Reading mode: Light Dark

The end of Bayer?

Go to last post
I found the information regarding Canon's patent application for a new image sensor type, but it's quite obvious from reading through it that it's probably several years away from practical application.

So this isn't really a foveon-style sensor, it's more of a way to do the bayer filter electronically. It's like an electronic version of the old tri-color filter method that's been around for well over 100 years. First the RED exposure, then the BLUE, then the GREEN, then combine them together.

The first big problem is the fact that the image sensor data must be read after each filter color change. That would impose some severe restrictions on the system, not the least of which is the fact that the shutter must remain open from the start of the first color to the end of the third.

It sounds interesting, but it's still a long way from practical for a professional DSLR camera.
 
Mauka said:
To think that single photosite RGB is coming, and AA filtering
nearing an end, is truly enticing.
Click to expand...
It seems to be a very small minority of people who think that only bayer-CFA captures need AA filters. Concentric-RGB pixels benefit from them too. Without them, spatial detail is misplaced, because multiple points of light are snapped to the same grid point. A narrow point of light drifting across a greyscale sensor that is properly AA-filtered will always have changes to the image, even for tiny sub-pixel movements. Unfiltered, the point will illuminate a single pixel uniformly, and then quickly jump to the next pixel (without microlenses, it will even black out between pixels).

In a static image, everything is displaced without an AA filter (unless the focused image isn't even sharp to begin with).

--
John
 
Mauka said:
With Canon's new announcements, and the 1DM2N being more of a
running update, I can't help getting a bit excited that we may have
seen the last high end Canon anouncments that will house Bayer
technology.

To think that single photosite RGB is coming,
Click to expand...
Single photosite RGB will continue to offer no improvement in the quality of images, because humans too have separate RGB sensors.

--
David J. Littleboy
Tokyo, Japan
 
Mauka said:
With Canon's new announcements, and the 1DM2N being more of a
running update, I can't help getting a bit excited that we may have
seen the last high end Canon anouncments that will house Bayer
technology.

To think that single photosite RGB is coming, and AA filtering
nearing an end, is truly enticing.
Click to expand...
AA filtering is a mathematical requirement for correct digital imaging, regardless of the type of sensels used.

As the MF back users are finding out, you can't leave off the AA filter and shoot fabrics.

--
David J. Littleboy
Tokyo, Japan
 
WSLam said:
Everyone knows the inherent problems with the Bayer approach, but
looking at today's cameras, Bayer designs continue to put out VERY
high quality images.
Click to expand...
Exactly.
WSLam said:
I too hope Bayer will come to an end one day.
Click to expand...
Not me. If you have to measure all three colors at every point, you need three times the data for essentially no improvement in image quality.

Even worse, you need to store three times as much charge at each sensel site, increasing noise radically compared to using that whole charge storage capacity (well depth) for one color.

Foveon is a horrendeously bad idea.

--
David J. Littleboy
Tokyo, Japan
 
I used to have a Kodak 330 and never used it with the AA filter on, the picture quality really went downhill when it was attached. Photoshop could easily get rid of 99% of the moire. I'm also pretty sure the Foveon cameras don't need an AA filter in them, though I'm not 100% sure about it - the Sigma people in here may know for sure (they may use some sort of hot-mirror type filter?)
 
Rich said:
I used to have a Kodak 330 and never used it with the AA filter on,
the picture quality really went downhill when it was attached.
Click to expand...
A badly implemented AA filter may be worse than no AA filter as a practical issue, of course.
Member said:
Photoshop could easily get rid of 99% of the moire. I'm also pretty
sure the Foveon cameras don't need an AA filter in them, though I'm
not 100% sure about it - the Sigma people in here may know for sure
(they may use some sort of hot-mirror type filter?)
Click to expand...
(The Stgma cameras don't have an AA filter.)

The mathematics of discrete sampling (which is what digital imaging is) is very explicit in concluding that low-pass filtering is required for correct imaging. Any information over the Nyquist frequency shows up as "aliases" in the pass band that cannot be differentiated from real data.

Note the correct in there.

As a practical matter, you can decide that you don't mind having to redraw the sections of every photo that get messed up by hand, and that you prefer everything being at the wrong point due to the "snap to grid" effect, but at some point, just doing it right makes more sense.

--
David J. Littleboy
Tokyo, Japan
 
Thanks
 
The low pass filter does not "scatter the color information from each pixel". The AA filter slightly blurs the light in order to reduce aliasing, which is a problem even on monochrome sensors due to the grid-like layout of pixels on the sensor.

The bayer filter restricts the wavelength of light hitting each sensor. In reality, the sensor is completely color blind, it only reads luminace levels. But the luminance levels will be different in each wavelength (Red, Green & Blue). The A/D converter, which is aware of the layout of the Bayer filter, does the blending of the various readings to determine the "color" for each pixel.

Then there is an IR cut-off filter to block IR from hitting the sensor and affecting the luminance readings.

At no point is the color information being scattered, because the sensor itself isn't color-aware.

klinux
marrio said:
Nope. They do have to do with each other. The low pass also serves
to scatter the color information from each pixel in order to read
that information in all channels. You can a complete info o Lens
Work III.

Márcio
Click to expand...
 
Mauka said:
Leon Wittwer said:
The problem with a new sensor technology is that it has to arrive
fully competitive in cost and performance with existing sensors
that already do the job to most peoples satisfaction. In fact, it
likely has to be significantly better otherwise no one will buy it
as a replacement for the old system.
Click to expand...
Fair enough, but Canon is a forward thinking company. They are
quite aware of how far they can push Bayer, and have been for some
time. They also know it's shortcomings (they get an earful on
these boards). They are also aware that the megapixel race is hard
stop once started.
Click to expand...
There are also practical limitations here that will at least delay things for some time. Cameras are not just sensors. They also need lenses, AF system, etc. A good camera is a balanced piece of equipment engineering wise. Currently, Canon's AF system resolves slightly better than the current sensors. If you do a significant increase in pixels, then you need a new AF system. It is not clear that this would be a very easy task and still retain current AF capabilities. Also, lenses would then become the limiting factor. There would need to be a new set of lenses to outresolve the sensor. This would be very expensive for Canon as well as for the photographer. It is the lens aspect that probably will limit how many pixels we will see.
Mauka said:
Knowing that they will abandon Bayer, it is a
matter of when.
Click to expand...
I don't think there is any assurance that Bayer will be abandoned any time soon.
Mauka said:
The problem you have posed is real, but smart
companies deal with such problems by tandem development teams that
share info. To think that canon will take all of it's engineers
off of Bayer products and set them on this new track would be
unlikely. When the new is competitive, then the time has come.
Click to expand...
We will see.

--
Leon
http://homepage.mac.com/leonwittwer/landscapes.htm
 
Rich said:
I used to have a Kodak 330 and never used it with the AA filter on,
the picture quality really went downhill when it was attached.
Photoshop could easily get rid of 99% of the moire. I'm also pretty
sure the Foveon cameras don't need an AA filter in them, though I'm
not 100% sure about it - the Sigma people in here may know for sure
(they may use some sort of hot-mirror type filter?)
Click to expand...
How is this demographic called "Sigma People" derived? Isn't the sample self-filtering? People who like the aliased look are drawn to Sigma. People who don't are drawn away.

Personally, I think that you have to have something wrong with your perceptual mechanisms to think that an aliased SD9 image taken with sharp optics is a realistic capture.

--
John
 
Peek at the betterlight LF backs without AA filtering. To avoid the onslaught, I am aware that they are scanning backs, and of all of the many significant shortcomings, but all the same it is intersting to see RGB and no AA in action.
 
Hi Leon,

We agree on one thing. Canon will need new lenses (at least some-they already do).
Leon Wittwer said:
This would be very expensive for Canon as well as for the
photographer.
Click to expand...
No doubt, but having Sony do it first would be more expensive.
 
pdcsq said:
But what low-pass filtering has to do with Bayer interpolation ?

To me these things are completely independent. One is to prevent
aliasing, the other is to overcome the effect that current sensors
are in fact color blind.
Click to expand...
You have already got some answers - and here you got one more.

A Foveon sensor (and also all other sensors that sample all colors at all sites) can get away without an anti alias filter - a Bayer (and all other CFA sensors) cannot.

A Bayer sensor without an anti alias filter have problems with colored aliasing if the lens is too sharp. It looks awful.

The SD10 on the outher hand has only micro lenses. The micro lenses create a box filter that removes the higher frequencies. It does not filter away the frequencies at nyquist frequency. But - it does not really matter as the pictures look good nevertheless. And that is what counts.

Roland
 
David J. Littleboy said:
Single photosite RGB will continue to offer no improvement in the
quality of images, because humans too have separate RGB sensors.
Click to expand...
This is not really so. A technically better way of recording images is better even if the human sight has its faults.

Some years ago (100 I think) some photgraphers concluded that all sharp images was not neccessary because humans see only a small part sharp. So - they used lenses that was very badly corrected - and only the central part was sharp. Of course - it was the emperors new clothes - everyone could see that the pictures looked more artistic than photographic. So could of course those photographers - but they insisted nevertheless.

Roland
 
That is today. In the future we will need a 66Mp bayer sensor to compete with the 22Mp Foveon type. Maybe the 22Mp will be easier to make.

Márcio
 
As I understant "blur" is equal to "scatter". It blurs or scatters the light for each light information be read by a pixel sensitive to (bayer pattern) each color. The low-pass is not needed for a Foveon type sensor, Sigma cameras doesn't have them.

Márcio
 
Just adding that there is aliasing artifacts that maybe could be corrected in software.

Márcio
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top