PDA

View Full Version : News: Canon's patent application for a stabilized sensor


Cem_Usakligil
January 28th, 2010, 05:56 AM
In a recently published patent application by Canon, a stabilized image sensor system has been described. As we know, Canon has currently no DSLR in-body stabilization but only stabilized (IS) lenses.

In the patent application, numbered 495,813 (http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-adv.html&r=1&f=G&l=50&d=PG01&S1=20100003025&OS=20100003025&RS=20100003025), the accompanying drawings clearly show an "image stabilization control circuit" which moves the image sensor. This might mean that Canon is considering to introduce in-body image stabilization for future DSLRs.

The patent application also mentions lens stabilization integration, suggesting that the lens stabilization and sensor stabilization mechanisms can interact/integrate for a better and more effective motion correction than it has been possible by using only one of the stabilization methods till now.

As with the K-7-DSLR of Pentax, the image sensor in Canon's patent applications can also be rotated. The K-7 uses this functionality to adjust horizon tilts automatically, but the rotation of the image sensor can also be set manually.

The patent application also describes a system whereby an LCD is used in the viewfinder showing the image recorded by the sensor. This might mean that Canon may introduce a new generation of DSLR cameras which no longer use the mirror mechanism and the prism. This is a currently quite a hot topic on various fora. The future of DSLR: where will the evolution take us to? Food for thought, eh? :-)

Cheers,

Doug Kerr
January 28th, 2010, 06:44 AM
Hi, Cem,

In a recently published patent application by Canon, a stabilized image sensor system has been described.
Thanks for this catch.

Note that US patents and patent applications can be retrieved (by number) in "familiar" form, with all drawings integrated, in PDF format, at this site:

http://www.pat2pdf.org

Note that the "pre-grant publication number" of the application is 20100003025. That is what is needed to identify it at this site. (The display format at the Patent Office site suggests that this is the "application number", but that's not quite accurate.)

(495,813 is the application serial number. The application cannot be retrieved under that number at the pat2pdf site.)

The patent application also describes a system whereby an LCD is used in the viewfinder showing the image recorded by the sensor. This might mean that Canon may introduce a new generation of DSLR cameras which no longer use the mirror mechanism and the prism. This is a currently quite a hot topic on various fora. The future of DSLR: where will the evolution take us to? Food for thought, eh?

I fully expect that the "electronic viewfinder" formulation will become the norm for mainstream high-performance still cameras very soon.

Thanks again for the good catch.

Bart_van_der_Wolf
January 28th, 2010, 06:55 AM
In the patent application, numbered 495,813 (http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-adv.html&r=1&f=G&l=50&d=PG01&S1=20100003025&OS=20100003025&RS=20100003025), the accompanying drawings clearly show an "image stabilization control circuit" which moves the image sensor. This might mean that Canon is considering to introduce in-body image stabilization for future DSLRs.

Hi Cem,

Thanks for that link. I havent looked at the new patents for a while, but assume Canon is still ranking near the top in number of patents worldwide.

But on the other hand, it might just be a defensive tactic to obstruct competitors.

The patent application also mentions lens stabilization integration, suggesting that the lens stabilization and sensor stabilization mechanisms can interact/integrate for a better and more effective motion correction than it has been possible by using only one of the stabilization methods till now.

Covering all bases?

The patent application also describes a system whereby an LCD is used in the viewfinder showing the image recorded by the sensor.

It will be hard to even approach, let alone beat, the resolution of an optical viewfinder, so by itself it would not be a good alternative. I get good visual feedback about the quality of focus with the current setup. On the other hand, with a combined zoomable Rear LCD, who knows what the future brings. Anyway, it also better bring increased battery capacity in a low weight compact design.

This might mean that Canon may introduce a new generation of DSLR cameras which no longer use the mirror mechanism and the prism. This is a currently quite a hot topic on various fora. The future of DSLR: where will the evolution take us to? Food for thought, eh? :-)

I wouldn't put it past their capabilities to introduce an alternative platform alongside the current DSLRs, if only to sell new lenses (even if they stick to the EF series for it).

Indeed, it is fun to speculate.

Cheers,
Bart

Doug Kerr
January 28th, 2010, 09:19 AM
Hi, Bart,

It will be hard to even approach, let alone beat, the resolution of an optical viewfinder, so by itself it would not be a good alternative.

In a sense, the resolution of the optical viewfinder itself is nearly infinite, so indeed we should not crave to beat it, nor use that possibility as a "straw man".

All I know is that, on my EOS 40D, I can invariably do a better job of manual focus with live view (using the "old style" Canon rear monitor, whose feeble resolution I can enhance by "viewing zoom") than with the optical finder. I'd be happier is that were available behind an eyepiece (and in fact I am going to order today a Hoodman attachment to turn the rear monitor into an eyepiece EVF when needed, notably when shooting outdoors).

Certainly for the EVF formulation to be fully competitive (in terms of functionality) with the SLR formulation requires the development of small high-performance display elements, but there do not seem to be any conceptual impediments to that. And of course there are many "system" imperatives to make that workable (different approaches to AF, etc.)

The SLR formulation has waxed and waned over the history of cameras. I keep a 1922 Graflex (SLR), several Speed Graphics (non-SLR), a Yashica Lynx 5000 rangefinder full-frame 35-mm camera (non-SLR), a Canon T70 (SLR), and (until recently) a Sony MVC-FD91 (EVF) in our museum, partly to illuminate that. I'm sure we haven't seen the last swing of that pendulum!

We have to be careful not to confer inherent sainthood on things that we (today) really like. (I do recall when I couldn't imagine why anyone would want a color display on a computer!)

I (a telephone engineer by former trade) even resisted the change of our telephone service here to an IP model (AT&T's U-verse service). It was actually Carla who broke that free (works a treat, actually). Ah, the wisdom of the Cherokee!

Best regards,

Doug

Winston Mitchell
January 28th, 2010, 10:16 AM
Putting resolution issues side, how well can an EVF be made to work in available darkness?

Doug Kerr
January 28th, 2010, 10:47 AM
Hi, Winston,

Putting resolution issues side, how well can an EVF be made to work in available darkness?

Well, simplistically speaking, an EVF will work with any scene luminance that would produce a usable exposure at maximum available aperture for any shutter speed up to some limit based on the system design, but we might think in terms of 1/8 - 1 second (not all present day EVF systems pursue that fully, of course).

Certainly there are occasions where we will be working with scene luminances lower than that with which, if we are dark adapted, we can frame and even focus through an optical finder. And for such work, having only an EVF available could be a real limitation.

Part of the secret is having the EVF operation, in effect, in such a situation, use an "ISO sensitivity" higher than would be used for the shot proper (unless of course the highest ISO sensitivity is to be for the shot itself).

Best regards,

Doug

Bart_van_der_Wolf
January 28th, 2010, 11:04 AM
In a sense, the resolution of the optical viewfinder itself is nearly infinite, so indeed we should not crave to beat it, nor use that possibility as a "straw man".

It can be beaten with zooming-in just like an optical magnifier would, as long as it's based on real data samples.

All I know is that, on my EOS 40D, I can invariably do a better job of manual focus with live view (using the "old style" Canon rear monitor, whose feeble resolution I can enhance by "viewing zoom") than with the optical finder.

What makes a big difference, in my experience, is the quality of the viewfinder's groundglass. For wide aperture lenses and exchangeable groundglass (actually laser etched patterns, often transferred in plastic), the actual plane of focus becomes much clearer to see than with the standard version which must be able to accommodate all types of lenses.


I'd be happier is that were available behind an eyepiece (and in fact I am going to order today a Hoodman attachment to turn the rear monitor into an eyepiece EVF when needed, notably when shooting outdoors).

Yes, such an LCD display loupe helps with accurate focussing, but it's not too useful for action photography.

Certainly for the EVF formulation to be fully competitive (in terms of functionality) with the SLR formulation requires the development of small high-performance display elements, but there do not seem to be any conceptual impediments to that. And of course there are many "system" imperatives to make that workable (different approaches to AF, etc.)

Cost might be an issue, don't know for sure how important that is.

We have to be careful not to confer inherent sainthood on things that we (today) really like.

That's true, never say never.

(I do recall when I couldn't imagine why anyone would want a color display on a computer!)

Well, you're in good company with e.g. Bill Gates who supposedly said:"'640K is all the memory anybody would ever need". Of course he denies having said that:"I've said some stupid things and some wrong things, but not that. No one involved in computers would ever say that a certain amount of memory is enough for all time." and "Meanwhile, I keep bumping into that silly quotation attributed to me that says 640K of memory is enough. There's never a citation; the quotation just floats like a rumor, repeated again and again".

Cheers,
Bart

JohanElzenga
January 29th, 2010, 03:16 PM
The patent application also describes a system whereby an LCD is used in the viewfinder showing the image recorded by the sensor. This might mean that Canon may introduce a new generation of DSLR cameras which no longer use the mirror mechanism and the prism.

Or it may mean that Canon is considering the introduction of a system like micro four thirds and/or Samsung NX10 in addition to its current line of DSLR cameras... I'd say that is far more likely.

Asher Kelman
January 30th, 2010, 04:08 PM
Or it may mean that Canon is considering the introduction of a system like micro four thirds and/or Samsung NX10 in addition to its current line of DSLR cameras... I'd say that is far more likely.

Johann,

There are two potential new markets. First the "rangefinder market without the rangefinder and next the
larger sensor as in the Leica S2.

Both these areas would open new, legitimate markets of Canon lenses and bodies.

Imagine compact cameras with either full frame 35mm or 60mm sensors and no mirrors!

Asher

JohanElzenga
February 1st, 2010, 01:03 PM
True, but I doubt that Canon will be very interested in a new large sensor system. Canon needs to make a profit, just like any other company. And the profits are in the mass markets, not in the super specialized markets. That's why I think that an APS-C based system is more likely than a 60mm system.