ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | Q&A
Ask Your Question

Hokuyo to camera calibration

asked 2011-04-25 05:28:21 -0600

Jose gravatar image

updated 2014-01-28 17:09:35 -0600

ngrennan gravatar image

I am trying to do the same type of calibration as you guys did in Scanner to camera calibration

The problem that I'm having is that the corners cannot be found properly in the intensity image.

I was wondering if someone could give me some references to the "interpolation derived from color interpolators, preserving dominant edges" algorithm? That way I could try and implemented to enhance the intensity image. Or if the algorithm is already implemented as open source where can I find it?

Thanks in advance.

PS: I was going to attach the intensity image but I don't have enough "karma" to do so.

edit retag flag offensive close merge delete

2 Answers

Sort by ยป oldest newest most voted

answered 2011-04-25 08:28:59 -0600

vpradeep gravatar image

updated 2011-04-25 08:29:32 -0600

Hi Jose,

I am not familiar with the "interpolation derived from color interpolators, preserving dominant edges". However, the pr2_calibration stack deals with this issue by naively upsampling the image before using OpenCV's checkerboard detector. The OpenCV checkerboard detector seems to perform much better on upsampled intensity images.

You can see an example of this approach in pr2_calibration/laser_cb_detector/src/laser_cb_detector.cpp.

edit flag offensive delete link more


Thank you for the quick response, I have tried upsampling the image. Although, it does give me better results they are far from prefect, that is why I decided to try and find a way to enhance the image.
Jose gravatar image Jose  ( 2011-04-25 09:37:09 -0600 )edit
Are you also running the subpixel refinement step (`cvFindCornerSubPix`)? This should significantly improve your results.
vpradeep gravatar image vpradeep  ( 2011-04-25 10:27:53 -0600 )edit
Yes and I'm actually using the corner finding algorithm from Vezhnevets Vldimir. This algorithm gives me better results than the OpenCV cvFindChessboardCorners.
Jose gravatar image Jose  ( 2011-04-25 12:24:20 -0600 )edit

answered 2011-04-25 07:34:55 -0600

fergs gravatar image

updated 2011-04-25 07:35:11 -0600

I believe you can find the source as part of the pr2_calibration stack (, however, there are multiple notes on the stack documentation saying it is still very unstable and PR2-specific.

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Question Tools


Asked: 2011-04-25 05:28:21 -0600

Seen: 889 times

Last updated: Apr 25 '11