Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
11-21-2010, 07:39 AM   #1
Senior Member




Join Date: May 2010
Posts: 113
Very technical question about in-camera stabilization

so I am trying to do some math using the images taken with my K-x and I really need to know exactly where the lens's optical axis is situated on the sensor (or rather, on the image of course).

and I require absolute precise information, I need to know exact pixel position of the axis.

I realize that cameras are not built with such extreme tolerances so the optical axis is probably not at the sensor center, but I think I know the way to find this center to precision I need, BUT ONLY IF IT IS CONSTANT.

however the Pentax in-camera stabilization system works exactly by shifting sensor about the lens, correct?

this means that optical center should be different for every image as long as SR was switched ON while taking it.

it's still ok though, since there is an option to turn the stabilization off.

however here we got the catch. Does the sensor really "freezes in place" when I turn off the stabilization, or it still floats freely when I shake the camera, although not trying to move intentionally while shaking?

if it still floats, then this is a deal breaker for me...

judging by the ever audible in-camera movement when shaking camera, I am afraid it's not fixed even when SR is off

does anybody know for sure?

11-21-2010, 08:44 AM   #2
Senior Member




Join Date: Aug 2010
Location: Nottingham
Posts: 172
I don't know, but common sense says that in a 1 second exposure, 1 pixel of movement would lead to 1 pixel of blur. We don't see that if the camera is mounted on a tripod with a timed release. Can you use a tripod for your maths pictures?
11-21-2010, 08:48 AM   #3
Senior Member




Join Date: May 2010
Posts: 113
Original Poster
I can but I do not know in which position the sensor is remaining when camera is not moving, in case that it's floating.
11-21-2010, 08:59 AM   #4
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,157
The sensor is held in the central position with SR off.

11-21-2010, 11:55 AM   #5
Inactive Account




Join Date: May 2010
Location: Roma
Posts: 4
hi, the sensor moves when SR is enabled (for example k7 SR is able to move the cmos up to 1,5 mm from the axis), but when SR is turned off the only force that keep the sensor in his central position is a magnetic field. If you use the camera on tripod is almost impossible to move the sensor during an exposure with SR turned off, but not impossible.

you just need a very strong and fast force, like a shock, that i hope it is not a very frequent stress for a camera

Last edited by isd88; 11-21-2010 at 12:03 PM.
11-21-2010, 01:53 PM   #6
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
We need to know more I think - what are your real stability needs?

The camera is designed to put an image on the sensor with at best one pixel precision during an exposure; SR exists to ensure this even if the camera jiiggles a bit. So optic axis stablity should be ok within an exposure.

Are you worried that there might be a few pixel drift between exposures?

I suppose there might be a camera orientation effect - the sensor rest position visa vis the optic axis might change if the camera is held up-side down. But this should not be hard to detect experimentally. If it is a problem, don't invert the camera or use a Canon or Nikon maybe?

Dave
11-21-2010, 07:49 PM   #7
Pentaxian
Class A's Avatar

Join Date: Aug 2008
Location: Wellington, New Zealand
Posts: 9,193
QuoteOriginally posted by olenl Quote
Does the sensor really "freezes in place" when I turn off the stabilization, or it still floats freely when I shake the camera, although not trying to move intentionally while shaking?
According to my understanding the sensor will try to stay in place but since it isn't tightly coupled to the camera, it will move against the camera frame slightly. It depends on the control loop parameters how tightly the coupling will be.

You may want to read about Falk Lumo's findings regarding shutter-induced blur by the K-7's. He speculates about sensor position control loop parameters.

Last edited by Class A; 11-21-2010 at 07:55 PM.
11-21-2010, 11:46 PM   #8
Veteran Member
Manel Brand's Avatar

Join Date: Sep 2008
Location: Porto
Photos: Gallery | Albums
Posts: 854
QuoteOriginally posted by olenl Quote
and I require absolute precise information, I need to know exact pixel position of the axis.
Now, I'm curious about that. Why do you need such a degree of precision?

11-22-2010, 02:33 AM   #9
Senior Member




Join Date: May 2010
Posts: 113
Original Poster
QuoteOriginally posted by Manel Brand Quote
Now, I'm curious about that. Why do you need such a degree of precision?
I want to map lens distance-from-center to angle function precisely so i can measure angles between objects registered on the sensor through mapped lens...

Clearly I need reliable distance to center measured for each pixel for that.
11-22-2010, 02:39 AM   #10
Senior Member




Join Date: May 2010
Posts: 113
Original Poster
QuoteOriginally posted by newarts Quote
We need to know more I think - what are your real stability needs?

The camera is designed to put an image on the sensor with at best one pixel precision during an exposure; SR exists to ensure this even if the camera jiiggles a bit. So optic axis stablity should be ok within an exposure.

Are you worried that there might be a few pixel drift between exposures?
yes, that's what I am asking about.
also I turn off SR to avoid "jiggling", and I wonder if that's enough to fix the sensor .

Some other replies keep me optimistic though.
QuoteQuote:
I suppose there might be a camera orientation effect - the sensor rest position visa vis the optic axis might change if the camera is held up-side down. But this should not be hard to detect experimentally. If it is a problem, don't invert the camera or use a Canon or Nikon maybe?

Dave
actually it's kind of hard, the only way I see is attaching something directly to the lens but that would mean I need a lens that can focus to its length which I don't have.
11-22-2010, 04:02 AM   #11
Veteran Member
Manel Brand's Avatar

Join Date: Sep 2008
Location: Porto
Photos: Gallery | Albums
Posts: 854
QuoteOriginally posted by olenl Quote
I want to map lens distance-from-center to angle function precisely so i can measure angles between objects registered on the sensor through mapped lens...

Clearly I need reliable distance to center measured for each pixel for that.
Thank you. Ok, but why do you need all those mesurements, in other words, what's the objective of all that procedure? What are you trying to achieve?
11-22-2010, 04:28 AM   #12
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
QuoteQuote:
I suppose there might be a camera orientation effect - the sensor rest position visa vis the optic axis might change if the camera is held up-side down. But this should not be hard to detect experimentally. If it is a problem, don't invert the camera or use a Canon or Nikon maybe?
You are right & the set-up may not be trivial.

One test would be to secure the camera to a long(1meter?), stiff, U shaped aluminum channel using the camera's tripod mount. Also secure a complex to the channel on the optic axis & focus on the target. Take photos of the target with the camera in various orientations and compare the results.

Dave

PS Any mechanical softness of the lens mount will confuse the result
11-22-2010, 05:15 AM   #13
Pentaxian
Wheatfield's Avatar

Join Date: Apr 2008
Location: The wheatfields of Canada
Posts: 10,157
QuoteOriginally posted by olenl Quote
yes, that's what I am asking about.
also I turn off SR to avoid "jiggling", and I wonder if that's enough to fix the sensor .

Some other replies keep me optimistic though.

actually it's kind of hard, the only way I see is attaching something directly to the lens but that would mean I need a lens that can focus to its length which I don't have.
Dude, if the sensor wasn't fixed in position during exposure (SR off), how the heck do you think it's going to make a sharp picture and have the much vaunted (near) 100% viewfinder accuracy?
Just curious if you can answer this..
11-22-2010, 05:27 AM   #14
Inactive Account




Join Date: Dec 2008
Location: Ames, Iowa, USA
Photos: Albums
Posts: 2,965
QuoteQuote:
so i can measure angles between objects registered on the sensor through mapped lens...

Clearly I need reliable distance to center measured for each pixel for that.
You want to map image plane distances to object space distances and convert those to the angle between the objects with the camera as the angle's vertex right? I think maybe knowing the precise center of the sensor might not matter much for the following reasons.

I think a small shift of the center of the sensor perpendicular to the optic axis must have an extremely small effect on the relation between distances in the image and object planes. If this were not the case, there would be obvious distortion of an image from the center to the edge of the frame.

The lenses we use are designed to be rectilinear - a spacing of Y in the object plane is mapped into a particular distance Z in the image plane (Y times the optical magnification) and this image distance is constant across the image plane, ie, shifting the sensor's center a bit one way or the other has no effect on the ratio M between Y and Z.

Z/Y = M

There is a well established relationship between distance from the image plane to object plane, D, focal length, F, and magnification, M.

D=F(1+M)^2/M

This can be solved for M, ie, Z and Y.

Isn't that enough? Can you give us an idea of the distances, angles, & focal lengths involved so some quantitative error estimates are practical?

Dave

Last edited by newarts; 11-22-2010 at 06:23 AM.
11-22-2010, 05:58 AM   #15
Senior Member




Join Date: May 2010
Posts: 113
Original Poster
QuoteOriginally posted by Wheatfield Quote
Dude, if the sensor wasn't fixed in position during exposure (SR off), how the heck do you think it's going to make a sharp picture and have the much vaunted (near) 100% viewfinder accuracy?
Just curious if you can answer this..
it surely does not move when the camera is held stable to take a picture, but when camera moves, in case of not solidly fixed sensor, it could float and stop in a slightly different position due to inertia so next image would be taken with slightly different position of lens optical axis, so direction of particular pixels would change a little.

that is of course if sensor is not re-centered by magnetic field or something like that, which was the point of original question.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
axis, camera, center, dslr, image, in-camera, photography, sensor, sr, stabilization
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Decentering - Technical Question Sailor Pentax SLR Lens Discussion 1 04-24-2009 04:32 PM
Is your in camera stabilization as good as a chicken? vitalsax Pentax DSLR Discussion 18 01-10-2009 03:40 PM
Camera Stabilization aamir515 Photographic Technique 10 11-20-2008 02:28 PM
Technical question re: PSE 5.0 back up and restore Ed in GA Digital Processing, Software, and Printing 1 11-14-2007 09:14 AM
Technical Question AmdamJayne Post Your Photos! 7 05-16-2007 12:37 AM



All times are GMT -7. The time now is 10:19 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top