Firework Galaxy Tips/Suggestions

[waterbourn]

So I feel like I am getting worse at this instead of better.

This was the first time I used my Edge 8 (with 0.7x reducer) to image with my dslr. I am also in the suburbs so I know light pollution is awful. LED street lamps are around near me too. Bhere is what I just can't figure out. I stacked my pictures 2 times. Both times I used DSS and the same settings except for one setting. The first time was with the "Use Color Balance" box NOT checked in the RAW settings within DSS and the second time I did it with it checked. When it is not checked then everything turns red when I autodev and even after I wipe I just can't get anything out of the data. When I use the whitebalance version from DSS then I at least see the galaxy when I autodev. So I am just confused about this. I always see do not white balance, but where is the red coming from (light pollution?). i can't seem do anything with the data when it is like that. I do not understand what the difference is and what I should really use.

Secondly, even when I use the version that is not all red the data has a lot of issues. i see dust motes, gradients, etc even though I did use darks, bias, and flats. The only thing I could figure out to try and salvage the data was to mask the galaxy and use the isolate feature to get rid of the noise everywhere else. It kinda worked, but are there other ways?

Here are the files if anyone wants to see some awful data. Is the main option to get out of the light pollution to really improve the data?

DSS white balance option NOT used: https://drive.google.com/file/d/0B0WZUC ... sp=sharing

DSS white balance version: https://drive.google.com/file/d/0B0WZUC ... sp=sharing

Here is what the DSS version looks like with no white balance box checked and autodev is used

Autodev.JPG (108.13 KiB) Viewed 5798 times

Here is when white balance box is checked and autodev is used

Autodev WB.JPG (315.21 KiB) Viewed 5798 times

This is the best I could do using the DSS white balance option and a heavy use of the isolate feature. after the first autodev an wipe, I had to manually dev as the second autodev was too aggressive at least for me to do anything with.

firework v1.jpg (165.03 KiB) Viewed 5798 times

[Cheman]

Hi Waterbourn
After looking at your files, I think you did an outstanding job with the whitebalanced image. I could not do as well, but I probably did not spend as much time on it as you. As far as the red, I turned up the aggressiveness setting in wipe from 75 to something in the low 80's and the red went away. Now for your data. I dont know why its so noisy, hows many subs did you stack? I'm now typically using 40 or 50, the more the better, but there is a point of diminishing return. I use 75 to 100 flats, and the same bias. I dont do darks with my camera, but I do use hot pixel removal in DSS. Looks like you need to clean your optical train to get rid of the dust motes and such. Flats will take care of it, but I feel you should make it as easy on yourself as you can. It appears you had a lot of image shift between frames, did you take the pics over multiple nites? try to frame close to the same each time. A program that can really help with that is astrotortilla. Its a bit of a pain to set up, but once you do, it makes frameing from one nite to the next a lot easier. I am lucky in that I have a rolloff roof observatory, and that makes thinks A LOT easier. I have a great deal of admiration for those who set up and tear down each nite. Again, I think you did a great job with the challenging image.
Che

[waterbourn]

Cheman,

Thanks for your feedback. This was 42 subs at 120 sec.

I do need to clean my optical train. Those dust motes were everywhere. What do you use? Air? Camel hair brush? I was only using 30 flats and they were fro a previous session.

All these were taken the same night within a span of a couple hours. I seem to always struggle with getting image shift. I don't know if it is because phd2 gets a little off when it acts up or what? I will look into that program. Thanks.

[Cheman]

Hi Waterbourn
I dont clean very often as since I'm permanently set up, things stay attached and are less exposed to things. It depends on how dirty and what it is. One good thing for telescope optics is Dr Clays method found here http://www.arksky.org/asoclean.htm. I have also successfully used collodion. Collodion is used in the cleaning of optics such as telescope mirrors. The collodion is applied to the surface of the optic, usually in two or more layers. Sometimes a piece of thin cloth is applied between the layers, to hold the collodion together for easy removal. After the collodion dries and forms a solid sheet covering the optic, it is carefully peeled away taking contamination with it. There are acouple types of collodion, make sure you use the right one. Careful it is flammable! Many cleaning methods and products can be found via google. What ever you use, do your homework, optics and coatings are delicate and expensive. Collodion works similarily to First Contact cleaning solution, http://www.photoniccleaning.com/ although, I dont know if they are the same thing.
If you flats were taken from a previous session, and the configuration( including camera rotation) changed at all, those flats will not work.

Che

Edit: PS. as far as cleaning your camera sensor, there is much info on the internet for that too.

[admin]

Hi,

White balancing is a very misunderstood subject unfortunately and - counter intuitively - you're actually better off with non-whitebalanced data.
White balancing by the camera (not an issue with RAW) or by DSS (which uses dcraw) clips the data and makes noise worse.

The red bias would normally not be an issue - Wipe will have it for breakfast - were it not for the many dust donuts. Flat frames are an absolute must and will most likely take care of the problem entirely (though a sensor clean can help a little as well).
Wipe hates 'dark anomalies' (e.g. anomalous data that suppresses the real background). In its presence, Wipe will back off locally and cause halos around those dark objects, as it was designed to never *ever* clip your data. To get Wipe to ignore the dark donuts you can mask them out using the mask editor. If you can't really see what you're doing, just let Wipe do its thing (and mess up), just as long as you can see where they are. Then you launch the Mask editor, create your mask, save it, click Undo and launch Wipe again. Launch the mask editor and load the mask that you just created (and saved). Now Wipe should do a much better job;

Screenshot-Untitled Window.jpg (183.61 KiB) Viewed 5758 times

Now that we got rid of the bias, process as normal (though the dust donuts will cause trouble with ST's various algorithms as they will try to preserve them and not clip them).
That said, doing a quick Develop, Life Isolate preset and then jumping straight to the Color module will show that StarTools will achieve a great (better) color balance.
I found it a little on the green side (as shown by the MaxRGB mode), so I reduced green a little.
Most galaxies (but not all) have a yellowish core (older stars) and a bluer outer region (younger stars). You'll typically find (purple) HII regions in the arms. And you'll indeed end up pretty close to what you came up with already.

First thing you need to do though is take those flats!