Hi guys,
The other night I hooked up a Nikon D610 through my ED80 and took 10 30 second exposures of the Orion Nebula at ISO800. I also took 10 darks, 10 flats and 10 bias/offset shots.
The individual light frames show the usual blue and reds that you'd expect from M42, however when I stack it all in DSS and adjust midtones to bring out the detail I get an image with a distinctly greenish tint.
At first I thought it was my flats or darks playing up so I tried stacking without them but I get exactly the same tinting. Cutting G and boosting R and B channels just ends up with a purple sky and the centre of the nebula still has a slight greenish tint.
Or should I just give up on trying to get a decent image out of DSS and just stack and do all the colour processing in Photoshop?
With DSS be carefull as to what you apply to the image, minimal is best, don't let it manage color, just stack don't save any changes it makes, let photoshop or other fix your color problem, you can use whichever method you prefer for stacking be it median, average whatever.
Take you exported Tiff or highest bit image you can import into photoshop.
Duplicate the image, and on the duplicate stretch it to find where least detail exists..... Ie blank sky, no nebula no stars etc.
You can use the eye dropper tool to see where your color balance lies.
Using the original image
Go to Tab
IMAGE ... LEVELS.... Down the bottom right select the middle eye dropper, left click in the blankest bit of sky you found and it should balance the color for you. Just make sure you don't clip the histogram.
Quick,n, dirty but it usually works.
I'm sure others will chip in with ideas for you.
if you do too much with DSS, like go overboard with hot pixel removal and cosmetic fixes, it does all sorts of horrors.
Alternatively, you could let your Nikon program decode the raw files into TIFFS first then stack, worth a try, it's been a long time since I used a dslr, but sometimes it used to have problems with RAW formats.
Hi Martin, You are doing everything out of order, taking darks, flats, etc
when you don't understand how to do the basics in DSS. Forget darks etc for now, and enable in camera noise reduction. You're not wasting much imaging time because your subs are short.
Most DSS output images look like yours, washed out and nearly monochrome. You have to make adjustments to arrive at the colours that
you want. Tick the box under the three sliders so they all move together, and move them to the left to get a sky colour that is fairly dark, but not black. Click the luminance tab. Leave the three sliders set to zero alone.
Adjust the highlights slider to quieten down the washed out highlights in the image.
Adjust the one above it [33.3] to show more or less nebulosity.
Adjust the top one [black] to darken the sky if necessary.
Click the saturation tab and provisionally set it to about 16-18.
Back to histogram, untick box and adjust RGB sliders individually to
get the colour how you want it. You should end up with the three colours merged into a single tall cone, and the dotted coloured lines should also nearly merge into one. Miniscule adjustments make big
changes. Finally, readjust saturation to suit you. Export image to other software to make any other adjustments such as contrast, sharpening,
etc. Colour sliders in DSS are difficult to move in very small steps.
This is how I do it, others may have different routines. Hope it helped.
raymo
Hi Martin, You are doing everything out of order, taking darks, flats, etc
when you don't understand how to do the basics in DSS. Forget darks etc for now, and enable in camera noise reduction. You're not wasting much imaging time because your subs are short.
Most DSS output images look like yours, washed out and nearly monochrome. You have to make adjustments to arrive at the colours that
you want. Tick the box under the three sliders so they all move together, and move them to the left to get a sky colour that is fairly dark, but not black. Click the luminance tab. Leave the three sliders set to zero alone.
Adjust the highlights slider to quieten down the washed out highlights in the image.
Adjust the one above it [33.3] to show more or less nebulosity.
Adjust the top one [black] to darken the sky if necessary.
Click the saturation tab and provisionally set it to about 16-18.
Back to histogram, untick box and adjust RGB sliders individually to
get the colour how you want it. You should end up with the three colours merged into a single tall cone, and the dotted coloured lines should also nearly merge into one. Miniscule adjustments make big
changes. Finally, readjust saturation to suit you. Export image to other software to make any other adjustments such as contrast, sharpening,
etc. Colour sliders in DSS are difficult to move in very small steps.
This is how I do it, others may have different routines. Hope it helped.
raymo
Raymo why say he doesn't know what he's doing taking bias, dark and flat images? This is the best way to process images (though really you probably need more of each than 10 to avoid noise in these images creeping through in the final output).
DSS gets a bad rap because most people don't know how to use it. Part of the reason colours come out muted with DSS is that it's compressing the dynamic range massively to try and show everything. Colour in most astro images is compared to the luminance signal, so they end up looking washed out unless you boost the saturation a lot. As other's have noted, they controls for adjusting the colour balance and saturation are pretty clumsy in DSS so it's best to this in other programs.
Martin:
The only thing I'd change from what you've got there are:
1) adjust the top end of the luminance so that you don't have a slight downward curve at the right-hand end. I'm not sure if this is just an artefact of the way the curve is displayed but if it's real it'll lead to weird looking stars with darker cores.
2) bring up the mid-point of the luminance so that you've got a broader inflection point going from dark to bright. This is where to centre your RGB histograms.
3) move the RGB histograms to the centre of the new inflection point. This means that the most pixels are now your mid-tones. This should result in a pleasing image with a broad range of shadows and highlights.
4) Make sure each of the R, G and B channels has its centre spike overlapping. This works because of the fact the most images have their mid-range luminance in each channel - grey tones. It's the variance (above or below this) that gives things their colour tone. If you're not happy with the look this gives then you can alter the image later in another tool, but this will be at least neutral.
5) increase the saturation to 15% if you've removed all saturation boost from the camera. 10% is a safe boost if your camera already gives colourful images.
Now save and open in another program like Photoshop to make your final adjustments.
Cam, I didn't say that he didn't know what he was doing taking darks, flats etc. I imagine that he took them correctly, and knew why he was taking them. My point was that he was missing out a step on the ladder by not knowing the basics of using DSS on any image, let alone ones that he has gone to the bother of supplying darks and flats for.
To my way of thinking, all that effort was wasted if the benefits that said
darks, flats, etc confer can't be utilised. A bit like paying $1000 to hurl a
race car round a track, and sitting on 100kph because you have little or
no idea how to handle all that power. I hope this doesn't offend anyone,
it's just my opinion.
raymo
I haven't had any real joy out of DSS yet either. Some guys have made detailed comments on how to use it on my Orion Nebula thread. I have trouble working out the sliders too but I'm going to persevere with it and see if I can master it.
Where abouts in the Pilbara are you Martin?
Graeme, I am an old, almost computer illiterate, man, who struggles with
the digital age. I am at the bottom of the pond as far as astroimaging
ability is concerned. I don't use darks, flats, biases, or RAW subs, but I
have made the best of a bad job by getting the most that I can out of DSS. All my processing is done in DSS, and downsizing done in PS.
All images I have posted here have been produced that way, except for
the single frame ones, which have been tweaked in Canon's Zoom BrowserX.
My point is that DSS is not that hard to master; if I can do it, anybody can.
Sorry for the hijack.
