From the July 2012 issue

How I take pictures

September 2012: Learn from a master the step-by-step process behind creating a beautiful astroimage.
By | Published: July 23, 2012
Over the years, many people have asked me to detail my “work flow.” They want to know: How do I take a picture? That may seem like a simple question, but, unfortunately, there is no easy answer because every image (as beginners soon discover) has its own personality.
In a perfect world, however, I try to follow a pattern. I begin by analyzing the nature of my target. Does it need deep exposures to allow massive stretching (see my November and December 2010 columns), or is it a bright nebula (like the Rosette) where short exposures will do? This will dictate whether I shoot 90- or 15-minute exposures and whether I combine 25 or five.
Do I shoot a separate luminance that I’ll combine with RGB data (my typical procedure for a galaxy) with my 14.5-inch f/9 reflector, or do I simply shoot RGB images (like I do for most nebulae) with my 71mm f/5.6 refractor? Tailoring your equipment and imaging parameters to the subject matter is the first step toward achieving superior results.
The Trifid Nebula (M20), shown here, is among the brightest targets in the sky, along with other summer nebulae like M8 and M17. Because of this, only short shots are necessary. For this image, the author took five RGB exposures 10 minutes in length. As always, knowledge of the subject matter is required to achieve good results. All photos: Tony Hallas
Once I have enough exposures, I go through and reject the worst. Issues like inferior seeing, bad guiding, or focus still crop up in some shots despite my best efforts.
After narrowing my selection, I use MaxIm DL version V to apply my darks and flats (see my September 2010 column), and to combine the pictures. I’m a big fan of the multi-iterative “SD Sigma Combine” for my combining, and I save the resulting 16-bit FITS file as a “master” file of the respective images. I also use MaxIm DL to align and combine my color frames into a master RGB, saved in 16-bit FITS. (For more about this process, see my January 2011 column.)
From here, I use a software algorithm on these master files that reverses the blur mostly caused from poor seeing and sharpens the images. (My favorite software for this process, called deconvolution, is AIP4WIN.)

Next, I bring these files back into MaxIm DL and convert them to 16-bit TIFF files. The advantage of this is that I can compress the files linearly to fit within the 16-bit boundaries if necessary.

M27 is a bright planetary nebula, so it too provides a lot of light. Nine frames of 20-minute RGB exposures, combined with nine frames of 30-minute exposures of Hydrogen-alpha light, were all it took to create this image.
When I am satisfied with my data reduction, I import the files into Photoshop CS5 to stretch the data using “Curves.” I also use “Levels” to reset the black point because “Curves” brightens the image. Once I have the stretched versions of the luminance and RGB, I use Photoshop to clean up the images. This might include removing satellite or airplane trails, adjusting color gradients, and dealing with any vignetting. Then, all that remains is the “fun” part: emphasizing the subtle elements of the image and using noise reduction to smooth it out.

Sometimes after my first “journey” through the data, I realize a different approach would work better. With difficult photos, I rework the image several times, slowly “getting to know” the data and the image better. This is common — accomplished imagers often realize that the best approach to the processing only becomes obvious after the first try. So don’t feel discouraged if your initial attempt didn’t come out perfect. Just go back and make it better.

Making an A-grade image requires many steps. As with so much, you’ll find that good planning will help considerably.