Please have a play with my data... losing the will to live

[Russe]

This is M33.

21x 600s exposures

FLATS, DARKS, BIAS applied

stacked in DSS

I've attached a single exposure as jpg in comparison.

I think my callibration files might be off or my settings in DSS are not appropriate.

I've uploaded the Tif on dropbox:

https://www.dropbox.com/s/5fs4k0shyubb3jp/M33.TIF

Here is a single 600s exposure as a jpg.

https://www.dropbox.com/s/mwa9qjz2g9qjn9y/L_2013-09-15_02-58-36_M33_1518_ISO1600_600s.JPG

What do you guys think?

Please give me an idea...

[Russe]

If you do - and it turns out my data is not so [removed word] after all - please give me an idea of your work flow in PS.

I just follow Doug German's YouTube tutorials.

[russellhq]

Hi Russe, what were the sky conditions/light pollution like when you captured this? 

[Russe]

Well... LP is hefty in Cardiff...

Yesterday was totally clear. But had clouds earlier on, while I was setting up.

Here's what I do after DSS:

Convert the image to 16 bit

crop (black edge from DSS)

rx-astro gradient removal

levels (?several)

curves (?several)

match colour (luminence, colour intensity)

reduce noise

unsharp mask

oh I forget - after gradient removal, before stretching, I protect my stars with a star mask in a layer mask inverted and slightly blurred

[StuartJPP]

I had a quick play with your data in PixInsight (sorry I was playing around with it at the time). I wasn't expecting much looking at the original TIFF image as there didn't seem to be much there initially. After about 20 minutes of playing with it in PixInsight and then jumping to PS, I am pretty surprised how much is in there. I am not saying that this is as good as it gets, there is probably more to be teased out with more work. I am quite surprised with this.

Looking at the original TIFF, it seems like it was done over multiple sessions or you re-framed at some point. I have cropped slightly to remove this.

I can't really give you a workflow purely in PS as I tried a bit of everything in both PI and then PS. Mostly histogram stretching and playing with curves. If I get a chance I will try to do it all in PS, though these days I am using PI more and more. I have attached the history from both to give you an idea. I first did a heavy stretch in PixInsight before any processing.

I am sure someone will be along with a purely PS workflow...

You have captured some decent data in there...don't give up!

[Russe]

Thank you

Thank you

Thank you

Thank you

Thank you

You really put a smile my face!!!!!!!!!!!!!!

[Russe]

It's two sessions because of the meridian flip... The camera battery had to be changed too... (Waiting for my AC adapter)

I wasn't careful enough when I reframed the target and lost some edges. I stacked then via intersecting frames.

[rjc404]

I had a quick go on my version of Photoshop (CS6).

Spent about ten minutes stretching the data, balancing the colour and levelling with curves before applying a mask which you can see around the edge of the galaxy as I was a bit harsh as I working so quick but gives you an idea. Also pushed the saturation and vibrancy up a bit.

Hoping it will clear up in a bit so I can get out with my own kit as not working tomorrow and don't need to wake up early.

[Russe]

how many levels and curves layers do you add? ~2 black / 8-10 from grey and 2 or 3 curves?

[RikM]

I didn't want to stretch it so hard as it tends to explode into noise, curse of the DSLR I am afraid. There is some good data in there though. TBH I would have expected more from a such a respectable exposure time.

All in CS3

Levels - set black point to 27 for all channels

Curves - 2-3 standard curves to stretch, resetting black point after each iteration

Gradient X - run two passes Course/aggressive then fine/mild

HLVG - run medium

Reduce background noise - Blur background at 1.5 pix radius using inverse layer mask to protect bright bits

FIx galaxy colour balance - Reuse layer mask in positive mode to mask off background, apply curves in red (drop bottom and lift top), and in blue (raise bottom and lower top), flatten

Enhance star colour - Select bright stars with colour range select, expand 12 pix, feather 6 pix, copy to new layer, increase saturation 60%, set blending mode to 'colour', drop layer opacity to 60%, flatten.

Brighten DSO - duplicate layer, set to 'screen', apply hide all layer mask, paint on mask over galaxy with large round soft brush tool set to white and 3% opacity. This is cheating!

Saturation - do same again on new duplicate layer with saturation lifted 60%. This is cheating!

Local contrast enhancement - do same again to layer with several tweaks made to curves to highlight contrast differences near galaxy core. I like cheating!

Set final levels and save jpg for export - drop black point a bit and hope it uploads okay because sometimes images get an extra stretch on upload!

Sorry I couldn't do any better

[lensman57]

Hi,

I have done a quick process in StarTools and PS, the data is not bad considering that you have tackled a very difficult galaxy.

Hope that you approve.

A.G

[Russe]

Just to prove I'm not a total leech... here's my attempt:

https://www.dropbox.com/s/z8ynh7m31sp75f9/M33%20first%20go%20with%20PS5.jpg

What really irks me is, you guys' pics are blue!??!?!?!

Where's my colour???

I've done exactly what I have written above, apart from adding a luminance layer and stretching the histogram some more.

[Russe]

Wow - didn't realise that so many of you had a go!!

Thanks a million!

Could someone enlighten me with regards to colour?

And, what did you read to improve your photoshop skills?

[lensman57]

Wow - didn't realise that so many of you had a go!!

Thanks a million!

Could someone enlighten me with regards to colour?

And, what did you read to improve your photoshop skills?

Hi,

I for one have very little photoshop skills to brag about, even in my younger days when I did a lot of serious fine art photography I had no time for photoshop, hell there was no photoshop. It was all done in the darkroom. The only astronomical software that I use are StarTools and a trial version of the Pixinsight. I use the PS in the last stage for a little cosmetic process and saving and resizing images. But as you see there are some real experts with PS around, in the right hands it is a very powerful if a bit time consuming tool.

Regards,

A.G

[RikM]

If you get the basic colour balance right, then you can increase the saturation. If you do that as a separate layer underneath a luminance layer you won't add in too much new noise.

There have been so many helpful things passed along by the members of this forum, most things I have picked up on here

Outside of SGL, I bought Lodriguss' guide to astrophotography on CD-ROM which is very useful. Another set of tutorials which has been a great help is Scot Rosen's DSLR-LLRGB workflow which I can highly recommend.

[russellhq]

I struggled a with the noise and the colour, but here's my attempt.

[Pat mcevoy]

The best I could do needed a lot of stretching are u sure dss stacked all of your pics

Just a thought

All the best pat

[glowingturnip]

A bit late to the party, but here's my attempt.

By the way, did DSS bork up your colour settings ?  The file came over with a heavy red cast, which I've seen with DSS too, and I've been worried that it's clipping data if one of the colour channels is too far one way.  

After too much trial & error I use these settings:

raw/fits settings:

auto white balance

AHD

set blackpoint to zero

register settings:

auto detection of hot pixels off

stacking params:

align rgb off

dark hot pixels off

all cosmetics off

and final save, save without applying the settings.

Basically I think all the cosmetic pixel stuff messes it up, and so can the colour balance in the settings and in the final save.  I'm still not really happy with the colour balance in this one though

Cheers

[Russe]

My postprocessing skills are starting to get better... here's another go at this data, where I lost the will to live:

[GuillermoBarrancos]

I think it´s a impressive result for a long focal length 127Mak and DSLR!