Rho Oph and Blue HH

[jjjnettie]

33x5 minutes
Canon 550D @ ISO 800
Tamron 90mm @ F5.6
Flats, Bias and 50 darks
DSS and CS3

the potential is there for this camera, I know it is.....
I'll shoot this one again tonight at ISO 400 and see how it goes.

BTW, I used Jpg capture not raw, as DSS won't recognise the raw from the camera.

[Logieberra]

Nice one! This is my new fav part of the sky! Can't wait to do one of my own!

[multiweb]

Sure is - there's always a way around things. Great shot. More data will make it one for the cool wall. You've got good colors and excellent focus. :thumbsup:

[Hagar]

Very nice JJ but 50 darks. Looks good though. Nice resolution and detail. Colour looks real good as well. You have captured a lot of the free dust in the area. You have to be happy with this result.
I still think 50 darks is overkill.

[jjjnettie]

Doug, I've spent all this morning processing this one. multiple times.
The first attempt was using the standard 1:1 light and dark. But the colour blotch/streaky noise was impossible to work around.
But I'd kept the camera shooting darks, wanting to experiment with using more.

Now strangely, when flicking through the images on the camera, you look at one star, it stays put in each frame, but the colour blotch near it, will move diagonally to the right a little bit each time. This is where the streaking is coming from.
I'll keep playing with the parameters in DSS and see what I can work out.

[multiweb]

Have you tried dithering? Offsetting your subs will greatly help when stacking as far as data rejection goes.

[ozstronomer]

JJJ that's a great shot, loads of detail, you should be happy with that one :thumbsup:

[adman]

Thats fantastic. You have really taken it to the next level recently. Good on you.

Adam

[h0ughy]

this is a target for astrofest this year with the 70mm, love what you got JJJ.

in levels, how about trying to adjust the indivitual channels of r g and B in photoshop - thats all i did - raise the black point only to the bottom of the curve. Love the image JJJ!!!!!!:thumbsup: go and get more data:thumbsup:

my only real problem is the banding that is in the image - this has to have come from the camera

[jjjnettie]

This is a 100% crop of the finished image. Make of it what you will.

[h0ughy]

that is a lot of chromanance noise - maybe the sensor is struggling - or the software in the camera interpreting the data its capturing. My pentax was utterly horrible, then pentax released a firmware update that helped to fix the noise - the camera was still horrible for Astro work though. I not sure abou this JJJ with the 550D? its not modded so the sensor isnt getting a full "spectrum" there maybe thats why its struggling? i know if you have light pollution about you get that type of noise badly. You can go for a long time getting data but eventually you get an extinction in the "town/city" light. getting rid of the noise is almost impossible because if the poor signal/data. What are your darks like noise wise?

[jjjnettie]

This is a 100% crop.
I've sent actual raw files to Terry Lovejoy to analyse and we're both just shaking our heads over it.
It's not an astro imaging camera, that's all.

[h0ughy]

ok JJJ - my view and i will say my very much unqualified opinion...... when you stretch the dark you get the image shown - which is very very similar to the noise in the image:shrug: - which does seem to be a lot, but then i am spoilt with the 40D

[jjjnettie]

Thanks Steve.
I'm my own worst critic, I know. I get roused on all the time about it.
I am proud of what I've achieved so far.

[jjjnettie]

You are spoiled indeed. I hated having to hand back Ponders 20D. It's smooth as silk in comparison.
I think it's time to try some very left of field experimentation. Drastic noise calls for drastic measures. If it works out, I'll post it.:D If not, no one will ever know my shame. :rofl:

[Octane]

Noise, or not, that looks awesome to me.

Jeanette, what combine algorithm are you using for both your darks and your lights?

H

[Ross G]

A very good shot.

Great colours.


Ross.

[jjjnettie]

See, now this is where I feel my lack.
I have no idea about the technical stuff. It's just gobbledegook to me. :lol:

I use a Median Combine. Is that what you mean?

[Octane]

Yep.

Try this, median combine for your darks. And, that is awesome that you took so many darks. You have an SNR of over 7 -- brilliant.

For your lights, try sigma kappa (sigma clip, in some other software). As I recall, DSS is like a pretty front-end for IRIS, you should have sigma kappa combine option. Give that a whirl.

H

[gregbradley]

I like this image a lot JJ.

With darks and a DSLR you have a major disadvantage in that the amount of dark noise depends on the temperature at the time.

If you take your darks at 15C but when you imaged it was 25C for example they won't match up very well at all.

Your image was probably taken at around 10C? What was the temp when you did your darks?

With CCD chips, there is a data sheet on each one from Kodak. Very typically dark noise doubles with every 6C increase in temperature.

One way around this is to use what is called adaptive darks. Where the software measures your image and adapts the darks you collected as a kind of mathematical approximation.

They usually work fairly well. I occassionally use them with my astro CCD but prefer not to as they are really a bandaid approach.

But if you didn't use them then that is one tool you can use.

I usually use sigma reject combine. Its a fancy term but it these names mean are different ways of working out an average for each pixel for all the darks you took. Sigma reject means if you have say 10 darks with an average of one particular pixel of say 15 and then 3 have 90, the 3 x90 values are excluded from the averaging process as they are not usual and are considered random.

Sometimes the sensor is hit by cosmic rays for example that give a worm like wriggle on the sensor that can show up in your darks but they are a random occurrence not usual standard repeatable dark noise.

Noel Carboni has an action that is supposed to get rid of horizontal banding. I wonder if it would work here.

The banding is fixed pattern noise - ie a fixed pattern in the chip itself from the manufacturing process. I imagine that would fade to far in the background with longer subexposures as well as with lower temperatures.

The modification of course increases the sensitivity of the camera and that fixed pattern noise would be pushed far into the background like a carefully held secret!

Greg.

[DavidU]

A glorious image jjj. You are doing well to tame the 550D.:thumbsup:

[pmrid]

Jeanette, now you've got me doing it. Seeing your Antares region and Crux shots has persuaded me to give the 1000D one last chance. I just set it up on top of the C14 on the Titan mount so it ought to be as good as I can make it. I'm just running an hour or so on Crux until it gets a bit darker and Antares has risen a bit higher and I'll do 2 or 3 hours on it and post the comparisons. I plan doing ISO400 at F5.6 and 300 sec subs. I'll do darks and bias shots when I'm done with that lot.
Peter

[jjjnettie]

I keep a thermometer next to me to keep an eye on the temp during the night. I started imaging at approx 6C, the session ended at 1.5C. Darks were taken at the beginning and the end of the run.
I have the Carboni actions. The Colour Blotch remover and the Horizontal Banding action were the first things I tried back when I first got the camera.
Like you said though. All this is just a bandaid solution.
I should be using a dedicated astro ccd. Because I'm not going to be happy until I start producing the smooth, sharp images that you guys are putting out. :)

[h0ughy]

thats garbage JJJ - you produce a lot of good images. I think the image has a wonderful framing and great colours within (no noise ;) ) if we can only fix the problem........and bandaid solutions are good when they make it all worthwhile

[jjjnettie]

Please don't take me wrong, I'm very proud of what I've produced so far.
And this struggle with the 550d can only improve my skills even more.
Thank you to everyone for your kind words and input. It's very much appreciated. :)

[midnight]

Jeanette,

Your framing is very nice and something that I just don't really think about until afterwards.

I would certainly be very happy with those images. Well done!!

We must have been targetting the same areas last night as I had enough of Antares due to light pollution and thought bugger it I'll drop by good old Crux!

Well done and thanks for sharing. and good luck with the 550D - just keep prodding it and talking to it nicely (then maybe not!!) :poke:

Darrin...

[irwjager]

Still a pretty respectable image JJJ - Luminance wise I'd be pretty happy with that myself, save for some streaks here and there.

Have you tried getting rid of the color noise by applying a big Gaussian blur to the color information (use layers for this)? It usually does wonders for chromatic noise.

I'm not familiar with the 550D, but I know that on some other cameras shooting in JPEG doesn't do you any favors when it comes to astrophotography. On these cameras shooting in JPEG will put it through all sorts of on-board image processing (including hot and dead pixel removal as determined in the factory) filtering and sharpening to make an image look 'pretty'. Add to that that these image processing steps need to be quick & dirty (because people expect to see the finished result on the screen virtually instantly) and things become sub optimal very quickly (not to mention compression artifacts and the conversion of your 10 or 12-bit source to 8-bit for JPEG).

I also know that, on these cameras, if you take dark frames in JPEG, chances are that any hot and dead pixels are already (sub optimally) removed from your darks. And if they are still present, JPEG is quite bad at encoding sharp spikes, resulting in artifacts (and corruption) around a hot or dead pixel.

Long exposures in JPEG mode in some cameras also do their own dark frame subtraction, so subtracting dark frames *again* will cause a lot of grief, probably similar to what you're experiencing now. You can tell if this is the case if after an exposure of a specific duration the camera takes an equal amount of time to 'process' the image. In reality this 'processing' is the camera shooting a dark frame.

It all depends on whether your camera allows you to turn all this off for JPEG processing (my hunch is that it doesn't). If it doesn't then shooting in JPEG is usually a bad idea!

[jjjnettie]

I shot in JPEG because DSS won't process this particular type of RAW properly. It seems to be processing properly right up to the very end when it spits out an image that is only a 1/4 slice of the full picture.
I can shoot raw, then convert to tiff in DPP, but the resulting TIFF's end up being over 100mb each.
DSS , when processing that much data, then tends to run out of memory about 2 minutes before it finishes stacking. dammit
In the past few months I've conducted many experiments with different settings , comparison imaging, software updates, firmware updates, etc
etc etc To no avail.
So for the time being I'll continue using jpgs. ;)