Hello all...
Recently got an Astronomik OIII filter to combine with my Ha filter in order to try my hand at Synthetic RGB work.
When I've taken Ha subs to combine with RGB files for "HaRGB", I've split the Ha into its component red, green, blue channels...discarded G and B and used the Red channel as my Ha file - seems to work better than just converting the entire image to greyscale.
Question: Do I do something similar with the OIII sub? i.e use the green or blue channel?? Then use that single channel to work with to produce the synthetic RGB image?
You'll find that when imaging with a hydrogen alpha filter, that the red channel will be busiest, as you've mentioned. Similarly, the oxygen III filter will show more detail in one of the other two channels.
You'll find that when imaging with a hydrogen alpha filter, that the red channel will be busiest, as you've mentioned. Similarly, the oxygen III filter will show more detail in one of the other two channels.
Pick and choose and hack as appropriate.
Regards,
Humayun
Yeah right H - I'm ahead of myself!!
Thanks for the reply, and you're probably correct...experiment!
Cheers
Doug
Interesting. I had assumed you'd use the whole Ha data and put it all in one channel, all OIII in another channel etc.
Using the above method, what do you do if you want to add SII? It's primarily red also. If you were to only use the blue or green from it, be very faint? Or do you take from it's red channel only, and put that in the synthetic green channel or whatever? As I said, I had assumed you'd put it all in one channel.
Interesting. I had assumed you'd use the whole Ha data and put it all in one channel, all OIII in another channel etc.
Using the above method, what do you do if you want to add SII? It's primarily red also. If you were to only use the blue or green from it, be very faint? Or do you take from it's red channel only, and put that in the synthetic green channel or whatever? As I said, I had assumed you'd put it all in one channel.
Re Ha - all the useable data is on the red channel, so I split the Ha into RGB components and use the red as my Ha for subsequent work. I've found a straight change of the original RGB file into grayscale loses all the contrast.
With this in mind, I wondered if most of the data in an OIII image resides in either the green or blue channels - which one to use!!??
Inspection of the images suggests there's an equal distribution of data over the G and B.
May try and develop my own method
Eddie takes stunning NarrowBand images with, I think, a Mono camera. From the sound of it, you guys might be using a colour camera which may add unwanted complexity??. A Mono camera gives all the images in grayscale and you assign whatever colour you like: HOS = Ha-Red, OIII-Green, SII-Blue. SHO ("Hubble Colours") = SII-Red, Ha-Green, OIII-Blue. There are other combinations as well.
A good intro - http://starizona.com/acb/ccd/advimnarrow.aspx
Eddie takes stunning NarrowBand images with, I think, a Mono camera. From the sound of it, you guys might be using a colour camera which may add unwanted complexity??. A Mono camera gives all the images in grayscale and you assign whatever colour you like: HOS = Ha-Red, OIII-Green, SII-Blue. SHO ("Hubble Colours") = SII-Red, Ha-Green, OIII-Blue. There are other combinations as well.
Charles
Hi Charles - correct I'm using a modified Canon 40D DSLR. The narrowband filters are the Astronomik EOS clip in variety which sit over the sensor in the camera body. I've achieved passable results with the 12nm Ha filter so far... http://s327.photobucket.com/albums/k...urrent=20b.jpg
...and inspired by Eddie's work, I purchased the equivalent OIII filter to try my hand at widefield synthetic RGB - so as you see I don't shy from unwanted complexity!!
Detail on the 10min images I took last night using the OIII filter really is very faint (as expected) and I will have to push exposure times way out of the HEQ5 comfort zone I reckon.
But as you've said, a mono camera is the tool of choice for this stuff and this I don't have. So my question still remains - in the "RGB" produced by using the OIII filter, where does the useable data reside...in the blue or green channels? Or, does one just convert the RGB to greyscale and use that?
I'm starting to confuse myself I fear!!
Doug
Not that I've got experience, but I would suggest setting the camera to mono and see how that goes. The Starizona docs show the Ha in the red, the SIII in the deep red and the OIII as turqoise. They also mention you could try HOO - Ha=red, OIII-green and OIII-blue. I just got my camera and filters last month but haven't used them yet.
...but I would suggest setting the camera to mono and see how that goes.
Charles
Hmmmm interesting Charles, hadn't considered this. I suspect though that this would result in a loss of data - electronic pixel wells not being filled or some such thing!! Hopefully someone else can flesh out the science of my uninformed statement!!!
Doug
All the HA is the red channel. So discard the G and B. For O3 you will find that the O3 is recorded in both the G and B channels almost equally. So in this case dicard the red channel. This is why O3 is a lovely cyan colour. Cyan is the absence of red in an additive colour system.
The Bayer filters in a colour DSLR stop any useful light getting to the other channels. If you do use them you are basically only adding noise.
For something like the Helix Neb which is nearly all HA and O3 you will get what looks like a 'natural' colour image if you assign colours thus.
HA red to RED Channel
O3 green to GREEN Channel
O3 blue to BLUE Channel
All the HA is the red channel. So discard the G and B. For O3 you will find that the O3 is recorded in both the G and B channels almost equally. So in this case dicard the red channel.
Bert
Thanks Bert - confirmed!!
I remember those threads too - interesting idea, well executed.
Cheers
Doug