2

I'm trying to calibrate the integrated Camera of my notebook.

I'm using a 9x6 cheeseboard with a length of 300mm. It's printed on a Konica bizhub 452c and fixated on a drawingboard.

Using the tutorial-Code I'm getting strange undistorted Pictures, which shows that the calibration is bad (example below).

http://answers.opencv.org/question/64905/bad-camera-calibration/

I have feed about 70 pictures in the algorithm (different positions etc.) trying to get trainingpoints as far as possible to the picture edges.

I have tried for days to get an expectable calibration, but I'm only able to minimize the hole-effects on the sides.

Any Help would be appreciated.

If they are need I will provide the calibration-pictures.

regards

Moglei

moglei
  • 21
  • 2
  • Hard to offer any comment without looking at the calibration photos. – Francesco Callari Jul 02 '15 at 13:53
  • I really tried, but can't resist: minimize hole-effects of your cheeseboard? – Photon Jul 02 '15 at 20:50
  • I'm not alowed to post pictures here. The example is in the link. – moglei Jul 03 '15 at 12:45
  • @ Francesco Callari : waht would be the best way to post the 72 pictures that are used to run the calibration – moglei Jul 03 '15 at 12:46
  • @ Photon : with hole-effect I mean the circular distortion on both side of the example picture. I thought this would be a proper discription of the problem. – moglei Jul 03 '15 at 12:49
  • possible duplicate of [How to verify the correctness of calibration of a webcam](http://stackoverflow.com/questions/12794876/how-to-verify-the-correctness-of-calibration-of-a-webcam) – Benjamin-M-Dale Sep 24 '15 at 04:03

2 Answers2

2

I was having the same problem. I calibrated over and over again, but couldn't get any results better than the image you linked to, and sometimes worse. I read this question and answer from the OpenCV Q&A site, and it helped me solve my problem. In the link, you will see that the person answering the question writes that the problem is related to deficiencies in a couple of OpenCV functions that only become apparent when dealing with cameras with strong radial distortion. For me, I was able to "solve" the problem simply by zooming in. The "fish bowl" effect of radial distortion is most pronounced near the edges of the field-of-view, so by zooming in, you are effectively "cropping" your image and thereby reducing the extreme radial distortion. This may not be practical for your application, if you require the widest angle possible, or if your camera doesn't have zoom, but it worked for me!

mannyglover
  • 2,139
  • 14
  • 18
  • While this might be a valuable hint to solve the problem, a good answer also demonstrates the solution. Please [edit] to provide example code to show what you mean. Alternatively, consider writing this as a comment instead. – Toby Speight Nov 17 '16 at 15:35
1

This question is quite old, probably the issue has already been solved. I experienced the same issue with a wide-angle camera, my solution was to use a fisheye model that was able to correctly estimate camera intrinsics and lens distortions.

Mouze
  • 706
  • 5
  • 10