backtop


Print E-mail del.icio.us 43 comment(s) - last by Oregonian2.. on Jun 18 at 2:36 PM


Comparison for action photography

Comparison in low-light photography
New Kodak camera technology could make dark, blurry photos as thing of the past

Eastman Kodak Company today announced what it considers a “groundbreaking advancement” in image sensor technology that will help reduce the accidental taking of dark and blurry digital photos. Kodak claims its new sensor technology provides a two- to four-fold increase in sensitivity to light (from one to two photographic stops) compared to current sensor designs.

“This represents a new generation of image sensor technology and addresses one of the great challenges facing our industry – how to capture crisp, clear digital images in a poorly lit environment,” said Chris McNiffe, General Manager of Kodak’s Image Sensor Solutions group. “This is a truly innovative approach to improving digital photography in all forms, and it highlights Kodak’s unique ability to differentiate its products by delivering advanced digital technologies that really make a difference to the consumer.”

Image sensors convert light into electric charge to capture images. Today, the design of almost all color image sensors is based on the Bayer Pattern, an arrangement of red, green, and blue pixels that was first developed by Kodak scientist Dr. Bryce Bayer in 1976. In this design, half of the pixels on the sensor are used to collect green light, with the remaining pixels split evenly between sensitivity to red and blue light. After exposure, software reconstructs a full color signal for each pixel in the final image.

Kodak’s new proprietary technology, invented by John Compton and John Hamilton, adds panchromatic, or clear pixels to the red, green, and blue elements that form the image sensor array. Since these pixels are sensitive to all wavelengths of visible light, they collect a significantly higher proportion of the light striking the sensor. By matching these pixel arrangements with advanced software algorithms from Kodak that are optimized for these new patterns, users can realize an increase in photographic speed, directly improving performance when taking pictures under low light.

Inventor John Hamilton explains in the Kodak blog, “One way that helps to think about this is to look at it in terms of luminance and chrominance. In the original Bayer design, the green pixels are used to recover most of the luminance information from the image. Now, we are using panchromatic pixels - which are more sensitive than green pixels, because none of the photons get filtered out or wasted - to act as the luminance. This gives us a more sensitive luminance channel in the final image, which raises the sensitivity of the entire sensor.”

John Compton adds, “The real advantage is that the panchromatic pixels are more sensitive, since they detect all wavelengths of visible light (rather than filtering light to detect color information).” Hamilton points out that the improved images don’t come from using panchromatic pixels, but rather the more accurate luminance data.

Kodak’s new technology also enables faster shutter speeds (to reduce motion blur when imaging moving subjects), as well as the design of smaller pixels (leading to higher resolutions in a given optical format) while retaining performance.

As for which situations benefit most from the improved sensor, Hamilton says, “In situations where you want more sensitivity to light. In a low-light situation, these new patterns will produce a lot less color noise than a Bayer pattern sensor. You can run the shutter faster, which gets rid of a lot of motion artifacts. It will cut down on camera shake or, if you're taking a picture of a moving object there will be less blur.”

“Another way to think of this is that you have the same number of photons coming into the new sensor as you would with the Bayer pattern,” adds Compton. “It's just that the new filter arrays waste fewer of the photons since fewer of them end up absorbed in a color filter.”

The inventors say that the technology is appropriate for use with both CCD and CMOS image sensors, and that Kodak is developing this technology for consumer markets such as digital still cameras and camera phones. The technology may also be applied to other image sensor devices for industrial and scientific imaging. The first Kodak sensor to use this technology is expected to be available for sampling in the first quarter of 2008.

Kodak’s announcement comes on the heels of advancements in low-light photography made by South Korea's Electronic Technology Institute. The institute announced in late May that it had developed an image sensor for use in extreme low-light conditions and would be suitable for use in tunnels, night clubs and bars.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

Great noise reducer
By PitViper007 on 6/15/2007 10:49:42 AM , Rating: 3
While all three sets of photos show a marked improvement, that last one really tells the tale. That's a true low-light situation, shot at a high ISO. The traditional sensor, shooting at ISO 1000 is predictably very noisy, but the new sensor at the same ISO shows a greatly reduced noise level. Consider me sold on this. I'll definitely be getting a camera with this new sensor, once they hit the market.

PitViper




RE: Great noise reducer
By bpt8056 on 6/15/2007 11:23:54 AM , Rating: 2
I think the noise level in the 3rd picture (top) for ISO 1000 is greatly exaggerated. There aren't any samples that I can find over at dpreview.com that shows anything like that.


RE: Great noise reducer
By noxipoo on 6/15/07, Rating: 0
RE: Great noise reducer
By AnnihilatorX on 6/15/2007 12:21:19 PM , Rating: 1
you beat me to it

Well I guess they are okay for demonstration even if they are fake. The general public won't know

I see no problem with this if they really have a product that works


RE: Great noise reducer
By Oregonian2 on 6/15/2007 1:33:41 PM , Rating: 4
Nope, they were taken at the same time using two cameras that were one above the other. The photo on the top was below the camera on the bottom.

I turned the photo sideways and did a stereoscopic 3D free-view of it (crosseyed fashion) and got a decently good 3D depth, just using the reduced version "above".

In other words the perspective is *slightly* different between the two images!

I've been doing 3D photography since the late 70's or so.


RE: Great noise reducer
By Oregonian2 on 6/15/2007 1:46:18 PM , Rating: 2
Actually the depth is much more pronounced if one assumes the cameras were side by side with the bottom image being the right camera. So with a bit more careful look, I'd say they were side by side, but there was a little up-down misalignment of the camera lenses too (as a guess to explain why there's some depth visible looking at it the "wrong" way).


RE: Great noise reducer
By Oregonian2 on 6/15/2007 1:47:34 PM , Rating: 2
PS - I've been looking at the last (third) one.


RE: Great noise reducer
By spluurfg on 6/16/2007 2:55:57 AM , Rating: 2
I'm guessing they've been photoshopped just to show what the sensitivity advantage would look like. They don't tell us what sensor size is used -- the 'improved' output on #3 would still be lousy if it were from an APS-C or full frame DSLR sensor at ISO1600. Also, they're talking about proposed changes in the Bayer layout -- they might not have even made a prototype yet.

I think one advantage that they don't seem to mention would be an increase in dynamic range -- since the RGB photosites are absorbing 1/3 of the light each, and the panchromatic photosites are absorbing the full wavelength, you essentially have photosites with varying sensitivity. Hopefully this should increase the ability to record highlights and shadows, similar to the Fuji super-CCD in the S5 Pro.


RE: Great noise reducer
By greenchasch on 6/18/2007 1:57:03 PM , Rating: 2
I dont think you can conclude that just from looking at the pics.


RE: Great noise reducer
By Oregonian2 on 6/18/2007 2:36:18 PM , Rating: 2
All we know is that they were taken at the same time with side by side cameras (as I mentioned earlier in this thread, and how I know it was done that way). They may have post-processed both pictures to increase effect, but I doubt it. They probably would have just set their controlled lighting conditions to be such that the edge-case of the cameras would be shown -- where 1.5 stops would be significant.


RE: Great noise reducer
By drunkenmastermind on 6/16/2007 8:31:42 PM , Rating: 2
Look closly numbnut, they are different. Anyway I think the whole lot are crap, I got better images 5 years ago.


RE: Great noise reducer
By AnnihilatorX on 6/15/2007 12:19:44 PM , Rating: 1
They do look like they have been deliberately photoshopped.
Especially the third one

The photo composition is 100% identical as far as I can see which is impossible if they are using live people and to be shot with different camera anyway


RE: Great noise reducer
By IGoodwin on 6/15/2007 12:46:09 PM , Rating: 2
The pictures are different. If you look at the last picture taken in the low light levels and look at the girl sitting down, you can see that her left arm is at a slightly different perspective.

Looking into the glass behind, you can see the flash from two camera's from, what appears to be a head. Looks like they had two cameras side by side, triggered by the same click.

That would appear to be a valid test. If these are fake, then they are better at it than you think.


RE: Great noise reducer
By IGoodwin on 6/15/2007 12:50:31 PM , Rating: 2
Yes, I'm aware that flash photography wasn't used. Still, the image in the glass does look like it has glowing eyes.

Still doesn't alter the different perspectives in the two images.


RE: Great noise reducer
By fic2 on 6/15/2007 1:29:46 PM , Rating: 3
A better thing to look to see they are side by side:
First picture series - the chair that the woman on the left is sitting in. In one picture the arm/back curve is more fully visible.
In the seconde series - the wood thing with the yellow border in the background has it's bottom corner just at the edge of the image on the left side. In the right side photo it is definitely inside the frame by quite a bit.
The third series appears to be top/bottom instead of side to side. The woman in the foreground you can see part of her left arm below the sleeve in the top picture. Also the background woman you can see her left shoe in the top picture. And the thing on the pedistal in the bottom right corner changes vertical position.


RE: Great noise reducer
By phattyboombatty on 6/15/2007 4:42:16 PM , Rating: 2
Also, if you look at the second picture, the sticks (or whatever they are called) that the guy is juggling are in different positions in the two versions. It is most apparent if you look at the stick closest to the guy's left hand. Most likely the two cameras used have a different shutter lag producing a time delay of a few milliseconds between the two pictures.


RE: Great noise reducer
By phattyboombatty on 6/15/2007 4:43:57 PM , Rating: 1
Should have said right hand, not left hand.


RE: Great noise reducer
By TheDoc9 on 6/15/2007 11:49:00 AM , Rating: 2
I'm sold on this as well, I was extreemly disapointed at my latest digital camera purchase for just this reason, and have since learned that most if not all digital cameras suffer from this in some form. I like my current camara and have since tweaked the hell out of it to minimize motion blur and noise, but I've never had to stand so still when taking a picture as compared to even a throw away 400 speed film camera.


RE: Great noise reducer