Slashdot Mirror
Kodak Unveils Brighter CMOS Color Filters
brownsteve writes "Eastman Kodak Co. has unveiled what it says are 'next-generation color filter patterns' designed to more than double the light sensitivity of CMOS or CCD image sensors used in camera phones or digital still cameras. The new color filter system is a departure from the widely used standard Bayer pattern — an arrangement of red, green and blue pixels — also created by Kodak. While building on the Bayer pattern, the new technology adds a 'fourth pixel, which has no pigment on top,' said Michael DeLuca, market segment manager responsible for image sensor solutions at Eastman Kodak. Such 'transparent' pixels — sensitive to all visible wavelengths — are designed to absorb light. DeLuca claimed the invention is 'the next milestone' in digital photography, likening its significance to ISO 400 color film introduced in the mid-1980's."
Of course, you achieve this increased light sensitivity at the expense of losing 1/4 of your color resolution. Maybe if you want the increased sensitivity it might make more sense to pick up something like the Canon 1D Mk III, which, at least according to Ken Rockwell, gives great results all the way up to ISO 6400. I'd hate to lose 1/4 of my color resolution *all of the time* to get the added sensitivity that I only need for a small fraction of the shots I take.
and color in the 70s.
I refer you to Tri-X b/w, and to Fujichrome 400 around 1972. a really nicely balanced and warm film. if you pushed it to 1200, you could peel the grains off the base and go bowling with them, but the picture held up remarkably well on the small screen. it was THE go-to magic film for 16mm newsfilm when it came out.
if that was a negative film, it would have been asa 800 with little more grain than the "fast" 125 color film of the time.
if this is supposed to be a new economy, how come they still want my old fashioned money?
As you state, DSLRs already have fairly decent sensitivity, so this is not likely to be a good compromise for them.
Modern 'compact' digital cameras, however, which stuff 7-12 megapixels on 1/1.8" and 1/2.5" sensors (smaller than your fingernail) could benifit enormously from this. These sensors are already past the diffraction limit of most of the lenses, so a drop in color resolution may not be too damaging (the eye being less sensitive to color resolution, than luminance anyway). Kodak is claiming a 1-2 stop increase in sensitivity, which would be a great benefit to anyone using a compact inside, or in other poor light. (I have yet to own a camera that performs well above ISO 200)
As with all such tech announcements the proof is in the pudding, and until we can compare full size samples to conventional bayer sensors, its hard to tell if this is the next big thing or not.
That's a neat trick. I wonder how they can do that?
This is really not anything new to the image industry, just a new application. There is already the CMYK colorspace for printers, which is effectively an RGB + black to get deeper colors. I don't see this as really revolutionary, as much as "Can't believe this hasn't been done yet." Though, at least they admitted this too :)
My biggest hope for this is to reduce per pixel noise by being able to reference the fourth plane, but I doubt they will get there for a while, they still have to work out the color conversions.
Kodak has rediscovered what evolution found millions of years ago -- design a dual system such as the rods and cones of the biological eye. The average human eye has about 120 million sensitive, panchromatic rods and only 6 or 7 million color-sensitive cones (many in the central fovea). The brain merges the limited amounts of color information with the larger volume of B/W image data to paint color into the image that we think we see.
Two wrongs don't make a right, but three lefts do.
They divide each sensor of the regular bayer pattern to 4, half white, half color. This way one can also report a 4-fold increase in the number of pixels, without really increasing the resolution. (which actually will be a boon for digital photography, since no one needs the current resolution anyway, because the optics doesn't keep up, but a megapixel race is on...)
But does anyone know why sensors use RGB and not CMY? a Cyan filter would let green and blue through, but keep red out, instead of blocking two parts of the visible spectrum for each pixel. This way, by simply switching color space, the camera becomes twice as sensitive to light. I.e. instead of use or something like that. One could even combine the two methods, and use white pixels, to gain a slight further increase in light sensitivity (from 8/12 to 10/12). Is there any reason that current cameras use RGB?
Sure, "faster" sensors will be a boon to the consumer market, and will surely have some applications in the pro market as well -- existing light press photography come to mind.
For me, though, the problem is not so much speed as it is noise and dynamic range. That's because a lot of the time I still do fine-art level landscape and studio glamour photography -- neither of which are speed starved, but even the finest digitals could still use even less noise and wider dynamic ranges.
While DSLRs have a huge advantage over handhelds in this regard, it would still be nice to see improvements in s/n such that the darker zones maintained their clarity and detail. Even the finest Canon cameras suffer to a degree in this regard, at least for people with very high standards. Some of us have those standards because that is what our clients demand - and in some cases we still must use film to meet their criteria.
It's a virtual law that to obtain the best noise performance you need to use the lowest ISO speed that the camera can attain. So instead of bottoming out at 100, like most DSLRs, I'd like to see 25. Or better, 12.
For more info, visit http://www.normankoren.com/digital_tonality.html
They posted a full press release with images and sensor layout diagrams, additionally there is an excellent discussion in their news forum with a lot of good information. http://www.dpreview.com/news/0706/07061401kodakhig hsens.asp
Only problem is that Foveon (at least current implementation) is crap. The three colors have too much overlap and they also aren't very sensitive, either. Fine, you get rid of some of the bayer artifacts, but in return you lose most of the extreme colors and lots of sensitivity.
One of the problems with DLP projection TVs with a "color wheel" was that since every color lets only 1/3 of the light through, the picture was dim. So they added a fourth element "clear" that lets out all the light to get every projected pixel a blast of light they need and the remaining portions of the color wheel adds only additional brightness for each color.
This technology seems to be kind of similar. The transparent sub pixel detects over all lumninosity and the remaining pixels "adjust" for color. Very close to what we have in our retina too. Almost all our cylindrical cells respond only to luminosity and the cones respond, to varying degrees, three colors. A poster was complaining about losing "color resolution". I think millions of years of evolution has shown us the balance. You need about 90% of the pixels responding to luminosity and just 10% to color. The same ratio in our retina.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
The patterns they suggested in the article were not as elegant as the Bayer filter (where each color formed an evenly spaced grid). They may be hiding the actual pattern for now or there may be some technical reason for those patterns that I don't understand, but I would suggest this pattern (C = Clear):
it keeps the same 4clear:2green:1red:1blue ratio but the different color pixels all form a regularly spaced grid.
Prov 9:8 Do not rebuke mockers or they will hate you; rebuke the wise and they will love you.
While I like Kodak's idea quite a bit, here are a couple of other ideas.
1) Sony was building cameras for a while with four color channels. There was the normal green, but also a different green they called "emerald" for one of the four Bayer pattern locations. Unfortunately, this was a solution in search of a problem, it never really caught on because there just wasn't any perceived benefit.
2) I do visual effects for films. For the last 50 years or so, people have been using bluescreen and greenscreen effects. The idea is to put a constant color background, and process the image so that any pixels of that color become transparent. Over the years, more and more lipstick has been applied to this pig -- so that you can now often extract shadows that fall on the greenscreen, pull transparent smoke from the greenscreen plate -- these things have become even more possible through digital processing.
Still, it sucks. Greenscreen photography forces so many compromises that I often recommend shooting without it and laboriously hand-rotoscoping the shots.
But -- say you had a fourth color filter, with a very narrow spectral band. Perhaps the yellow sodium color -- commercial lights that put out very narrow-band yellow are sometimes used for street lighting. If you had a very narrow-band sodium filter over 1/4 of the pixels, you could pull perfect mattes without 99% of the artifacts of traditional greenscreen and bluescreen photography. Finally (and this is killer!) you could make glasses that the director of photography and other lighting crew could wear that block just that frequency, so they could see the set as it really is -- without the sodium light pollution.
Still, Kudos to Kodak for thinking outside the box.
Thad Beier
I love Mondays. On a Monday, anything is possible.
It's done on TV all the time and nobody complains (chrominance is separated from luminance and often transmitted at much lower resolution). As has been pointed out below, your eyes are made up of rods (which see black and white) and cones (which see color), and only a fraction of those cones are devoted to each individual red, green, or blue spectrum. So your color resolution is already significantly lower than your luminance resolution. You can even see photos demonstrating this with a 9x decrease in color resolution (3x in each linear direction). You're most sensitive to green, which is why the Bayer sensors commonly used in digital cameras divide each 4 pixels into GRGB.
I wonder how this is going to compare to the Foveon sensors. They capture RGB data at all pixels - filtered based on depth rather than location. Now if only those babies cost less.
Sure I'm paranoid, but am I paranoid enough?
Researchers here at Rensselaer Polytechnic Institute recently came up with a super non-reflective coating -- it basically has nano-spikes that help absorb light from all angles and at all frequencies. Seems like it would be good to use for the dark pixel. http://news.rpi.edu/update.do?artcenterkey=1956
No... not really.
First of all, the Bayer pattern is...
RG
GB
Giving up one of the four sites to wide-band sensitivity as Kodak proposes, the same spatial pattern still has exactly the same sensitivity to red and blue; nothing has changed there. Red and blue sensor sites still alternate at the exact same spatial rate. But the new pattern has 1/2 the spatial (not intensity) sensitivity to green (which we are most sensitive to, remember); it has the same sensitivity to luma; and it probably has considerably enhanced sensitivity to infrared and ultraviolet, though that remains to be seen, and such an advantage is not as generally useful to most photographers (though those who enjoy IR and/or UV photography will love this thing if the sensor is truly wide-band.)
But there are complications; such as, Bayer filters tend to produce significant moire patterns, and the filters applied to prevent that reduce the available spatial resolution by as much as 1/2 along each axis anyway.
I've written numerous RAW image plugins for Bayer (and other) patterns, and believe me, it isn't as simple as 1/4 the color. This is a new configuration, and I've not written code for it as yet, but I would bet my boots that when the time comes to do so, the color resolution of an image will not suffer much, if at all. You'll still have RGB info available at about twice the moire filter rate. Spatial resolution shouldn't suffer either, because luma information is still available from the new arrangement. In terms of color images, what I'm trying to figure out is what the perceived advantage is.
Thinking outside the box of color images, though, I can imagine a simple 1/4 resolution B&W mode that can do infrared and ultraviolet with the proper blocking filters... that'd be trippy. :-)
I've fallen off your lawn, and I can't get up.
In printing technologies, at least in the early '90s they were using a technique called either "GCR" (gray color removal) or "UCR" (under color removal) which basically transfer almost all of the "light density" information from the cyan-magenta-yellow films of a color separation to the "K" film (black) -- because black ink is quite a bit cheaper than the alternatives. I have seen images printed with up to 90% of the density in the black that are virtually indistinguishable from images printed from a "normal" color separation by the naked eye, and sometimes if a high enough line screen value is used (+200 LPI) it is hard to tell that a print is a GCR'd image even with a magnifying glass.
So it stands to reason for me at least that if I devote more attention to capturing the "amount" of light with "one CCD eye" completely open, and the "quality" (hue and tint) of the light with my "other three CCD eyes" that are filtering for spectra, I should be able to do the same thing digitally that they have been doing optically in printing for yearsand still yield a superior result.
I'd love to hear a discussion about the best way to use the digital bits in a 32 bit "GCR" digital world by the way. For example, using 10 bits (1024 levels) for luma, 8 bits (256 hues and tints) for green, and 7 bits (128 hues and tints each) for red and blue, or whatever the optimal case could be
Thoughts?
...Open Source isn't the only answer -- but it's almost always a better value than the alternatives...