Image for Processing Discussion

[Apollo XX]

Last month I was out in the street playing with my DSLR on the hinge tracker. I took some exposures over toward the Leo area in hopes of randomly catching comet pan-stars K1, but since I didn't actually know where the comet was, it really emphasized the meaning of the word "random". K1 is near Leo now, so maybe I'll go back there. Anyway, I took some exposures in the area of Ursa Major just for kicks. We were discussing processing in Andy's Abell cluster thread, and so I submit these examples for your perusal.

This first image shows the full frame of the exposure. The details are; Nikon D5000, 50mm prime lens, F2.8, ISO800, 60 seconds. The bright double at the bottom center is the Mizar/Alcor duo, the bright star up towards the middle is Alkaid, and the bright double in the upper left is Cor Caroli. My feeling on this frame is that I got as much exposure as I could under the conditions without burning things out.


This next image shows the frame after a trip through Gimp with some tugging on levels and curves. The sky background has a distinct red hue to it. I used the "daylight" white balance setting on the camera to allow the natural colors of the stars saturate the exposure. I've used tungsten in the past, and although the adjusted sky background is a pleasing blue, all the stars are white.


This image shows the original exposure cropped to extract more of the captured detail. That's M51 to the upper left, and M101 to the lower right.


In this final image we have the same cropped section from the adjusted image. The clipping has deteriorated some of the galaxy data, and any more pulling on it just eliminates them. I tried a couple of things like inversion then value invert, and although it resulted in a black sky and very pretty star colors, it completely eliminated any other subtle details.


I know there isn't much to work with here, but any suggestions on how to improve the quality of captures like this would be appreciated.

Mike M.

[AndyG]

Mike,

Could you post the raw NEF files from the camera somewhere like dropbox or google drive? Those jpg files are smaller, but have lost some data due to the lossy jpg compression format, and I think we might be better off working with the raw data .

[Bruce D]

Gee I thought the final image looks pretty good Mike, looks like you got a couple of faint fuzzies too!

[Apollo XX]

Mike,

Could you post the raw NEF files from the camera somewhere like dropbox or google drive? Those jpg files are smaller, but have lost some data due to the lossy jpg compression format, and I think we might be better off working with the raw data .

Andy, I'll e-mail you the file later this afternoon. It's just one 10-ish MB NEF, but I don't have it with me right now.

Gee I thought the final image looks pretty good Mike, looks like you got a couple of faint fuzzies too!

Thanks Bruce! We definitely got quite a few in the whole frame, and the crop section has at least M51 and M101. What I'm hoping to see is if someone with more experience and/or a dedicated astrophoto processing program can process out the reddish skyglow without eliminating the desirable data.

Mike

[Apollo XX]

Mike, how'd you the NEF conversion? Or did you just use the JPEG versions?

The raw NEF file doesn't include "white balance", but the raw outputs from the Bayer RGGB sensor array. White balance is used to convert this raw data to the colorspace of the display format (e.g. JPEG); that is, white balance is the RGGB weights used to compute the final output.

There are two ways to fix your problem: rework from the original file, adjusting the RGGB coefficients. There are several programs that can do this. You get correct coefficients by measuring the RGGB values of a known G2V white star. The Astrondon web site has an video on how to do this. Just beware that the color of stars that saturated may be incorrectly adjusted (depending on the software); you'd have re-saturate the afterwards. (In general, I avoid saturation in my subs, since your into non-linear territory then.)

Or, you can adjust the RGB levels, or curves, in Gimp. I'd first select out the stars, because they'll be saturated and you don't want adjust their color.

P.S. Theoretically, once you've determined a white balance weights, you could upload to your camera as a custom white balance. But I have no idea how to do that.

Here's a quick color level adjustment in Gimp

Thanks Bill, and I'd say that sky color is definitely more toward what I'd like to see. And, you retained the galaxies! I haven't worked with the NEF yet, but hopefully will get to this weekend and I'll look into your suggestions on how to adjust it. In the meantime I'm e-mailing you the original NEF file for you to take a look at if you want.

Mike

[Apollo XX]

Bill and Andy, thank you for the pointers, examples and suggestions! They've been big help and will go a long way towards helping me get a handle on adjusting astro photos. I've taken some of what both of you have offered and applied it to this admittedly limited exposure, and really brought it along I think. Below is the result of some tweaks (and a crop) I made to the jpeg this morning, a rendition I'm calling "Minimalist M51". I'm still a wee bit amazed that a silly little camera placed in the street and pointed at the sky for 60 seconds is capable of revealing the spiral nature of a galaxy 23 million light years from earth.



Mike M.

[Apollo XX]

Mike,

You can also use Gimp to convert from Nikon NEF raw (and other camera raw formats) with the UFRaw plugin.

Bill, thanks for the link to this interesting piece of software. I've just downloaded it and hope to try it tonight. You may be interested to know that for Windows users it is now a stand-alone only application, and no longer available as a Gimp plug-in. He does say that you could possibly build the plug-in on your own, but of course one first has to have the skills to do that. For the record, I rely very heavily on the skills of others when it comes to constructing computer software...somewhere around the level of "completely".

Mike

[Apollo XX]

O/k, this is my final iteration of this file- I swear! Thanks to some help from Andy and Bill and some time spent playing around with things, I think I was able to extract a reasonable image from this single 60 second exposure. In the middle we have the bright star on the end of the Big Dipper handle, also known as Alkaid, with M101 down to the left and M51 up to the right. I also like that it has the Mizar/Alcor duo over to the left and the trail of stars that I often use to hop to 101. Details: Nikon D5000, 50mm prime lens, F2.8, ISO800, 60sec., tracking via hand-driven hinge tracker:



Mike M.