From: "Milton Aupperle" <milton@outcastsoft.com>
Date: September 30, 2008 5:57:37 PM MDT
To: Astro_IIDC@yahoogroups.com
Subject: UnFiltered Versus Light Pollution Filtered Urban Imaging
Hi Folks;
Seeing was quite good last night, so I decided to run some tests
comparing unfiltered and IDAS Light Pollution Reduction (LPR) filtered
imaging of M27. A lot of people believe that using an LPR filters is
basically pointless, as the the pollution and star light you lose with
the filter is more than one losses just shooting without a filter, so
I wanted to see whether this holds true here in urban Calgary.
Darkness and transparence wise, I could easily see the bottom two
stars of the constellation Lyra at zenith last night, which means a
visual limiting mag of some where between 3.5 and 4.2. On a very good
night (twice since I moved here in July), I can see all 7 stars in
Lyra, which would be Mag 4.6. I am surrounded by a mix of mercury
vapor lamps, sodium lamps and incandescent lighting from a major road
way 4 blocks west and a huge shopping mall to the north in this area
which makes my sky a darkish gray color, so it's fairly broad spectrum.
For the tests, I was using my EXHAD Grasshopper 1384x1036 Mono 16 bit
Camera on the Celestron 8" with a Focal Reducer to give me a focal
length of 1100 mm (F 5.5 effective). I shot 1, 3 and 7 minute exposure
using a Astro Don IR Cut Filter and then 1, 3 and 7 minute using an
IDAS LPR Filter with the Astro Don IR Cut Filter. I'm only showing the
comparison of the 7 minute exposures here:
http://groups.yahoo.com/group/Astro_IIDC/files/Other/NonVSLPRFiltered.jpg
Figure 1 is the unfiltered RAW image and Figure 2 is the LPR Filtered
RAW image, both 7 minute exposures and exactly the same camera
settings. The LPR Image would have been about 10 degrees lower in
altitude than the unfiltered image, so it would have been more
affected by light pollution. The background for the Figure 1 image is
39,900 out of 65,535 and for the Figure 2 image is 17,000, so the
unfiltered image background is 234% brighter than the LPR filtered
image. Also using PhotoMetric tools in Astro IIDC I measured the FWHM,
HFD and pixel count values for stars in both images and found that
difference in star brightness is 1.4 to 1.5 times higher in the
unfiltered image than the LPR image.
To reduce the background to 10,000 (out of 65,535) and isolate the
star and nebula from the light pollution I used Astro IIDC's cutoff
and expand. To achieve this I had to use a cutoff of 34,000 for
Figure 3 the unfiltered image and 7,500 for Figure 4 the LPR image,
which created some issues.
The primary issue is that the higher the cutoff value, the higher the
image contrast is and the lower the dynamic range becomes. This makes
the stars "pop" and edges of the nebula appear harder, however you
lose details in the nebula and you wind up with stars as white
"holes". This leads us to the LPR Filtered Image 5, which I have
processed a bit more to bring up the brightness, however without
blowing the stars into white holes and also bringing up the subtler
details in the nebula too.
My conclusions is that although the LPR filter does cause light loss,
it is not useless. Firstly, I can expose the CCD for up to 2.34 times
as long before I reach the same background light pollution density I
would without a Filter. Secondly, the loss of "Good Light" is only 1.4
to 1.5 times for the filtered image. So if I exposed the CCD for 1.5
times longer, I would still have a background that 0.9 to 0.8 times as
dark as unfiltered and better maintain the Dynamic range for subtle
features.
One remaining thing I have not yet worked through is how using the LPR
affects the Color Balance with my Astro Don "E" Series RGB filters.
When I used the LPR filter with my Bayer Color Flea camera, I used to
have very different Red and Blue gains than I would without the LPR
when imaging the moon. Typically that was less Blue and more Red gain,
but the exact amounts of difference I have yet to work through.
Hope this is useful to some of you.
Milton Aupperle