Would any kind image processor have 10 mins to spare ???

[Stevie P]

I have had a few goes at this in Photoshop but feel I need some advice to get the best out of it, I would be grateful if anybody was willing to spend a few minuets on it and see what they can get out of the image, and advise on whats good (if anything) and whats bad.

The download is approx 60Mb, directly from DSS, no cropping and no adjustments, it consists of 30 x 60s Lights iso 1600, darks, flats and bias frames.

Camera Canon 1000D, 200p and HEQ5 pro. I could only manage 60s exposure because of trailing, some star trailing is evident in image.

To anybody willing to take a look, "I Thank You", and if you could offer any advice on your workflow, it would be much appreciated

Thanks again

Steve.

https://www.dropbox.com/s/gqh09ye0608dks3/m31_09_09_2013.TIF

[Davey-T]

Have a look at posts in Southerndivers thread in imaging, seem to remember having similar prob's in DSS but can't remember now , getting old

Dave

[lensman57]

I have had a few goes at this in Photoshop but feel I need some advice to get the best out of it, I would be grateful if anybody was willing to spend a few minuets on it and see what they can get out of the image, and advise on whats good (if anything) and whats bad.

The download is approx 60Mb, directly from DSS, no cropping and no adjustments, it consists of 30 x 60s Lights iso 1600, darks, flats and bias frames.

Camera Canon 1000D, 200p and HEQ5 pro. I could only manage 60s exposure because of trailing, some star trailing is evident in image.

To anybody willing to take a look, "I Thank You", and if you could offer any advice on your workflow, it would be much appreciated

Thanks again

Steve.

https://www.dropbox...._09_09_2013.TIF

I think that you have done a great job with 60s exposure, I have done a quick process in StarTools and finished off in PS. The fact that you have managed to to get DSS to stack the frames with so much trailing is great but looking at the stars there appears to be a bit of jerky motion from the combination of the mount and the large scope, anyway I hope you approve. There is very little colour in the capture BTW but this is typical of the DSLR and you probably need much longer subs in great nos to get any colour. Thanks and good luck.

A.G

[Stevie P]

I think that you have done a great job with 60s exposure, I have done a quick process in StarTools and finished off in PS. The fact that you have managed to to get DSS to stack the frames with so much trailing is great but looking at the stars there appears to be a bit of jerky motion from the combination of the mount and the large scope, anyway I hope you approve. There is very little colour in the capture BTW but this is typical of the DSLR and you probably need much longer subs in great nos to get any colour. Thanks and good luck.

A.G

Thanks for taking the time to have a look, I think I am trying to pull to much from too little data, tending to get to much whitness leading to loads of noise...

[lensman57]

Thanks for taking the time to have a look, I think I am trying to pull to much from too little data, tending to get to much whitness leading to loads of noise...

Hi,

The core is not too bad because it is so bright, I have only tried once for M31 but it was a failure . The dark parts and the outer rim are the noisy bits. I would also wait untill it is almost overhead as this helps with the signal and the reduction of noise. I also think that you'd probably need two sets of subs, one for the core and the other for the outer rim and and the dust lanes these could be processed separately and then combined in PS .

Regards,

A.G

[glowingturnip]

Here's my attempt:

And if you're interested in the interim files and final tiff they're uploading here (might take a while) - http://www.stupey.com/Astronomy/StevieP/.  Apologies, I only have Photoshop Elements so had to drop it down to 8 bit when I got there.

Steps taken:

Maxim:

- deconvolve

- gradient

- digital development processing

- small curves adjustment (dropped the very low end a bit to lose some of the noise, boosted the next lowest a tad to get some of the fainter detail out

Photoshop:

- despeckle, dust&scratches, colour noise

- layer masking - made 4 layers of the image, 1) top one for the bright details with a layer mask, highpass overlay 40% 5px, unsharp mask 40% 3px, all to bring out those dustlanes, tried to avoid washing out the core 2) first layer underneath with a broader layer mask for faint details, smart blur 1px, bit of lvl adjustment to bring out faint detail but drop some fog, 3) a layer of everything that's not black, just to keep the background stars and 4) everything else, just for the background, smart blur 3x, saturation low to remove any colour cast, lvls to remove some fog.

I couldn't find any colour information in there either, which is a shame coz those dustlanes come out a lovely brown (see my attempt in the link below).

Also, i'm not sure that there is much tracking error in there, more lots of coma - if you look at the stars in the middle they're quite round, but the ones in the corners are all distorted, 'pointing' to the middle.  You might want to crop for a final image ?

I'm surprised you had such a high histogram and so much noise for 60s exposure to be honest, maybe play about with lower ISO's ?

Cheers,

Stuart

[Stevie P]

Here's my attempt:

m31_09_09_2013.jpg

And if you're interested in the interim files and final tiff they're uploading here (might take a while) - http://www.stupey.com/Astronomy/StevieP/.  Apologies, I only have Photoshop Elements so had to drop it down to 8 bit when I got there.

Steps taken:

Maxim:

- deconvolve

- gradient

- digital development processing

- small curves adjustment (dropped the very low end a bit to lose some of the noise, boosted the next lowest a tad to get some of the fainter detail out

Photoshop:

- despeckle, dust&scratches, colour noise

- layer masking - made 4 layers of the image, 1) top one for the bright details with a layer mask, highpass overlay 40% 5px, unsharp mask 40% 3px, all to bring out those dustlanes, tried to avoid washing out the core 2) first layer underneath with a broader layer mask for faint details, smart blur 1px, bit of lvl adjustment to bring out faint detail but drop some fog, 3) a layer of everything that's not black, just to keep the background stars and 4) everything else, just for the background, smart blur 3x, saturation low to remove any colour cast, lvls to remove some fog.

I couldn't find any colour information in there either, which is a shame coz those dustlanes come out a lovely brown (see my attempt in the link below).

Also, i'm not sure that there is much tracking error in there, more lots of coma - if you look at the stars in the middle they're quite round, but the ones in the corners are all distorted, 'pointing' to the middle.  You might want to crop for a final image ?

I'm surprised you had such a high histogram and so much noise for 60s exposure to be honest, maybe play about with lower ISO's ?

Cheers,

Stuart

Thank you for taking the time to have a look, great job.

My Photoshop skills are very limited, so really never sure what the right thing to do and in what order, so your work flow is much appreciated.

Would longer exposures give the "missing" colour ?

One of the many problems I had when attempting to process the image was noise, and everything I tried just seemed to make it worse.

This particular set of images was taken at ISO 1600, next chance I get, I will try at ISO 800.

Thanks again for you time and advice

Steve

[glowingturnip]

not sure why the colour's not there tbh, maybe someone else can chime in.  Would assume longer exposures will help though.

As a fellow DSLR user, I have a lot of problem with noise too, though as I understand it, beyond a certain ISO, depending on your camera, you're not adding any more useful information per photon but are adding noise (I can't find the post I read it in, sorry).  I guess, find a happy medium ISO, and then just take looottts of frames to average out the noise.

Has anyone invented a cooler bag for DSLR's ??

[Naemeth]

I'm not really an imager yet either, but from what I've picked up:

More subs increases the signal to noise ratio (SNR) as you can subtract the noise in each sub from each other.

Longer subs increase detail and brightness.

In an ideal world you'd have 20 or 30 minute exposures (assuming you managed to cool the DSLR that is! Better with a CCD for less heat noise) and lots of them, but that's not always practical.