OIII Causing Graininess

[FuriousRabbit]

I have begun doing some bi-color narrowband imaging and have some basic questions as I'm still very new to this.

First my Ha data looks great once it's stacked and aligned, however after doing the same process with my OIII data there is a very noticable 'grainy' look to the OIII stack. All of my OIII subs suffer from the same level of graininess and I don't know if it's something I should be addressing before stacking the subs.

Second. Once I'm done stacking my Ha and OIII data I'm using pixel math to create a synthetic green, which works beautifully. However I'm confused about the process before I actually generate the new image using pixel math. Should I be using DBE and noise reduction techniques on both the Ha and OIII stack before I create the new image?

I'm currently creating the Ha stack, then the OIII stack, and I begin using DBE, etc on the resulting image generated from pixel math.

[jerryyyyy]

I am not the expert around here, but am in the same boat stacking a bunch of images of the Cocoon.  A couple issues I have noted at the basic level:

1.  Many nebula are mostly H-alpha so you do not get much O-III signal (and just noise).  What is the target?
2.  Type of filter probably makes a  difference.  I regret having the Baader and probably should have Astrodon, but I knew not what I was doing at the time of purchase.
3.  How many subs are you getting?  The stacking routines will clean up some of the mess through dithering when they can use the better math with a lot of subs.  Look at the documentation of the Preprocessing Script and see the cautions the newer versions give when you choose the wrong settings for your number of subs.
4.  I could also use a better telescope

[FuriousRabbit]

I am not the expert around here, but am in the same boat stacking a bunch of images of the Cocoon.  A couple issues I have noted at the basic level:

1.  Many nebula are mostly H-alpha so you do not get much O-III signal (and just noise).  What is the target?
2.  Type of filter probably makes a  difference.  I regret having the Baader and probably should have Astrodon, but I knew not what I was doing at the time of purchase.
3.  How many subs are you getting?  The stacking routines will clean up some of the mess through dithering when they can use the better math with a lot of subs.  Look at the documentation of the Preprocessing Script and see the cautions the newer versions give when you choose the wrong settings for your number of subs.
4.  I could also use a better telescope

Interestingly enough we may be in the same boat here.

1. I'm working on NGC7000, specifically the cygnus wall. I can definitely see data in my OIII subs.
2. I'm also using Baader filters. 1.25"
3. I was only able to get 7 subs of OIII data, each sub @ 900".
4. I've long suspected my scope of causing trouble, I just haven't yet bit the bullet and bought a new one. I have the Orion ED80.

[Nocturnal]

How could your ED80 be causing 'trouble' related to graininess in O3 images? Same for your filters. Why would you even suspect your Baaders as a possible culprit? As long as the passband is in the right spot (ie. they let O3 pass through the filter) you will get most of the O3 there is to get, give take a few %. If the filter is too wide it will let more non-O3 through which will cause your O3 to be washed out but it will not appear grainier.

Just trying to prevent you from trying to fix things by replacing items that are perfectly serviceable

Most likely the problem is that there is so much more Ha than O3 in your target area. If you try to blow up the O3 to comparable levels during post processing you end up amplifying the noise. The only way to counter that is to take lots more exposures. If O3 is 10% as intense as Ha and you want the same SNR you'll need 10x the exposure time.

[DMouse]

Hi, I agree with Sander, its not the scope. I read somewhere that for a good narrowband image you need a ratio of 1:2:3, of Ha, OIII, SII. So for every say 5min Ha sub you need one 10min OIII and one 15min SII to get a nice balance of signal to noise ratio, and hence a clean final image. Also about 15 subs for each channel will help reduce the over all noise, as will dithering. I recently tried to get close to this and it does seem to work.

Clear Skies

Paul

[Dimitris Platis]

This is a very well knowm problem...obviously not caused by the scope...
I agree 100% with the previous posts.
Sometimes it is very difficult to maintain the 1:2:3 ratio....but at least you should add on exposures to SII/OIII to keep the noise low.

[Nocturnal]

I recommend not putting to much weight on such ratios (pun intended). They fall into the same category as "1/3 histogram for flats". They'll work in some conditions but they're shortcuts rather than deterministic.

If possible, review the target before imaging and determine the relative intensities of different NB channels. It may also be possible to use catalogs for this but I haven't tried that.

Juan doesn't want any non-PI software mentioned but there is a program that automatically stacks incoming frames. You can use that to monitor your progress while you capture. Check if there is any signal coming in and what the graininess is like.

[jerryyyyy]

I have been processing and processing today and the H-alpha signal is just a lot stronger in the CoCoon Nebula.  I am collecting a few more subs only on O-III and S-II.  I can definately see the S-II in the Cocoon, but little of the O-III.  I have had less trouble with other targets like M20, which I guess has more O-III...



If I were a real astronomer, I guess I would look all this up before selecting filters etc...  I think most people seem to do the Cocoon in RGB and the images I saw I think tricked me into thinking that the blue was O-III.  Duh.