[Leica] Color Astronomy With Monochrom

[Howard Ritter]

Bob—

What celestial objects are you planning to image? What equipment are you expecting to use? This astronomer’s opinion is that unless you’re planning to do the sort of color-critical imaging that’s usually done by battle-hardened astroimaging veterans with highly developed color sensibilities and idiosyncratic preferences, dedicated electronically cooled CCD astrocameras that cost as much as—to much more than—a prosumer DSLR, attached to an astronomical telescope with computer-controlled sky-tracking mount that each cost as much as the camera did, you’re not likely to see any advantage from using your M Monochrom to make (L)RGB images and combining them to get a final color image over using a color M to make a color image directly. Think of the difference between off-roading in your suburban SUV and those who do it in purpose-built Jeeps with stump-yanker gearing, tires with knobs the size of hen’s eggs, and winches front and rear.

First, there is no a priori reason that the Bayer color filters on a digital camera should produce a “truer” or more pleasing color balance than dedicated narrowband filters as used in dedicated CCD astrocameras. The primary reason for using a panchromatic sensor with outboard color filters is to get better detail in the final image and to allow separate processing of each color channel for fine-tuning the final image. Given that each photosite in a Bayer-filtered sensor contributes to the structural detail of the image just as much as each photosite in an unfiltered sensor does, that the individual color channels can be accessed readily in raw or DNG files, and that the structural resolution of the final image is far more important esthetically than the color resolution, this practice has all but ceased to offer advantages. Modern higher-MPx sensors and sophisticated de-Baying algorithms have essentially eliminated any advantage to making and combining RGB images from a panchromatic sensor. Any small differences in color in the final images obtained through on-chip Bayer filters and through narrowband filters on a panchromatic CCD chip are less than the differences produced by different manufacturers’ filter sets and different imagers’ own personal processing techniques.

The following is in the event that you want to proceed anyway!

In general, the “truest” (a highly subjective term of art!) astroimages in the age of CCD astrocameras have been obtained with panchromatic (monochrome) cameras and narrow-band filters, making two, three, or four exposures in different wavelengths and combining them in post-processing. Different exposure lengths may be used for each color, and in post-processing, each raw image can be mapped to a selected color. There is also a great deal of literature, and controversy, on what constitutes the “real” color of faint, nebulous deep-sky objects, which are seen visually in even the largest telescopes as shades of grey, and of how best to obtain that “reality” in the final image. All of this is why I said that “truest” is a subjective term. For images of the planets and the Moon, the advantages of taking and combining multiple monochrome images at different wavelengths are nonexistent, and the practice is a waste of time and effort. For these objects, a “one-shot color” astrocamera will suffice, or alternatively, for the essentially colorless Moon, a single unfiltered image from a panchromatic camera will be perfectly satisfactory.

Most commonly, R, G, B, and L (a luminance “filter” is a colorless piece of glass, used rather than no filter at all in order to preserve the same focus as with the colored filters, the purpose being to provide overall luminance information) filters are used in a panchromatic astrocamera, one that has a sensor with no integrated Bayer filter. For some gaseous nebulae that radiate primarily (or exclusively) by fluorescence, filters for, e.g., the hydrogen-alpha or doubly-ionized oxygen (O-III) filters are used. Since the image quality is most sensitive to physical resolution, whereas color may be applied rather crudely by comparison and still produce a visually excellent image, it is not uncommon practice, in the interest of optimizing the use of limited resources, to combine a luminance (i.e., B&W) image made with a large-aperture telescope (for best detail) with RGB images made with smaller instruments. 

However, the most modern one-shot color astrocameras, whose sensors have built-in Bayer filters, are said to produce color images that are equivalent to those obtained with panchromatic unfiltered sensors and outboard filters. Again, plentiful literature and controversy.

For the astroimaging community, the requisite filters are generally available in the two barrel diameters that are standard in astronomical telescopes, 1.25” and 2”. Some are available as 2” x 2” unmounted glass. Google “color filters astroimaging” and look for articles and for links to manufacturers, for example the German Astronomik and the American AstroDon. And these filters are indeed associated with visible color: the R filter is very red, the G is green, and the B is blue, while the L is colorless.

Another option, which I assume is the one you’re considering, is to use what we’d all recognize as a “camera”, usually a DSLR but in your case obviously the M Monochrom, which is the only B&W digital camera on the market. The evolution of high-MPx, low-noise sensors has allowed the performance of premium general-purpose consumer/prosumer 35mm cameras to rival that of purpose-made astrocameras, especially for imaging brighter objects that do not require hours of exposure and, consequently, active cooling with a built-in thermoelectric cooler. The “hot hand”, no pun intended, in this respect is the Nikon D810A, a camera aimed at the astroimaging community, that lacks both the IR filter (to better capture deep-red and H-alpha wavelengths) and the anti-aliasing filter in order to improve structural resolution.

Bottom line: Unless you plan to put your Monochrom on an astronomical telescope with a computerized drive to make long-exposure astroimages, I’d stick with using an M9 or Typ 240 for images made with camera lenses. If you plan to make wide-field sky images with camera lenses, I don’t think that LRGB offers any advantages.

Coincidentally, I have gotten my scopes set up after a move from Ohio to NC, and have just received my M Monochrom in trade-up from my sensor-corroded M9. I’m going to be trying some astrophotography with it. In addition, the M8 would seem to be an excellent candidate for making images of nebulae that emit primarily H-alpha, so I’ll be trying that as well. I’ll post what I manage to get.

—howard



> On Feb 28, 2016, at 1:10 PM, Robert Adler <rgacpa at gmail.com> wrote:
> 
> I am hoping to get some guidance from the astronomy experts here. A little
> research shows that by using what are called "LRGB" (Luminance, red, green,
> and blue) filters with monochromatic sensors, one can combine the images
> taken with these filters to come up with astrophotos with very true color.
> 
> Does anyone know where to get these special filters (they are not
> associated with visible color, but more wavelength transmission). I would
> either need Series VII (for the 21/1,4) and E46 for the 35/1,4 or 4x4
> filters for a Lee holder (which I would somehow need to fit to my Leica
> (Seven5 filters would be great, but probably impossible to find).
> 
> Any leads or experience would be greatly appreciated.
> Thanks,
> Bob
> -- 
> Bob Adler
> www.robertadlerphotography.com
> 
> _______________________________________________
> Leica Users Group.
> See http://leica-users.org/mailman/listinfo/lug for more information