Hacker News Reader

Comment by jcr | Hacker News Reader

I doubt I'm the only one who read the article with the hope of finding some new technique to remove light pollution from images to show off the stars... and was mildly disappointed to find out the beautiful images were just cut-n-paste jobs (albeit outstanding ones).

I'm wondering what kinds of techniques exist for removing light pollution?

Spectrum sensitivity? Spectrum filtering on exposure? Post processing?

View on HN

Tucson, Arizona uses sodium vapor lighting for all its street lighting specifically to allow the nearby observatories easily filter the light pollution.

I attended a conference in Tucson last fall and...being jetlagged, did a nice walk at 4AM in the desert behind our resort. It was really amazing, I hadn't seen shooting stars in decades.

It's only a couple hours from Tucson to the Kitt Peak Observatory, where they have astronomy programs for the public. I talked my wife into doing it when we were there visiting friends last year.

I highly recommend it. The astronomy is fairly basic, but the experience is incredible: You're up on a big mountain, surrounded by gigantic telescopes, in the middle of a vast desert wilderness. It's freezing cold outside after dark. When the program is over, you have to drive down the upper half of the mountain without headlights to protect the observations that are happening.

I went to scout camp in the Chiricahua mountains in SE Arizona and experienced the same thing. I'd sneak out if my tent at night to look at the night sky (until my scoutmaster chased a bear out of our camp...the guy was psycho!).

Similar situation in San Jose, CA: http://mthamilton.ucolick.org/public/lighting/Cooperation2.h...

I grew up in Tucson, I do remember amazing stars at night. I also remember seeing an observatory on Mount Lemmon. Never made that connection tho.

It helps that sodium lights produce more lumens/watt than the alternatives.

You most certainly are not.

I wonder if you could get it done with some kind of manual HDR: tracing out the skyline to explicitly say "stuff below the line should be from the low exposure photo, stuff above from the high exposure photo."

Disclaimer: I know nothing about photography.

The problem with light pollution is that the light bounces off of water, dust, and smog in the atmosphere, and back down into your eye/camera. So when you look up, the sky isn't darker, it's a lot brighter! You would need a way to subtract the extra light.

Theoretically, it would basically be just a matter of subtracting the "general" light background, and then apply a deconvolution to the remaining "signal" to reduce blurry areas of light into point-like stars.

However, deconvolution is one of those things that's great in theory, but almost impossible to pull off in practice, except under extremely controlled circumstances. And in this case, considering just how much dust/smog/etc. is in the atmosphere, and how it is non-uniform, I expect it would be simply impossible.

There's a lot of that kind of stuff that goes into signal-cleaning in observatories, especially those located near major cities (like the Lick Observatory outside San Jose). But it tends to be pretty site/telescope-specific, full of in-house custom code, so I'm not sure if there's anything good that a regular photographer could use to produce the same effect.

Though it won't help our eyes any time soon, something along the line of "Adaptive Optics" [1] could possibly address at least some of the issues you mentioned when creating images. Getting tech from the world's beset telescopes into consumer/pro camera gear might be asking a bit much at the moment, but eventually it could become possible. If it was done, a big issue would be having countless people shining lasers into the sky.

[1] http://en.wikipedia.org/wiki/Adaptive_optics

I read it as a brilliant and elegant hack, a sort of real-world magic. As Vi Hart said (referring to how she builds her comfortable-looking fast-cut accelerated videos of handwriting) it's "tricking your brain into seeing what was actually there."

Here's a shot of the Milky Way over Los Angeles, taken over Los Angeles [1]. The first comment explains the technique. There's still cut-n-paste going on, because the exposure lengths for stars and for city lights are so different---and also the stars are moving. But you might consider it slightly more "honest."

[1] http://www.flickr.com/photos/magneticlobster/5941836830/

Reply on news.ycombinator.com