Stars all seem saturated?

[Terry Danks]

Hi All:
I have been trying to figure out why I am losing all colours in virtually all stars . . . not just bright ones.

In my individual subs I measure the ADU value of a given star as 19,000. The same star in the RGB combination measures at 63,000, and appears burned out with no colour. My camera is an STL-11000 and saturates at 65,000 ADU. Needless to say, the bright stars are all totally saturated in all three channels.

[sreilly]

Actually the full well capacity of the STL-11000 is 50,000;

Imaging CCD Kodak Enhanced KAI-11000M
Pixel Array 4008 x 2745 pixels, 36 x 24.7 mm
Total Pixels 11 million
Pixel Size 9 x 9 microns
Full Well Capacity (NABG) 50,000 e-
Dark Current 1.5 e-/pixel/sec @ o degrees C
Antiblooming ABG only

[topboxman]

Hi Terry,

If you are using LRGBCombination tool, are you playing with Transfer Functions sliders (Saturation or Lightness)? If so, leave them at 50%. I assume your RGB images are still linear before RGB combination. Do you do any other processing besides calibration before RGB combination? Also, are you combining Luminance with RGB?

I don't recall stars get brighter after RGB combination in my images.

Peter

[Terry Danks]

Actually the full well capacity of the STL-11000 is 50,000;

I am confused.
When I put the cursor over a bright star in PI, the readout on the bottom bar for an integrated and combined RGB image is "65400," more or less, for each of the R, the G and the B channels.

Doing the same for an obviously saturated star in a single, non-integrated, monochromatic  sub, yields 65,500 as the displayed K value.

I took these to be the ADU values. Not so?  I'd supply a screenshot by can't. The values disappear when I open the snipping tool.

[topboxman]

I think there's confusion. Full well capacity unit is electrons and ADU is a digital unit based on electrons and gain. These two are not the same thing.

Peter

[oldwexi]

I took these to be the ADU values. Not so?  I'd supply a screenshot by can't. The values disappear when I open the snipping tool.

Hi Terry!
Putting the Cursor over the star press the readout button to view the Pixel values, doing this press in addition the Windows "Print" button
this creates a screenshot which you can load afterwards with:
 Ctrl+V

Gerald

[Terry Danks]

Hi Peter:
Thanks for the reply. Truth is, I am having a hard time getting a pleasing RGB image with any colour at all in the stars. I am also getting purple HII nebulosity where I expect reds. I just did a straight RGB combine. And the results are not surprising. The RGB values are comparable to the individual values of the R, G and B integrated subs. Not all near 65,000 as I first reported. But, the stars are still all white . . . no colour at all.

The screenshot (Thanks, Gerald for the instructions!) shows the actual RGB I am working with. Note the stars are . . . just white.
And the nebulosity is purple. 

I am combining the RGB in linear space using only the Channel Combo tool. Using the LRBG tool to use Ha a luminance comes next. I was able to get very pleasing nebulosity doing that . . . but just white stars.

I take it that saturated pixels on the STL will be 65,000 ADU. That corresponds to full well of 50,000e. Is that not correct?

[topboxman]

Hi Terry,

It may be a little early to make a judgement after RGB combination. Your histogram showed unbalanced colors meaning you have more work to do. You will then have to do some clean up like balancing the colors using Dynamic Background Extraction, Background Neutralization and Color Calibration. You may be surprised of the results.

Also, using Ha as Luminance can be difficult to do and can mess up the star colors. This requires quite a bit of skill I have not yet mastered. I blend Ha to red channel using a script done by third party and it's not part of PixInsight. The script can be downloaded from:

http://pixinsight.com/forum/index.php?topic=3446.msg43314#msg43314

This script is based on Harry's "Adding Ha to the Galaxy" video tutorial except that it's much easier and quicker to run the script. I would look at the video tutorial for educational purpose to get the idea of how it works.

I will be happy to a look at your calibrated and stacked Red, Green and Blue images and try to process them if you upload the images (just three files or zipped file containing RGB images) to a server like dropbox.com.

Yes, in your camera you can say that 65K ADU correspond to full well of 50K electrons. The gain is set in such a way the full dynamic range is very close to 16 bits (65535).

Peter

[Terry Danks]

Again, many thanks, Peter.

I have done so many things to this poor image now, I quite forget just what and probably did all kinds of detrimental stuff.
So I went back to the original stacked, calibrated and registered subs and started over doing very little processing. The resulting RGB has pleasing nebulosity . . . but no star colours and, again, all but very faint stars have readouts indicating saturation. The RGB is attached. Again, minimal processing has been done. No DBE, but I did do a background neutralization. I deliberately clipped the individual channels 1%. No cosmetic correction and no noise reduction or HDR. The issue that concerns me has only to do with the stars being nothing but white. Whatever mess I get into trying to blend Ha into the mix can come later. For now I'd settle for an RGB with some star colour.

I can do the dropbox thing, as you so kindly offered, if you think it worth your time.

[topboxman]

Hi Terry,

It will be worth my time to look at your images. I googled for images of Lagoon Nebula and noticed that stars within nebula area are not as colorful as you may have thought. Take a look at this APOD:

http://apod.nasa.gov/apod/ap130817.html

Stars surrounding the nebula are more colorful than inside the nebula. Maybe this DSO may not be a good sample to process to get colorful stars within the nebula.

Please upload your images so I and others would be happy to analyze and process.

My old image of M8 do not have colorful stars within the nebula but there are some colorful stars around the nebula:

http://peternagy.smugmug.com/Telescopes/Nebulae/15823498_3vBdTQ#!i=1306471066&k=mB3LxdB&lb=1&s=A

Peter

[Terry Danks]

Hi Peter. File are uploading to drop box. PM sent.

The APOD version has more colour than my stars. They look like blue SGs in an OB association. All but the really faint stars in my version are simply white. As in "white" white. Not "bluish" white. 

[topboxman]

Hi Terry,

Once the upload to dropbox is finished and successful, then left click on Dropbox icon in taskbar and click on Dropbox.com. It will then display a list of files uploaded and the rest should be self-explanatory and you should be able to get links to the files you uploaded. Dropbox is really a cool program for uploading images to a server.

I have to admit APOD do show some star colors and this was taken from someone living in Southern Hemisphere where Lagoon Nebula can be at near Zenith. From where we live, Lagoon Nebula is pretty low in the sky and maybe that's why we don't get colorful star colors?

Looking forward to process your images. I am kind of bored since I have limited clear skies and I don't mind processing other people's images and plus I need the practice. 

Tonight looks like it's going to be a very clear night and I plan to test my new Windows 8.1 laptop using USB3 ports.

Thanks for giving me Windows 8.1 tips at CN.

Peter

[Terry Danks]

Good luck with the new lap top tonight.
Here's the drop box link.

https://www.dropbox.com/sh/jegp8mdeihzni6c/AAAMMGrjN94Kfrk5AnoRewa-a

[topboxman]

Hi Terry,

Here is my quick process of your RGB images. Stars appear to be colorful surrouding the neubla.

https://imageshack.com/i/np9v20j

I have a feeling that the sub-exposures may have been a little bit too long. Lagoon Nebula is a pretty bright diffuse nebula. Can you tell us the following:

1) Scope
2) Sub-exposure times for each filter and total number of subs for each filter

I think you did a great job capturing Lagoon Nebula.

My process procedure:

1) LRGB combination.
2) Dynamic Crop
3) DBE
4) BN
5) CC
6) MMT to remove noise while image is still linear
7) HT
8: SCNR to remove green
9) HDR
10) LHE
11) CT using Sat

Thanks,
Peter

[Terry Danks]

Hi Peter:

Many thanks for your efforts. You have more star colour than I was able to retain.

My RGB data are as in the attached table.

I have done no CC as I was always at a loss as to what to choose for white reference in a field filled with HII
nebulosity? I have just reviewed Harry's tutorial and will give it a go tomorrow using the star method.

Also, in your workflow I do not know what your final step is. "CT using sat?"