Canon still pursuing Foveon-style multi-layer sensor design


Canon has patented a color-sensitive multi-layered sensor design, showing the company is still pursuing the technology. Like Sigma’s Foveon chips, the multi-layered design allows each of the sensor’s pixels to capture color information without the need for colored filters. The patent, discovered by the Japanese Engineering Accomplishment blog, suggests a system to promote resonance within the sensor, in an attempt to make the lower layers of the sensor more sensitive.

The Canon patent includes a structure (40) designed to induce resonance within the sensor, in an attempt to boost sensitivity to red light.

Canon already uses a two-layer sensor in the iFCL metering system introduced with the EOS 7D, to make it color aware. At present only Sigma, with its Foveon technology, uses a multi-layer design as its main imaging sensor. The principle is that different colors of light have different energies, allowing them to penetrate to different depths within a sensor. Existing designs have not been able to offer the same degree of light sensitivity as more conventional, filtered sensors – at least partly because the red can be lost in the sensor, rather than being recorded.

Canon’s iFCL metering sensor (as first used in the EOS 7D), uses a two-layer design to provide an understanding of the color, as well as the brightness, of a scene.

Although the fine detail is not clear (a combination of being written in repetitive ‘patentese’ and in Japanese means it doesn’t lend itself to precise machine translation), it seems Canon’s design uses a physical structure that causes light to resonate within the sensor, increasingly the likelihood of the red light being captured.

Canon is not alone in working on layered sensors – Sony has also published several patents in the area, hoping to avoid the risk of color moire and loss of color resolution that exist in the conventional Bayer design. (from Egami blog)



Well Im glad they try. Tho for now I would be happier if they maybe tried to at least match competition. I hope they try to do that too, cause as far as sensors go, Canon is quite a bit “meh” right now.


I think Canon would be better served by getting their existing sensors up to the Sony level of quality (I am a Canon user so this is not a fan club thing). On the other hand, just because Sigma can’t make a less noisy sensor doesn’t mean that Canon can’t. Even on my little DP1, if you keep the ISO low, the detail captured is amazing.


what makes you think these are competing developments?


It’s good to see real sensor innovation, rather than the fake filter innovation passed off as a different sensor (I’m look at you Fujifilm).


Sensor design innovation are “real innovations” if new technologies (X-Trans, Foveon) improve performance in some way. And this is true even if laypeople don’t happen to think so.

Gary Dean Mercer Clark

Sony and Canon had the opportunity to buy Foveon when it was for sale. Foveon had spent millions targeting its designs for use in cell phones. Toshiba was one of the cell phone manufacturers interested in this sensor for it’s phones but at the last moment pulled out and went with I believe Sony for its camera phones. At the time, Foveon was desperate for more investors seeking financial backing, promising huge returns at the time, discussing its development of the cell phone camera sensor for toshiba. Toshiba went with Sony cell phone sensors instead & Sigma bought Foveon. The rest is history. The current 46 MP (28-30MP bayer) sensor miniaturization effort foveon aimed towards cell phone market has been incorporated into the current foveon APC sized sensor found in the DPM compact and SD1M DSLRs. The image quality and size rivals full frame cameras like the Nikon D800 which amazes me. If you can stand the humor watch this video on


trust me…i love SD1M but only at ISO 400 and under….it’s practically not anything good if you need speed. i do own couple of sigma glasses and i love to but when it comes to providing services to clients…..SD1M is not considered by me.


They didnt buy it cause they want different approach. Sigma clearly showed that it doesnt work as intended.

Unless you can live with base ISO camera. Which I can. Just 99% of photographers cant.


Nice to see that Canon is doing something serious about sensors.
Anything that makes better use of available light is warmly welcome.


Interesting if Canon can get it to work.


No thanks. I don’t need this.

M Jesper

If they release this i would forgive them for using that same 18MP sensor in everything since the beginning of time. And it would explain why it took so darn long …


A 4th generation 18MP sensor is the same as the 1st generation is it? Research before you type in the future…

M Jesper

So they tweaked it 4 times, show me how much of an improvement that has been i dare you. And don’t try to pass off jpeg processing as sensor performance. Research before you type in the future huh… (ps. i wasn’t picking a fight, the opposite in fact as this is GOOD news, but congratulations on turning it around.)


Yes the 4th gene 18MP sensor is pretty much the same as the 1st gen. There has not been any noticeable change or improvement in that sensor


Canon is definitely workign on bringing their sensor tech up to speed. They are suffering from having such a large 0.5um process whereas Sony is on 0.18um I beleive and others are 0.25um. I expect the 7D II sensor to be entirely new and the precursor of a high MP FF sensor. But so far as has been said, they have been tinkering at the edges. However, the 1D X sensor does show some real improvements and while it can’t match Sony for low ISO DR, it’s high ISO performance is astounding and it’s shadows can be pushed further than 5D III and it has no banding.


canon should just buy sigma (or at least foveon, sigma ideonst making money with it) and thats that..


Too bad, not quite that simple, I would guess Sigma owns the sensor IP but not absolutely positive. Originally, and still is, produced by National Semiconductor which is now part of Texas Instruments. Good idea but many players at the table.

Joseph S Wisniewski

Foveon stopped having National fab their sensors and switched to Dongbu Electronics way back in 2006. (There was all sorts of interesting talk at the time, including allegations of arson and organized crime involvement).

Intel, National, and Dongbu were just people Foveon had fab chips. They don’t own the IP any more than print shop owns the rights to the books they print.


With so much pixel resolution, the question remains if the color layer technology has any future. That said utilizing resonance to detect a color seems superior to foven models as it should have better sensitivity.


Will Canon be paying Sigma for licensing the idea?

R Butler

I doubt it. Because:

1) You can’t patent ideas, only practical applications of ideas
2) Because this is about something Sigma’s sensors don’t do – hence the attempt to patent it.

Joseph S Wisniewski

There was a lot of layered sensor art before Foveon. Sigma’s patents are all about the diode structure they use in an effort to lower costs, and the use of the penetration depth vs. wavelength of photons in silicon to do the color separation.

Neither of these things turned out that well. A company with more resources, interested in layered sensors, would do something different.


Yes please please please… 🙂

I could start using all my nice Canon lenses again.
For now I’m 98% Ricoh GXR with “M Mount”…


How about calling this technology “X3”, which is the jargon adopted as an industry-standard by CIPA? “Foveon” is just one approach to X3, and not the one that Canon is pursuing, and the “X3” tag is also well-known.


Or, “RS”, as in EXMOR-RS? 😀

R Butler

That’s why we’ve said ‘Foveon-style’ – we think it’ll be more broadly recognised than ‘X3.’ We also stick to ‘multi-layer’ because Canon’s existing multi-layered sensor is effectively an X2 design.

@BLJ – Exmor RS is a totally different technology – it’s a conventional Bayer sensor with the light sensitive regions stacked on top of the circuitry – this is a sensor that perceives colour based on the depth to which light penetrates.

Joseph S Wisniewski

I wouldn’t call it “Foveon style”.

Like I mentioned earlier, there was a lot of layered sensor art before Foveon. Sigma’s patents are all about the diode structure they use in an effort to lower costs, and the use of the penetration depth vs. wavelength of photons in silicon to do the color separation. Those two things make a sensor “Foveon style”. Do it with Fuji’s interlayer organics, Nikon’s prisms, Canon’s dipoles, etc. and you’ve got something else.

Joseph S Wisniewski

Even Foveon wasn’t satisfied with X3. There’s a paper by Dick Merril that calculates six layers as being optimal.

At the rate processing power and memory s growing, I can’t see this RGB stuff lingering around too much longer in either filter array or layered sensors. Heck, if Sony hadn’t poisoned the well by making suck a botch of RGBE, four colors would probably be the standard by now. Fuji was propagating it across their film line before the film industry imploded and all the R&D money dried up.


Sony and now Canon appear to be pursuing this ‘multiple’ layered approach to sensor design. And every report that talks about it is always saying XXX is working on a Foveon-like sensor, etc. The thing I wonder is how patentable could it be if the concept for the sensor, despite whatever differences it might have is essentially the same ‘in principle’ as a Foveon. While I’m not sure whether Bayer arrays are any infringement from one designer to the other I would think anyone who attempts this approach would be breeching Foveon’s patents similar to Polaroid success suits against Kodak when they came out with their own ‘instant’ process.

Joseph S Wisniewski

If you read Foveon’s patents, you’ll see that Foveon owns no patents on the “concept for” a layered sensor. Make a sensor with an alternating PNPN junction structure and Sigma will sue the crap out of you. Make a sensor with three organically isolated PN junctions and Fuji will go after you.

Source Article from