A comparison of IR stereo vision and LIDAR for use in fire environments

This paper presents the development of a far infrared stereo vision system for room mapping in low visibility firefighting scenarios. In this paper, the calibration and rectification of the cameras are presented, along with a discussion of the correspondence matching algorithm selected. Finally, the results from tests of the system in high visibility, clear conditions and low visibility smoke conditions are presented and compared to the ranging capability of LIDAR in the same environments.

[1]  Yazhu Chen,et al.  A Stereo Thermographic System for Intra-operative Surgery , 2005, 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference.

[2]  John S. Zelek,et al.  Sparse Disparity Map from Uncalibrated Infrared Stereo Images , 2006, The 3rd Canadian Conference on Computer and Robot Vision (CRV'06).

[3]  Luigi di Stefano,et al.  A fast area-based stereo matching algorithm , 2004, Image Vis. Comput..

[4]  John S. Zelek,et al.  Dense Surface from Infrared Stereo , 2007, 2007 IEEE Workshop on Applications of Computer Vision (WACV '07).

[5]  Massimo Bertozzi,et al.  Pedestrian detection by means of far-infrared stereo vision , 2007, Comput. Vis. Image Underst..

[6]  Brian Y. Lattimer,et al.  Evaluation of Navigation Sensors in Fire Smoke Environments , 2014 .

[7]  Andreas Uhl,et al.  Infrared camera calibration for dense depth map construction , 2011, 2011 IEEE Intelligent Vehicles Symposium (IV).

[8]  Christopher Assad,et al.  Unmanned ground vehicle perception using thermal infrared cameras , 2011, Defense + Commercial Sensing.

[9]  Hélène Laurent,et al.  Human Detection with a Multi-sensors Stereovision System , 2010, ICISP.