February 7, 2013

Panasonic's Revolutionary New 'Micro Color Splitter' Sensor Filter Doubles Light Sensitivity

The most common way we get color images with digital cameras is with a Bayer pattern CMOS sensor, but there are plenty of variations on that design being used today. The upcoming Aaton Penelope Delta uses a Bayer pattern over a Dalsa CCD, for example, while the RED EPIC-M Monochrome uses the MX CMOS sensor foregoing color filtration entirely. By their very nature, though, color filters of any kind cut down the amount of light transmitted to the sensor. That's why Panasonic is developing a brand new type of color filter that will employ diffraction to split up the color spectrum, instead of filtration, and thus will be capable of doubling the light sensitivity of the sensor.

Panasonic's Sensor Filters By Way of 'Micro' Diffraction

Here's the bulk of information from Panasonic's announcement, with thanks to 43Rumors for the find:

Panasonic Corporation has developed unique "micro color splitters", which separate the light that falls on image sensors by exploiting light's wavelike properties. Applying them to actual image sensors allows bright color images to be achieved even under low-light conditions. This development makes color filters unnecessary by using the micro color splitters that control the diffraction of light at a microscopic level. Panasonic has achieved approximately double the color sensitivity in comparison with conventional sensors that use color filters.

The developed technology has the following features.

  1. Using color alignment, which can use light more efficiently, instead of color filters, vivid color photographs can be taken at half the light levels needed by conventional sensors.
  2. Micro color splitters can simply replace the color filters in conventional image sensors, and are not dependent on the type of image sensor (CCD or CMOS) underneath.
  3. Micro color splitters can be fabricated using inorganic materials and existing semiconductor fabrication processes.

Followed by a bit more on how all this works:

Since light separated by micro color splitters falls on the detectors in an overlapping manner, a new pixel layout and design algorithm are needed. The layout scheme is combined and optimized using an arithmetic processing technique designed specifically for mixed color signals. The result is highly sensitive and precise color reproduction. For example, if the structure separates light into a certain color and its complementary color, color pixels of white + red, white - red, white + blue, and white - blue are obtained and, using the arithmetic processing technique, are translated into normal color images without any loss of resolution.

They seem to be saying in their materials that there is no loss of resolution due to using this method. This would be in contrast to a traditional Bayer pattern filter which loses some resolution thanks to interpolation, since not every color is represented on every pixel (unlike your 3-CMOS or 3-CCD cameras). In practice it may not be perfect, but this development is definitely exciting, as it looks like it will improve sensor performance without needing to completely redesign the sensor or the manufacturing process. Developments like this are important as sensors reach their physical limit of light sensitivity and color reproduction at given sizes -- like the Super 35mm/APS-C size, which RED is certainly pushing to its limits with their new Dragon sensor.

Some Background Material

Some additional context may be useful when considering the 'Highly Sensitive Image Sensors Using Micro Color Splitters' featured in Panasonic's announcement. Panasonic's new sensor technology is by no means the first to deviate from the Bayer pattern standard (which our own Joe Marine explains here). It is also not the first to achieve full-color reproduction without a Bayer pattern, either:

Other Unique Sensor Designs

  • The Sigma/Foveon X3 sensor actually uses three stacked photodiode layers so that each pixel has its own complete RGB information (pictured right, courtesy Sigma). This is possible due to something called 'wavelength-dependent absorption depth' [PDF] meaning each layer only absorbs part of the spectrum. Used in the SD1 Merrill and other Sigma stills camera models.
  • Sony's CineAlta F35 HD digital cinema camera uses a CCD with a very unique color filter array. Instead of the bayer 'RGBGxGBGR' pattern, the F35's 5760x2160 sensor is arranged in a 'RGBxRGB' pattern that averages rectangular blocks of 6 (every 3 horiz. x 2 vert.) sensor pixels into each final pixel of the 1920x1080 image. This gives the F35 powerful color fidelity and SNR [via PVC's Art Adams].
  • The Aaton Penelope Delta uses an ISO 800 Dalsa CCD that will actually have the ability to physically oscillate in place by 1/2 pixel each frame. This 'time traveling' sensor theoretically increases its 3.5K RAW yield to 7K RAW by this action, due to alleviation of the fixed noise structure inherent to any image sensor.
  • Lytro and plenoptic/light field photography are not yet fully developed technologies, but in terms of unique sensors they at least deserve an honorable mention. Lytro technology allows for refocusing and perspective shift after the photo is taken because its micro-lens array gathers light from different directions simultaneously.

For more general reading material on imager comparison, you can check out Dalsa's article on CMOS vs. CCD sensors.

What do you guys think?

Link: Panasonic Develops Technology for Highly Sensitive Image Sensors Using Micro Color Splitters -- Panasonic.co.jp

[via 43Rumors]

Your Comment

15 Comments

I wonder what the turn around time is on this to get to market. I also wonder how this change in pattern will affect frame rates and rolling shutter, or is that even related?

February 7, 2013

0
Reply
kevin

Should be no changes to rates or rolling shutter. CCD basically remains constant, but with "micro prisms?".
Polished, this could work very nicely IMHO. I'm interested in how this might make up for the weak energy of long wavelength visible (red) energy. Might make tungsten, low color temp etc, shooting much better, not only by reducing filter effect but subtly raising photo-site activation. A thing to notice though is that the prisms produce a colour contamination spread and send it to adjacent sites...to be continued!

February 7, 2013

-2
Reply
jd holloway

I love it when you guys write articles like this. Much respect.

February 7, 2013

-2
Reply
Some young guy

Appreciate it, Guy :)

February 7, 2013

-1
Reply
avatar
Dave Kendricken
Writer
Freelancer

I second that.

February 7, 2013

-1
Reply

Basically all I understood from that was... technology continues to advance outrageously fast and by the time you get around to picking a professional camera there will be 10 better ones for half the cost...

I love living in the present.

February 7, 2013

-1
Reply
James

In fairness James the typical time to market or mass production on anything from this stage ranges from decades to infinity (as in never released). The only exceptions are on the insanely priced high end cameras who obviously can afford to employ the new low yield/costly sensor designs. Id be amazed if we see this in under 5 years...but could be wrong...

February 7, 2013

-2
Reply
Peter

absolutely awesome
Hope we see it in a camera soon (by end of 2013)!!

February 7, 2013

0
Reply
Luke

Neat-o, now start implementing it in the new cameras, while Canon is... oh, that's right, Canon... hahahahahahaha

February 7, 2013

2
Reply
Natt

Haven't the professional camcorders (2/3") always used diffraction on their 3-CCD systems?

At least that's how I learned during my job training how a camera works: Light goes through the lens, gets split up by a prisma and then hits the three CCD sensors.

So the real news here is that this is now a new method of on-sensor diffraction for single-sensor cmos cameras.
Diffraction as a way to split light into three colors isn't news by itself.

February 8, 2013

-1
Reply
Heiko

February 8, 2013

0
Reply
Heiko

The magic word is REFRACTION. Dont confuse with DIFRACTION.

February 9, 2013

-1
Reply
Simon

Ups, mixed that up.
To my excuse I am German and we don't use the Latin terms refraction and diffraction but the corresponding German terms Brechung and Beugung. So it's not like I don't know the difference, I just translated it wrong ;)

Still not a good excuse because I learned Latin in school and I wasn't bad at physics neither, so I should have known :)

February 10, 2013

-1
Reply
Heiko

Is this a complimentary colors scheme, the sort used in consumer but scrapped for rgb and Bayer on professional due to accuracy? Is this low light consumer?

February 8, 2013

-2
Reply
question

Judging from the images I'd say they are gaining about a stop of light, right? That's on a micro 4/3rd sensor. but how does this compare to bigger sensors like super35 or full frame? Isn't full frame already 2-3 stops faster than super35?

February 8, 2013

-2
Reply