Download DLR CalDe and DLR CalLab

Feel free to download the compiled IDL programs as well as some sample files. We kindly draw your attention to the copyright clauses at the end of the document.

1. About

The camera calibration suite consists of two separated programs, that is DLR CalDe (camera Calibration Detection tool) and DLR CalLab (camera Calibration Lab). CalDe aims at the measurement of the salient points of a specific calibration object. The measured points are read and processed by DLR CalLab, which finally provides the intrinsic camera parameters as well as the extrinsic parameters for stereo cameras and/or eye-in-hand cameras.


2. Installation

First, download and unpack DLR CalDe

and DLR CalLab

into a directory of your choice, hereafter named ${CALLAB}.

Second, download and install the free IDL virtual machine (*) into a directory, hereafter named ${IDL}. DLR CalDe and DLR CalLab work at least under Sun Solaris, Linux (32-bit), Mac OS X, and Microsoft Windows (NT, 32-bit).

* NOTE: As of 2010 (IDL version 8.0), EXELIS VIS seems to have broken this tradition and no new virtual machine is available anymore. The (registered) user can however download and install the full IDL package (e.g., IDL 8.5.1). ${IDL}/bin/idl will now take ${IDL}/bin/idlvm's place. Registration is not immediate and might take up some time. After clearance, I'm afraid you must email supporteu@l3harris.com and mention the desired IDL version and platform in order to get access to the Download and Licensing Center (DLC).

Further resources are available:

  • Sample image and points files: sample.zip .
  • A Mathematica notebook with useful methods by Wolfgang Sepp: CallabUtil.zip . It contains I/O, projection, distortion, and undistortion functions.

  • Matlab functions with simpler methods for undistortion and triangulation by Klaus Strobl: matlab_f_undistort_triangulate.zip . In detail it contains:

    • The function  f_undistort_features_iterations.m, which undistorts a vector of projections using the intrinsic parameters A, k1, and k2 by plain iterations.
    • The function f_undistort_features_jacobianiterations.m, which undistorts a vector of projections using the intrinsic parameters A, k1, k2, and k3 by Jacobian-based iterations. The Jacobian between the undistorted normalized coordinates and the residual error in the distorted image coordinates is used. Jacobian-based iterations are reportedly more accurate than the blind approach above, hence the method faster.
    • The function f_undistort_image.m, which undistorts a dense, 3-channel image using the intrinsic parameters by interpolation.
    • The function f_triangulate.m, which triangulates a 3-D solution from undistorted projections on a calibrated stereo camera in a non-optimal, LLSQ way.


3. Execution

Run CalDe via
${IDL}/bin/idlvm -vm=${CALLAB}/calde/calde.sav

Run CalLab via
${IDL}/bin/idlvm -vm=${CALLAB}/callab/callab.sav


4. Newsletter

Our newsletter informs on new versions of DLR CalLab and DLR CalDe. To receive mail please subscribe by sending an e-mail with the subject "subscribe callab" to Klaus.Strobl AT dlr.de from the desired e-mail delivery address.


5. BibTeX entry

You can use the following entry to cite this work:

@electronic{callab,
  author = {K. H. {S}trobl and W. {S}epp and S. {F}uchs and C. {P}aredes and M. {S}misek and K. {A}rbter},
  title = {{D}{L}{R} {C}al{D}e and {D}{L}{R} {C}al{L}ab},
  organization = {Institute of Robotics and Mechatronics, German Aerospace Center (DLR)},
  address = {Oberpfaffenhofen, Germany},
  url = {http://www.robotic.dlr.de/callab/},
  howpublished = {\url{http://www.robotic.dlr.de/callab}}
}


6. Copyright

DLR CalDe has been developed at the Institute of Robotics and Mechatronics of the German Aerospace Center (DLR), Oberpfaffenhofen (Germany) by Wolfgang Sepp, Stefan Fuchs, Klaus Arbter, Klaus Strobl, and Michal Smisek.

DLR CalLab has been developed at the Institute of Robotics and Mechatronics of the German Aerospace Center (DLR), Oberpfaffenhofen (Germany) by Klaus Strobl, Cristian Paredes, and Michal Smisek.

These program are distributed in the hope that they will be useful, but WITHOUT ANY WARRANTY. You are free to use and redistribute this software as long as
a) the software is used for non-commercial purposes only,
b) the software is not modified,
c) the credits to authors and the license terms are made clear to the user.

Any of these conditions can be waived if you get permission from the copyright holder.