Design of Vision System Based on Varying Lighting Condition for Multi-robots

A robot collects environment information mainly by its vision system. In order to improve recognition accuracy and positioning precision under variable lighting conditions, this paper describes a new pattern design and proposes a method to adjust the color threshold according to the credibility of current color threshold. Image morphology is used to optimize the recognition results. The vision system is tested in the project Family Life Support Multi-Robot Systems.

[1]  Christian Laugier,et al.  Automatic camera placement for robot vision tasks , 1995, Proceedings of 1995 IEEE International Conference on Robotics and Automation.

[2]  Sebastian Thrun,et al.  Probabilistic robotics , 2002, CACM.

[3]  Dong-Jo Park,et al.  Fast image segmentation based on multi-resolution analysis and wavelets , 2003, Pattern Recognit. Lett..

[4]  Hideki Hashimoto,et al.  Study on optimal camera arrangement for positioning people in intelligent space , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[5]  Manuela M. Veloso,et al.  Fast and accurate vision-based pattern detection and identification , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).