Lens Calibration: A complete guide how to calibrate a lens.

[Amamba]

Sorry if a silly question, but is the lens profile camera dependent ?

Is there any value in having people share their calibrated lens profiles on this board ?

I have a few lenses that I may end up trying to calibrate, and wouldn't mind having access to some calibration data for the lenses I no longer own.

[FotoLars]

Sorry if a silly question, but is the lens profile camera dependent ?

Is there any value in having people share their calibrated lens profiles on this board ?

I have a few lenses that I may end up trying to calibrate, and wouldn't mind having access to some calibration data for the lenses I no longer own.

Hi!

In practical terms, the lens calibration should work equally well for every camera which the lens can be connected to. Third party lenses from Sigma, Tamron, Tokina, etc which are used both with eg. Canon and Nikon houses should display more os less identical optical properties on either one.

What is important to remember is the size of the sensor: If a lens has been calibrated for a full frame camera, where the sensor multiplier is 1.0, then the calibration can be used for a cropped sensor too, that is, a sensor which has a multiplier value of 1.5, 1.6 or whatever the crop factor might be (it depends on the camera in question). If, however, the lens has been calibrated for a cropped sensor, then this calibration will not be useful for a full frame sensor (camera).


Cheers,

Lars Nilsson,
Lund, Sweden

[marbly]

What is important to remember is the size of the sensor: If a lens has been calibrated for a full frame camera, where the sensor multiplier is 1.0, then the calibration can be used for a cropped sensor too, that is, a sensor which has a multiplier value of 1.5, 1.6 or whatever the crop factor might be (it depends on the camera in question). If, however, the lens has been calibrated for a cropped sensor, then this calibration will not be useful for a full frame sensor (camera).

Thanks for that info, Lars. I didn't know that a lens profile for a 1.0x sensor could be used with a 1.5x or 1.6x sensor. And 1.5x profiles and 1.6x profiles can be used interchangeably? Or, are you saying that a lens profile should only be used for the sensor size it was calibrated with or smaller? In other words, a 1.6x lens profile should not be used for a 1.5x camera?

[FotoLars]

What is important to remember is the size of the sensor: If a lens has been calibrated for a full frame camera, where the sensor multiplier is 1.0, then the calibration can be used for a cropped sensor too, that is, a sensor which has a multiplier value of 1.5, 1.6 or whatever the crop factor might be (it depends on the camera in question). If, however, the lens has been calibrated for a cropped sensor, then this calibration will not be useful for a full frame sensor (camera).

Thanks for that info, Lars. I didn't know that a lens profile for a 1.0x sensor could be used with a 1.5x or 1.6x sensor. And 1.5x profiles and 1.6x profiles can be used interchangeably? Or, are you saying that a lens profile should only be used for the sensor size it was calibrated with or smaller? In other words, a 1.6x lens profile should not be used for a 1.5x camera?

Hi,

Going from larger sensors to smaller sensors will work, but not the opposite direction.

Now, in practical terms, the difference would not very big if one would use a calibration made for a 1.6 crop factor sensor on a 1.5 crop factor sensor. The widest angles will be missing, but the result might still be good enough - or better than with no correction applied at all; it will greatly depend on the lens and its distortion properties. It will be easy to test, though. Change the multiplier value for a given 1.6 calibration to 1.5 for a lens combination you have used, possibly poke around and fake (add) some wider value(s), apply it on a photo taken by an 1.5 crop factor sensor and see how good, or bad, the resulting image looks.

Jim at Bibble Labs explained a couple of years ago the benefits of using full frame cameras for doing calibrations: The results will be useable on all smaller sensors, meaning that people with the same lenses but cameras with cropped sensors don't have to worry about making a calibration for their combination; please have a look at his posting at http://support.bibblelabs.com/forums/vi ... 94&t=15867.

The best is, of course, to either have a calibration made for a full frame sensor or, if such a calibration doesn't exist, a calibration for the very sensor size you have. With the help of the guide, g3data and afx's scripts, it will be an easy thing to make a new calibration - if you still own (or have access to) the lenses in question, that is !


Cheers,

Lars Nilsson
Lund, Sweden

[marbly]

Lars, thanks for that additional info.

[greerd]

Great thread, thanks to all who contributed. To bad this thread wasn't a sticky, but I guess we would end up with to many. What about sub-forums!!!
Anyway I plan on giving this a whirl with my new 25/1.4 Panasonic just as soon as I find a good target. So bump...

[SaintGermain]

As I, and others, have said regarding lens detection in Hugin and its properties, please see http://forum.corel.com/EN/viewtopic.php?f=90&t=45656, I would not trust the values from Hugin very much - but that can always be checked against a photo; if the parameters are OK, the optical distortions will be reduced to more or less no distortion at all.

I have seen, when testing some given calibrations in the B5/ASP files, for example for my little Canon S95 pocket toy camera, that the values, calculated by someone I have no clue of, for this particular camera are not very accurate for the circumstances (eg. distance) I compared against in a real-world photo - but even with the not fully accurate vaules, the result is rather good compared with the uncalibrated photo. I tested at a few different focal lengths and distances, and there are noticable differences.

Hum maybe it was me who posted the calibration results for the S95 ?
I posted them on forum.hardware.fr (http://forum.hardware.fr/hfr/Photonumer ... m#t4196080) and dpreview.
I used the lensfun tutorial (http://lensfun.berlios.de/lens-calibration/) but I was not quite satisfied with them. In particular I used a tilted camera on a building (not so good apparently).

I would like to find the good parameters and do the work, but as I am not really an expert, I would need a little guidance.
Interested to help me ?

Regards,

[FotoLars]

Hi SaintGermain,

As I am currently travelling around in Germany, I don't have access to my little pocket friend Canon S95, but once back in Sweden again, I will make a calibration series for it and upload the results - in a couple of weeks or so...

If you want to try for your self, the hardest part is really to find a proper calibration target - the rest is very simple, and doesn't take very long time.
As you seem to be running Linux in your box, it will be even easier to get started.

Download the example photo I published earlier in this thread, prepare the software required (see the top part of the guide), practise the clicking and parameter extraction part with the scripts afx prepared - and you will see that the whole thing is rather easy. Then, once you have found a good wall, etc. and there have taken a number of photos ranging the entire zoom span, with less than 15% increase in focal length between each exposure, you will know how to correctly and quickly perform the real calibration for each focal length and how to merge these data into a proper formatted file for use in ASP.

Good luck, if you decide to try !



Cheers,

Lars Nilsson
Lund, Sweden
Presently in Dortmund, Germany

[SaintGermain]

Well given my first experience on the subject, I think it would be much better if you do it yourself (I'm really a newby in photography...).
I can however integrate the lens parameters in lensfun, that way all other software relying on it will benefit.

I'm really eager to see your results !

Best regards,

[SaintGermain]

Hello Lars,

Any news on this famous S95 lens calibration parameters ?
I'm very eager to see the results !

[nyukuri]

Hi,
I have the Nikon P7100, which is not yet in the database for lens calibration. But the Nikon P7000 is in the database. Since both have the same sensor and same lens, I thought I can simply select the P7000 from the database and apply on my images of the P7100, unfortunately this does NOT work! (Corel, are you listening???)
Now, is there a possibility to take the data of the P7000 and to create with it a profile for the P7100? I guess this should be somehow possible, I just don't know how. Can somebody help me, or a tech geek have a look into it????
THANKS IN ADVANCE!
Greetz from Africa,
nyukuri

[afx]

But the Nikon P7000 is in the database.

Nope. Just a placeholder entry. Easily identifiable by the
calibrated: false
entry.

cheers
afx

[claus_uetersen]

Is it possible to convert lightroom lens data to ASP?
There is a calibration download program which gives an readable txt-file.
Claus

[cookiebob]

Is it possible to convert lightroom lens data to ASP?
There is a calibration download program which gives an readable txt-file.
Claus

That would be awesome if s.o. knows a work-around

[df]

So, if you figured out a lens calibration for a 1.3 crop camera you'd just go into the calibration file and change it from 1.0 uncalibrated to 1.3 calibrated and when you used it on a 1.5 or 1.6 crop camera it would just work? Or do you need to do something else for it to work?

edit: assuming that you'd also plug in the numbers in the txt file for the various focal lengths as well.