OSC data in Compose module

[jhart]

I understand that clicking Open and loading an OSC stacked data set with the Linear, Bayer matrix, not white balanced option, will load it automatically in the Compose module (and its button turns green). Do I then need to open the Compose module and select one of the OSC options in the Luminance/Color parameter for proper processing?

[firebrand18]

It's actually one or the other way to open an image, not both; If you do the "Open and load a OSC stacked data set with the Linear, Bayer matrix, not white balanced" option, you then process from there (no Compose). When you get to the Color module, the Matrix dropdown will show a bunch of DSLR camera models you can emulate.

If you load through Compose from the start and choose "Linear", when you get to the Color module will show HOO/SHO and other variations you can use to create a bi-color or other composite.

Cheers,
Nick

[jhart]

Hi Nick,

Thanks for the reply. My confusion is that the documentation for the Compose module says that the "L + Synthetic L from R(2xG)B, RGB (Color from OSC/DSLR)" algorithm in that module is or should be used if OSC data is loaded with the Open button and "Linear from OSC/DSLR with Bayer matrix and not white balanced" is selected. Does that mean that Compose is activated even though it was not clicked on to load that data in that way? (It's button at the top turns green I assume meaning it is active in some way, and seems to stay active until Tracking is turned off). Do I need to go back into the Compose module and change or tweak its parameters? (If I click on the Compose button after loading OSC data in that way, that algorithm is not shown as being selected in the Compose Luminance,Color parameter).

I also do not understand how you should use the Color screen (vs the Luminance screen) in the Wipe module to make adjustments when the Color button is enabled after OSC data is opened with that "Linear from OSC/DSLR ...." option and apparently activating Compose in some way.

Jeff

[firebrand18]

Let's first narrow down what data you are looking to process; emission nebulae using narrowband Ha/Hb/OIII filters or broadband targets like galaxies, globs etc. Will have a big say in how to process.

Nick

[jhart]

Hi Nick,

I am trying to image DSO's: galaxies, globs, and smaller emission nebulae (or parts of larger ones). I don't use narrow band with my OSC camera, just an occasional LP filter like the L-Pro.

Jeff

[firebrand18]

Using the L-Pro broadband filter (I use it as well) on galaxies and globs, you can go down two routes:

Normal process for this OSC visual spectrum data is open ST, click load and choose the middle option, Linear from OSC/DSLR with Bayer matrix and not white balanced; you would not bother with Compose at all, and like I mentioned, in the color module, you have DSLR options available.

You can also open ST, click Compose right away (not Open) and as stated in the documentation on ST site "use L + Synthetic L from R(2xG)B, RGB (Color from OSC/DSLR)" to create a synthetic luminance dataset from Luminance, Red, Green and Blue, weighted according to the exposure times provided by the "Total Exposure" sliders. The green channel's contribution is doubled to reflect the originating instrument's Bayer Matrix having twice the amount of green samples. The colour information will consists of simply the red, green and blue datasets as imported. StarTools will be put into "composite" mode, processing luminance and colour separately yet simultaneously. This mode is suitable for OSC and DSLR datasets and is used internally by the "Open" functionality on the home screen when the user chooses the second option "Linear from OSC/DSLR with Bayer matrix and not white balanced."

The Compose option gives more flexibility in certain modules and in color module gives the Composite options, not the DSLR models.

I've processed many galaxies, globs and reflection nebulae with the L-Pro with the first, default option and works great; now getting into narrowband with Ha, OIII filters to create HOO, H(H+O)O, HOO with RGB etc, composites and using the compose module is the way to go in this more advanced mode. I am also learning this as I go along and by no means an expert.

Hope it helps; others may chime in as well to further comment.

Nick

[Mike in Rancho]

Yes, as I think quoted above, I believe center-option-open gets you into "compose mode" exactly the same as if you had gone into composite, loaded your OSC stack into each of the R, G, and B channels, and chosen the L + Synth L from OSC/DSLR (with 2xG) option, with the weighting default equal on all channels.

When you hit composite again, after having already loaded data, it is no different than hitting open again -- meaning ST thinks you are ready to start over at the beginning with new data.

Wipe is, to my limited knowledge of course, just to get to an "even field" across the image, handling any gradients, vignetting, dark/bias issues, and so on, in both L and color. And in color you can also check for various color casts and splotches across the image (which will propagate to later), and can often be handled by DAF and if necessary, aggressiveness.

[firebrand18]

Correct as Mike states. Note the color toggle option in Wipe is the only place you will actually get a hint of the color in your image until you get into the color module itself; I use this feature all the time to switch between luminance and color to see what Wipe is doing before moving on to the final global stretch (Autodev). You will get a different perception of the aggressiveness setting wipe uses when you compare the mono vs color preview; very handy.

There is a reason ST is a big proponent of processing luminance and color separately but in parallel until Color module; you get much better results tweaking small/medium/large detail with contrast, hdr, sharp, decon etc when only looking at luminance; color can make a mess of things and really confuse how to approach your final result. Keeping it towards the end works very well.

Cheers,
Nick

[jhart]

Thanks for all of that helpful information.

When clicking to see the color screen in Wipe, is it useful in adjusting any of the Wipe parameters? Are there any guidelines on how that screen should appear for purposes of further processing and a good final image?

Thanks again,
Jeff

[firebrand18]

Welcome to the wonderful world of endless processing, re-doing, updating, tweaking the same dataset; you think you finally get the image you like, then find something else to adjust and off you go!

The 75% default aggressiveness in Wipe works well in most cases, especially broadband data with the L-Pro; light pollution (Bortle scale) you are imaging from and other variants will affect the dataset, and tweaking more or less the percentage may yield different results.

If everything looks faint and washed out try higher setting but you want to avoid blowing away faint detail if you push too hard; masking out a galaxy for example may help remove gradients but keep galactic detail intact but not always necessary. With L-Pro, 75% works well for me in Bortle 8 location.

What I do after Bin is play back and forth with Crop/Wipe until get a good base, play back and forth with Wipe/Autodev (with ROI tweaks) jumping right to Color to see what it looks like; once satisfied, go back and redo everything including all the other modules for bringing out detail; I go through multiple versions of an image before settling on a final to post.

In Wipe, play with the Correlation Filter as well; makes a big difference to remove smeared noise due to Dithering and other junk; I normally use 0.7 but you can adjust to taste, stopping before you start bloating stars and blurring detail.

Nick