M31 for critique

[Photosbykev]

This is the first time I've manage to get a reasonable sequence of any astro feature and would appreciate some hard hitting criticism in order to improve the technique or even change the technique.

This is M31 taken on a Canon 5D MkII and Canon 600mm f4L prime lens

a fullsize uncropped image can be seen here http://www.photosbykev.com/wordpress/wp-content/gallery/astro/m31_composite_v1ff.jpg

Settings

Camera was controlled via USB and serial comms with APT

Iso 1600 @ f4 all images shot in raw with NR turned off

10 x 8 second

10 x 30 second

10 x 120 second

NEQ6 mount controlled with EQMOD and PHD guiding

no darks or flats yet so there is some noise but not bad for iso1600, they are something I need to do on a cloudy evening.

Processed in DSS and then exported to CS5 for basic levels and curves adjustment

[Sparrow]

I would say you've got some really good data there but it's crying our for flats to be added to the processing as it's being swamped by the vignetting.

There's some quite advanced techniques.equipment being used for taking flats, but to be quite honest, so far I just fix a piece of white A4 paper to the front of the scope, point it at the bright daytime sky and taking my flats. Even illumination is the key so that the flats give an accurate interpretation of the vignetting. I usually take mine so that the histogram is well towards the right of the graph and it seems to work well.

Obviously darks will help with noise, but don;t forget that they need to be taken at the same settings for exposure, and at the same temperature. Ideally they should be taken at the end of your imaging session.

Otherwise - I reckon you've got a very promising image in the making there.

Regards

John

[steppenwolf]

What John said above. We all try not to take flats when we start imaging (come on folks admit it!) and there are some excellent tools like Gradient Xterminator that will help to remove gradients but these tend to be pretty destructive with your data. Once you see the improvement that flats will give you, you will never look back!

This is your image with the Gradient Xterminator filter applied but there is no substitute for flats:-

[jgs001]

that's come out great Kev, what they said about flats... I use a laptop screen with Al's virtual light box held over the objective of the scope (or in this case, camera lens ), set the camera to Av mode and let it work out the exposure, works a treat.

Nice edit Steve.

[russ]

That's a great image, you've captured some excellent data. Nice job! Just needed those flats.

What John said above. We all try not to take flats when we start imaging (come on folks admit it!) and there are some excellent tools like Gradient Xterminator that will help to remove gradients but these tend to be pretty destructive with your data. Once you see the improvement that flats will give you, you will never look back!

This is your image with the Gradient Xterminator filter applied but there is no substitute for flats:-

wow that's amazing Steve. Although there's no substitute for flats, would you recommend buying Gradient Xterminator?

[steppenwolf]

Hi Russ, yes I would recommend the software as there are other gradients that can creep in that flats will have no effect on and Grad. X will work well on these. Grad. X wasn't designed for resolving vignetting problems per se but as you can see, at a pinch, it will do a fair job of it. What I don't like is the small 'halo' of darkness that it often leaves in its wake. However, it is a great piece of software that has saved several of my images from the celestial dustbin and it is sensibly priced too at £31.85

[russ]

Thanks Steve. I've downloaded the plugin and requested the trial key. Will give it a whirl a bit later.

[Photosbykev]

Thank you for the advice so far, I've just shot a Master flat using a EL panel. 30 exposures at the same iso but the shutter was adjusted to give me a peak about a stop short of pure white on the histogram (if that makes sense)

This is an unmodded sub frame from the flat set. for a white EL panel there is a lot of colour lol

When processed the master flat tif was surprising grey so I've stretched a copy using auto levels only purely to see what was going on it, I'll use the master file for stacking

This is the unstretched version

This is the stretched version

I've no idea if this is what a flat should look like but it seems logical to me ?

The camera is currently outside in the car shooting a set of darks that will be close to the same temperature as last night, it just hasn't warmed up at all

[steppenwolf]

Yup, that's a flat and as you can see, it shows the vignetting rather nicely!

[Photosbykev]

Yup, that's a flat and as you can see, it shows the vignetting rather nicely!

For a £6000 lens I want my corners back !

[Photosbykev]

I've taken onboard the advice and this is the second version processed with darks and flats. I also processed the 8 second exposures of the core separately and added them back into the image as a layer and played with the opacity to get a reasonable match.

It's getting better, I think

[Photosbykev]

This is a link to the 120Mb full frame tiff file if anyone wants to play with it M31_composite_v2_flat.tif

[Beyond_Vision]

Looks like you now have the flats sussed out. You just need longer subs with a total time of over 3 hours or more

Did you stack all the various exposure times as groups within DSS or did you combine the stacks from each exposure length in PS? With M31 you can use for example 10 minute subs or more to bring out the fainter parts. A less stretched version of the image can then be blended to stop the core from being blown out.

[Photosbykev]

Looks like you now have the flats sussed out. You just need longer subs with a total time of over 3 hours or more

Did you stack all the various exposure times as groups within DSS or did you combine the stacks from each exposure length in PS? With M31 you can use for example 10 minute subs or more to bring out the fainter parts. A less stretched version of the image can then be blended to stop the core from being blown out.

Looks like i did something wrong

I loaded all of the 30 and 120 exposures as lights and loaded the 30 and 120 second darks as darks, I didn't play with the groups at all. I assumed DSS would match the light and dark exposure time automatically I then processed the 8 second exposures separately and blended the result in CS5

What averaging should I be using? Average, Median or the Kappa-Sigma option

[Psychobilly]

Have a look at auto adaptive weighted average in DSS...

I find this works best if you have realtively high numbers of subs at each exposure

Billy...

[Photosbykev]

Have a look at auto adaptive weighted average in DSS...

I find this works best if you have realtively high numbers of subs at each exposure

Billy...

Thank you Billy

edit auto adaptive weighted average just gave me a pure white final image I must of done some thing very wrong lol

[Psychobilly]

Oh thats wierd...

You can tell its ages since i used it - Entropy weighted is the one you want...

Billy...

[Beyond_Vision]

DSS has probably applied a stretch to the image which has saturated it.

[Photosbykev]

Oh thats wierd...

You can tell its ages since i used it - Entropy weighted is the one you want...

Billy...

No problem running DSS with the new process now

[Beyond_Vision]

Kev this image I took was just a stack of 10 minute subs with no layering for the core. The HDR wavelet tool in Pix-Insight can be used to avoid totally blowing out the core.

I could have taken this a step further by using an unstretched image and blending but didn't bother.

http://stargazerslounge.com/imaging-deep-sky/117123-andromeda-unchained-reframed-reworked.html

Regards

Kevin

[MikeWilson]

Looks good, others have given all the advice that I would have given and much more besides

Just a quick question for you - 8 second exposures... I've never considered taking anything less than 30 seconds on M31. Why 8?

[Photosbykev]

Kev this image I took was just a stack of 10 minute subs with no layering for the core. The HDR wavelet tool in Pix-Insight can be used to avoid totally blowing out the core.

I could have taken this a step further by using an unstretched image and blending but didn't bother.

http://stargazerslounge.com/imaging-deep-sky/117123-andromeda-unchained-reframed-reworked.html

Regards

Kevin

Now that's a standard I need to work towards

[Photosbykev]

Looks good, others have given all the advice that I would have given and much more besides

Just a quick question for you - 8 second exposures... I've never considered taking anything less than 30 seconds on M31. Why 8?

The 8 second exposures @ iso 1600 f4 were giving me detail nearer the core of M31 so it was easy to shoot them and add them in

The logic behind using 8 was two stops between exposures so 8, 32, 120 seconds

[Photosbykev]

Last version until I get some more data. Thank you for pushing me in the right direction

[Tim]

Looks good Kev

Take care to keep the top of the curve flat when stretching, and it will help preserve your star colours. Also a big help is the star colour action in Noels Tools, or Peter Shah uploaded a similar tool somewhere here in the past. Might be worth asking him.

Basically what the tool does is grab the colour of the star from the edge of the disc, and then apply it to the body of the star, so that you get those nice golds and blues.

At any rate, nice going, you just need more data now to help smooth it out, no substitute for that im afraid