Is Canon now recording focus distance in EXIF?

[StuartR]

The latest version of Digital Photo Professional software (free with Canon DSLRs) now includes support for some basic "Lens aberration corrections" with certain camera / lens combinations.

The really interesting bit for me is in the help file which states:

When correcting RAW images taken with the EOS-1Ds Mark III, EOS-1D Mark III or EOS 40D, it [the focus distance] is set automatically according to the shooting distance information saved to the image.

This appears to mean (to me at least) that Canon now update an image's EXIF with focusing distance?

So what you ask? Well I'm hoping that software that needs this data to apply lens correction algorithms (DxO being my particular interest) can now automatically set the focus distance rather than require the user to guesstimate it.

If they are recording it in the EXIF then I can't see it (using Opanda) but I guess it could be "hidden" in the "MakerNote" area?

Any EXIF gurus out there?

[Ian]

Re: Is Canon now recording focus distance in EXIF?

The latest version of Digital Photo Professional software (free with Canon DSLRs) now includes support for some basic "Lens aberration corrections" with certain camera / lens combinations.

The really interesting bit for me is in the help file which states:

When correcting RAW images taken with the EOS-1Ds Mark III, EOS-1D Mark III or EOS 40D, it [the focus distance] is set automatically according to the shooting distance information saved to the image.

This appears to mean (to me at least) that Canon now update an image's EXIF with focusing distance?

So what you ask? Well I'm hoping that software that needs this data to apply lens correction algorithms (DxO being my particular interest) can now automatically set the focus distance rather than require the user to guesstimate it.

If they are recording it in the EXIF then I can't see it (using Opanda) but I guess it could be "hidden" in the "MakerNote" area?

Any EXIF gurus out there?

I must admit I had always assumed that the shooting distance was recorded in images from most DSLRs, though maybe in a proprietary format rather than exif.

I'll certainly be doing a bit of delving about this.

Ian

[Ian]

Re: Is Canon now recording focus distance in EXIF?

I must admit I had always assumed that the shooting distance was recorded in images from most DSLRs, though maybe in a proprietary format rather than exif.

I'll certainly be doing a bit of delving about this.

Ian

OK - I'm kind of right and kind of wrong

I'm right in that focus distance is routinely recorded in exif across the board, but wrong in that I thought it was recorded in a proprietary fashion (not exif) - in fact it usually is in the exif data but few exif readers see it.

One exif reader recommended, at least for Nikon users, is exiftool:

http://www.sno.phy.queensu.ca/~phil/exiftool/

I'm sure there are others.

There does seem to be a school of thought that suggests not all Canon lenses generate the distance data and that sometimes the recorded data is not meaningful?

Ian

[StuartR]

Re: Is Canon now recording focus distance in EXIF?

I must admit I had always assumed that the shooting distance was recorded in images from most DSLRs, though maybe in a proprietary format rather than exif.

I'll certainly be doing a bit of delving about this.

Ian

Ian

I recall a thread (with Stephen IIRC) many moons ago discussing conspiracy theories that implied that Canon deliberately withheld this data to stop software like DxO improving the quality of images. The theory being that the "optically corrected" images from mid-range consumer lenses might be as good as that obtained using professional L series optics thereby resulting in a drop in sales!

Now, I don't always believe conspiracy theories but I've always struggled to understand why Canon would leave focus distance out of the EXIF when other manufacturers include it....

[StuartR]

Re: Is Canon now recording focus distance in EXIF?

There does seem to be a school of thought that suggests not all Canon lenses generate the distance data and that sometimes the recorded data is not meaningful?

Ian

Interesting - delving further into the help file seems to broaden the number of camera models and, given the number of lenses it includes, I wonder if your point about the data not being meaningful holds water?

(unfortunately the table formatting in the help file gets lost when posting here)

__________________________________________________ ____________

Correcting Lens Aberration

You can easily make corrections to drops in peripheral illumination, distortions in the subject, and colour blur which occur by the physical characteristics of the lens or slightly remaining aberration.

Please note that with Digital Photo Professional ver. 3.2, you can only correct RAW images which you have taken with the compatible cameras and lenses listed below. JPEG or TIFF images cannot be corrected.

• Compatible Cameras
  EOS-1Ds Mark III, EOS-1D Mark III, EOS-1D Mark II N, EOS-1Ds Mark II, EOS-1D Mark II, EOS-1Ds, EOS-1D, EOS 5D, EOS 40D, EOS 30D, EOS DIGITAL REBEL XTi/400D DIGITAL

• Compatible Lenses
  • EF 14mm f/2.8L USM
  • EF 24-105mm f/4L IS USM
  • EF 14mm f/2.8L II USM*1*3
  • EF 28-70mm f/2.8L USM
  • EF 20mm f/2.8 USM
  • EF 28-135mm f/3.5-5.6 IS USM
  • EF 24mm f/1.4L USM
  • EF 28-200mm f/3.5-5.6
  • EF 28mm f/1.8 USM
  • EF 28-200mm f/3.5-5.6 USM
  • EF 35mm f/1.4L USM
  • EF 28-300mm f/3.5-5.6L IS USM
  • EF 50mm f/1.2L USM*1*4
  • EF 100-400mm f/4.5-5.6L IS USM
  • EF 50mm f/1.4 USM
  • EF-S 60mm f/2.8 Macro USM
  • EF 85mm f/1.2L USM
  • EF-S 10-22mm f/3.5-4.5 USM
  • EF 85mm f/1.2L II USM*2*4
  • EF-S 17-55mm f/2.8 IS USM
  • EF 100mm f/2.8 Macro USM
  • EF-S 18-55mm f/3.5-5.6
  • EF 16-35mm f/2.8L USM
  • EF-S 18-55mm f/3.5-5.6 II
  • EF 16-35mm f/2.8L II USM*1*3
  • EF-S 18-55mm f/3.5-5.6 USM
  • EF 17-35mm f/2.8L USM
  • EF-S 18-55mm f/3.5-5.6 II USM
  • EF 17-40mm f/4L USM
  • EF-S 18-55mm f/3.5-5.6 IS*3
  • EF 24-70mm f/2.8L USM
  • EF-S 55-250mm f/4-5.6 IS*3
• *1 Lens aberration correction functions cannot be used for the image taken with the combination of the lenses*1 and EOS 5D camera, since it is incompatible.
• *2 Lens aberration correction functions cannot be used for the image taken with the combination of the lenses*2 and EOS 5D camera whose firmware version is 1.0.5 or older, since it is incompatible.
• *3 Lens aberration correction functions cannot be used for the image taken with the combination of the lenses*3 and EOS 30D camera, since it is incompatible.
• *4 Lens aberration correction functions cannot be used for the image taken with the combination of the lenses and EOS 30D camera whose firmware version is 1.0.4 or older, since it is incompatible.

[yoshi]

Re: Is Canon now recording focus distance in EXIF?

Hi Stuart,
sounds interesting to me so I looked into it (in rather a nasty way). First off, I have to say, I might misunderstood the point you raised. In that case just excuse me, please.

I think you are talking about the automatic correction of lens aberration, using the focal distance information from exif.

My interest here is rather narrower, as I'm not a user of the three models listed in the latest DPP. So my comment is about the "lens distance" infromation in exif data. (though I'm not a exif guru in any sense)

I opened some RAW files taken recently by 20D using both an older version of DPP and a freeware image viewer Faststone. I'm surprized to see that DPP does not show "focal length" in thumbnail exif data (pic #2), while it ("focal length") has been available in Faststone for quite a long time (pic #1). This applies at least to some of the lenses not listed in the lens list you attached. My guess is that "focal length" must be available for all Canon lenses at least.

The Faststone, however, does not convert the focal length to "35mm equivalent" and this is kept vacant. This conversion can be done however quite easily with the help of a calculator.

To put it short, what I'm saying here is; The focal length data must have been available for long time, as 20D has it. Older DPP just does not show it, although it can be read by at least Faststone and maybe other viewers as well.

I regret again if this is not what you are talking about.


p.s. the left pic = exif obtained from FastStone
the right pic = exif from DPP (ver. 3.0.0.3): sorry it's Japanese but I can assure you there is no focal length there.
The file name shows these two are from the same RAW files. The lens used here is EF20-35mm (an old lens from film days).
yoshi



yoshi

[StuartR]

Re: Is Canon now recording focus distance in EXIF?

Hi Yoshi

I think, as you say, focal length is available (I can see it with Opanda and in DxO) and probably always has been available but what's been missing, until now it appears, is the camera to subject focusing distance.

Looking at the "MakerNote" section (again using Opanda) shows quite a lot of data that doesn't make much sense to the man in the street but must mean something to Canon. I'm assuming that the focusing distance is "hidden" somewhere here (or it's in plain view and I just can't see it! )

[yoshi]

Re: Is Canon now recording focus distance in EXIF?

Hi Stuart
Many thanks for the clarification between focal distance and focal length.
So focal distance is the distance between the lens and the subject.

Then I recall the event that in early 90's Nikon introduced "D" type lens which had a chip and contained the lens focal length data. When Af is fixed then the chip recognizes the "focal distance" (i.e. how far the subject is from the lens). Nikon also introduced a new flash system that effectively used this focal distance information that is given from the lens chip to the body to adust the flash and related exposure set. This was highly appreciated in the market as the subtle flash power adjustment was said to be really effective.

At that time, a pro whom I met at a seminar (known well also for flash skills ) as well as many magazines said;
although it looks a new technology developed by Nikon, it actually isn't. Nikon just gave a new name to the new lens series. That kind information of "focal distance" has been also available to other brands as well - Unlike Nikon they just did not use it for marketing purposes explicitly.

So I simply second your hypothesis that the information must be surely available to Canon system too. If it's not disclosed yet I have to suspect why not yet ...

yoshi