separating color channels

Questions and answers about processing in StarTools and how to accomplish certain tasks.
Post Reply
BABOafrica
Posts: 2
Joined: Mon Mar 14, 2016 4:36 am
Location: Nairobi, Kenya

separating color channels

Post by BABOafrica »

I have StarTools 1.4 and I have been using it for about four years.

I tried an experiment with my Fuji X-A1 DSLR, mainly to deal with light pollution. I took a series of 120s light frames with an Ha. Then I took a series of 120s light frames with an O3 filter. I processed both sets together as two "groups" using DSS and got a single *fits file output. I used StarTools to process as if the *.fits was just the normal output for a DSLR camera.

I was imaging IC 1805 (heart nebula). Blurry but at least infinitely better than light frames that get washed out with LP after 20 seconds.

Somebody suggested (on another website) that I try to separate the RED from the RGGB data and process it separately and then do the same for the BLUE.

Can StarTools do that ... basically creating an image file that has only pixels corresponding to the "R" in the RGGB matrix ???

BABO
User avatar
admin
Site Admin
Posts: 2330
Joined: Thu Dec 02, 2010 10:51 pm
Location: Melbourne
Contact:

Re: separating color channels

Post by admin »

Hi BABO,

Firstly, in all versions of StarTools, you can separate channels easily by just loading any "full color" stack into into the channel you need in the LRGB (<=StarTools 1.4) or Compose (>=StarTools 1.5) modules. StarTools will automatically generate a mono image with just the information from the channel you loaded. You can then save this image, reimport it, create a luminance frame out of it, etc.

From 1.5, the new Compose module lets you process color and luminance separately yet simultaneously, while also offering a super easy way to create a synthetic luminance frame from whichever channel data you have.

For that classic HOO bi-color look, in your case, in 1.5, you'd want to load the Ha stack into red, load O into green and also in blue. Set the for "Luminance, Color", select "L + Synthetic L from RGB, RGB". (see also here). Now set the exposure times for Ha (red) and O (green or blue, but not both; set the other channel to 0). StarTools will create a synthetic luminance frame dataset + a chrominance dataset which you can now process as normal.

From 1.6, the Color module has a number of popular color renderings built-in that create popular blends and color schemes. Start off with the "Duoband" preset and try some different Matrices.

Essentially, you're in luck, as it is now easier to do what you need (and more!) in StarTools than in any other software. :)
BABOafrica wrote:I have StarTools 1.4 and I have been using it for about four years.

I tried an experiment with my Fuji X-A1 DSLR, mainly to deal with light pollution. I took a series of 120s light frames with an Ha. Then I took a series of 120s light frames with an O3 filter. I processed both sets together as two "groups" using DSS and got a single *fits file output. I used StarTools to process as if the *.fits was just the normal output for a DSLR camera.

I was imaging IC 1805 (heart nebula). Blurry but at least infinitely better than light frames that get washed out with LP after 20 seconds.

Somebody suggested (on another website) that I try to separate the RED from the RGGB data and process it separately and then do the same for the BLUE.

Can StarTools do that ... basically creating an image file that has only pixels corresponding to the "R" in the RGGB matrix ???

BABO
Ivo Jager
StarTools creator and astronomy enthusiast
BABOafrica
Posts: 2
Joined: Mon Mar 14, 2016 4:36 am
Location: Nairobi, Kenya

Re: separating color channels

Post by BABOafrica »

Many thanks for explaining all that. Let me try it and see how it works with ST 1.4

Then I'll consider upgrading to 1.5 one of these days.

BABO



admin wrote:Hi BABO,

Firstly, in all versions of StarTools, you can separate channels easily by just loading any "full color" stack into into the channel you need in the LRGB (<=StarTools 1.4) or Compose (>=StarTools 1.5) modules. StarTools will automatically generate a mono image with just the information from the channel you loaded. You can then save this image, reimport it, create a luminance frame out of it, etc.

From 1.5, the new Compose module lets you process color and luminance separately yet simultaneously, while also offering a super easy way to create a synthetic luminance frame from whichever channel data you have.

For that classic HOO bi-color look, in your case, in 1.5, you'd want to load the Ha stack into red, load O into green and also in blue. Set the for "Luminance, Color", select "L + Synthetic L from RGB, RGB". (see also here). Now set the exposure times for Ha (red) and O (green or blue, but not both; set the other channel to 0). StarTools will create a synthetic luminance frame dataset + a chrominance dataset which you can now process as normal.

From 1.6, the Color module has a number of popular color renderings built-in that create popular blends and color schemes. Start off with the "Duoband" preset and try some different Matrices.

Essentially, you're in luck, as it is now easier to do what you need (and more!) in StarTools than in any other software. :)
BABOafrica wrote:I have StarTools 1.4 and I have been using it for about four years.

I tried an experiment with my Fuji X-A1 DSLR, mainly to deal with light pollution. I took a series of 120s light frames with an Ha. Then I took a series of 120s light frames with an O3 filter. I processed both sets together as two "groups" using DSS and got a single *fits file output. I used StarTools to process as if the *.fits was just the normal output for a DSLR camera.

I was imaging IC 1805 (heart nebula). Blurry but at least infinitely better than light frames that get washed out with LP after 20 seconds.

Somebody suggested (on another website) that I try to separate the RED from the RGGB data and process it separately and then do the same for the BLUE.

Can StarTools do that ... basically creating an image file that has only pixels corresponding to the "R" in the RGGB matrix ???

BABO
Post Reply