Filter 'promises' crisp images in low light...

[Blue Velvet]

A year from now, capturing a crisp, clear image of a candlelit birthday party could be a piece of cake - even with a camera phone.

Eastman Kodak Co. said Thursday it has developed a color-filter technology that at least doubles the sensitivity to light of the image sensor in every digital camera, enabling shutterbugs to take better pictures in poor light.

"Low light can mean trying to get a good image indoors of your kid blowing out the birthday candles. It can mean you want to take a photograph on a street corner in Paris at midnight," said Chris McNiffe, general manager of the photography company's image sensor business. "We're talking about a 2-to-4-times improvement in (light) sensitivity."

Analyst Chris Chute doesn't doubt that the new filter system, intended to supplant an industry-standard filter pattern designed by Kodak scientist Bryce Bayer in 1976, represents a breakthrough in boosting photo quality - especially when light conditions are not ideal.

The new method, which has been under development for more than five years, adds "panchromatic" cells that are sensitive to all wavelengths of visible light and collect a larger amount of light striking the sensor. Tailoring software algorithms to this unique new pattern enables faster shutter speeds, which reduces blurring when capturing a moving subject, McNiffe said.

http://news.wired.com/dynamic/stories/L/LOW_LIGHT_PHOTOS?SITE=WIRE&SECTION=HOME&TEMPLATE=DEFAULT


Just thought some of you might find this interesting...

 

[Westside guy]

That looks quite interesting - I'd imagine it'll take much more than a year before we see it in our dSLRs though.

 

[Silentwave]

DPReview says the trade-off is 1/4 the color resolution

I'll believe it when I see it.

1-2 stops is still not enough for a camera phone to get a good shot in most situations IMHO. Especially when the 'lens' is no more than a plastic disc.

 

[j/k/Andy]

1-2 stops is still not enough for a camera phone to get a good shot in most situations IMHO. Especially when the 'lens' is no more than a plastic disc.

Agreed, while a stop helps, it not that much. I think improvements will come with cheaper fast lenses (f/2.8) and most importantly lower noise, high ISO sensors.

 

[LethalWolfe]

I just saw this on another forum and agree w/some of the people there that this will most likely end up in consumer gear where you are dealing w/smaller sensors (the article specifically talks about cell phones) and situations where the reduction in color accurately is acceptable.


Lethal

 

[stcanard]

What got me was the example showing how hard it is to get a clear image at 1/60s

http://www.dpreview.com/news/0706/kodakiss_CoffeeShop.jpg

Judging by the subject matter I assume they aren't using a long focal length, maybe the photographer should worry less about technology and more about how to hold a camera...

 

[Butthead]

What got me was the example showing how hard it is to get a clear image at 1/60s

http://www.dpreview.com/news/0706/kodakiss_CoffeeShop.jpg

Judging by the subject matter I assume they aren't using a long focal length, maybe the photographer should worry less about technology and more about how to hold a camera...

Just goes to show you how valuable image stabilization technology is, Ken Rockwell gets sharp images from a compact digicam @ 1/4th sec(but not 100% of the time).

Holding a camera super steady isn't always an option...think, top of the line digicam, big lens, fast moving subject at night, arms extened to aim camera over the heads of other Paps...Paps chasing down Paris :0

This is great news!

 

[stcanard]

Just goes to show you how valuable image stabilization technology is, Ken Rockwell gets sharp images from a compact digicam @ 1/4th sec(but not 100% of the time).

Holding a camera super steady isn't always an option...think, top of the line digicam, big lens, fast moving subject at night, arms extened to aim camera over the heads of other Paps...Paps chasing down Paris :0

This is great news!

Sure, in the situation you describe. But not their example -- small lens, decent shutter speed and light, stationary subjects...

 

[Butthead]

DPReview says the trade-off is 1/4 the color resolution

I'll believe it when I see it.

1-2 stops is still not enough for a camera phone to get a good shot in most situations IMHO. Especially when the 'lens' is no more than a plastic disc.

http://www.dpreview.com/news/0706/07061401kodakhighsens.asp

the tradeoff is of course color resolution which is effectively a quarter of the traditional Bayer pattern. This interesting development will of course only be proven when we see it actually implemented and we can compare it to traditional Bayer.

.

Yes, engineering samples to be sent out in 2008, means at least 2009 before they hit the retail markets. Only then when it goes into actual production will we find out how it compares as if the color resolution is improved by then, and/or if this is that significant in terms of overall performance.

Take a look the 1st photo comparison, 2nd one at 1/10th ISO1000 with low noise, and more importantly, far greater dynamic range, you can see clearly items in shadow/darker areas that are almost completely gone in the upper photo. Very impressive, IMHO.

 

[Butthead]

Sure, in the situation you describe. But not their example -- small lens, decent shutter speed and light, stationary subjects...

How small is the lens, doesn't say. Do you consider 1/1000th a 'decent' shutter speed? I've botched more than I want to count both digital and 35mm shots at that 'decent' speed shown for comparison. Ken Rockwell is correct, even @1/1000 you will not get 100% sharp images, very high percentage, but not 100%. Anything that can improve upon picture taking, raise the percentage to a higher level, makes it more fool proof and is to be commended.

 

[Butthead]

Agreed, while a stop helps, it not that much. I think improvements will come with cheaper fast lenses (f/2.8) and most importantly lower noise, high ISO sensors.

Umm, hello, this Kodak technology allows for both of those, did you not read?

Sure it allows for smaller pixels to get 1/2 way decent (probably even better than Fuji's Super CCD, but then again, in another year Fuji may have their own revised sensor that costs less to manufactuer, or someone else may in that time) high MP sensor in compact digicams or even smaller cellphone cams. But the technology could also be used to produced either 33MP sensors for dSLR's or 144MP medium format Hasseblads that have the equivalent resolution of 4x5 large format film cameras, all at high ISO's...insanely great!

Comparing the ISO1000 shot, not only is the noise level (if this was shot with a compact pocket sized digicam sensor) looking like that of a dSLR, the dynamic range there is the most impressive part of that image, it looks phenominal compared to the image above, which is what you'd typically get with small pocketable digicams. Outstanding.

The Canon TX1 with this sensor would take insanely great 2k HD resolution video, not just noisy 720p. 5yrs down the line when flash mem gets cheap and into TB capacities, 4k resolution video (well Samsung's already got a silly, super noisy, completely useless in all but bright light, 10MP cellphone cam out in Korea) in a hybrid pocket sized digicam...w00t!

 

[stcanard]

How small is the lens, doesn't say. Do you consider 1/1000th a 'decent' shutter speed? I've botched more than I want to count both digital and 35mm shots at that 'decent' speed shown for comparison.

Looking at the image I linked, I'm going to assume the picture is not from a stalker or paparazzi so we can assume for that image they are using something with < 70mm focal length. Probably half that.

If you cannot get a reasonably decent image on that size lens at 1/60s you would be better served practising your shooting, rather than investing in new technology. Yes, you will botch some but handheld at that focal length at 1/60s you should get passably sharp. Especially considering the price you pay in colour resolution with those sensors, the tradeoff would not be worth it _in that situation_

It is quite possible that this technology will be useful, my point was that their choice of test image was awfully unrealistic.

I am one of those who is of the opinion that 90% of Ken Rockwell's opinions aren't worth the paper (or pixels) they are written on, so I will avoid even discussing that.

 

[cube]

This is not restricted to Kodak sensors. Anyone can use (license?) this basic tech.