestion though, you might want to resize your image to about 45% before posting. Not only does it make it easier to view on the forum, but it hides a multitude of sins. LOL
Compare the difference to the two, the second I resized 50%.
Quote:
Originally Posted by irwjager
Hi Austin,
I'd say it's definitely that pesky neighbor's spotlight and funky tree that's hampering your imaging efforts. It's probably a good thing to try & find out what it was exactly that caused the stray light, so you can take it into account (or better: block it!) next time you go out.
There's also a fair bit of vignetting happening (dark corners) - you might want to take some flats, so you can correct for it during processing.
jjjnettie gave some good tips (curves!) on how to get some more detail from your images. You might already know this, but keep your eye on the histogram as you play around with the curves and other tools. Make sure the levels in the histogram "hug" the left and the right (e.g. drop off just before they hit the far right and left), but never "crash into" the far left (=clipping in dark areas) or right (= clipping in light areas/overexposure).
Another great tip from jjnettie, which not just applies to forums, is scaling down your image. If your optics, tracking or seeing conditions are introducing errors/distortions in your image, scaling down will make them less noticeable. It sounds like cheating, but it isn't!
It's hard making anything meaningful from a heavily compressed JPEG, but I attached my two cents (I used nothing but StarWipe and StarBright + The GIMP only for cropping the usable part).
Keep up the good work!
Ivo
I love what you have done to my image. Amazed at how so much detail is still hidden in the image. (Obivously I still have a lot of things to learn). Thanks also for the advice.
In relation to subs and exposure time, how would an image with 10x1min compare to an image with 20x30sec subs.
Also, is there a way to work out how much I can leave the shutter open for before stars start to trail? (I've been using my best guess at getting an image with roundish stars and the longest exposure)
In relation to subs and exposure time, how would an image with 10x1min compare to an image with 20x30sec subs.
In an ideal world *with* tracking, there would be no difference. However, in the real world, there are many, many factors that come into play, most of which are related to the abilities of your imaging and tracking gear. However, since you don't have any tracking, shorter subs are obviously better in your case, in order to minimize trailing.
Shorter subs have a bit more noise in them relative to signal (and the signal is obviously dimmer), but if you take enough and stack them, you can overcome that disadvantage. At higher magnifications, with more, but shorter subs, you also get the benefit of having a better chance at capturing fleeting moments of 'perfect' seeing conditions (that's why some people use video for some bright objects - lots of frames to choose from).
Quote:
Also, is there a way to work out how much I can leave the shutter open for before stars start to trail? (I've been using my best guess at getting an image with roundish stars and the longest exposure)
Without tracking, the trailing starts immediately. In an ideal world, given the magnification you're using and size of the pixels on your CCD, you could calculate the time it would take for a star to move between pixels. You could then use that information to scale down your image appropriately to fit the 2 pixel trail into 1 pixel again.
In the real world, however, various factors (seeing conditions being one) 'smear' starlight over a larger area anyway, so you can get away not having tracking for a little while. How long that is exactly, amongst other things, depends on atmospheric conditions, your magnification, the quality of the optics and how much you're willing to scale down (or otherwise process - there's ways) the image afterwards.
There's definitely a lot you can do even without tracking. Trial & error and a bit of processing fairy dust will get you a long way!