Robust Machine Vision Framework for localization of unknown objects

In this paper an approach to the detection of unknown objects is presented. The proposed algorithm is applied to the rehabilitation robot FRIEND II for the localization of objects situated in complex scenes. Also, the method was designed to cope with changes in the illumination conditions. The approach used in this work is the inclusion of feedback control in the image processing chain used by the Machine Vision Framework of the robot. A closed-loop control system was designed at image segmentation level for improving the robustness and reliability of the feature extraction module. The design of the closed-loop is based on an Extremum Searching Algorithm which searches for the optimal parameters of the image segmentation method. The performance of the proposed framework is investigated in comparison with a traditional open-loop method.

[1]  A. Graser,et al.  Rehabilitation robot FRIEND II - the general concept and current implementation , 2005, 9th International Conference on Rehabilitation Robotics, 2005. ICORR 2005..

[2]  David G. Lowe,et al.  Object recognition from local scale-invariant features , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[3]  Chao Wang,et al.  System-controlled user interaction within the service robotic control architecture MASSiVE , 2007, Robotica.

[4]  John Krumm,et al.  Object recognition with color cooccurrence histograms , 1999, Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149).

[5]  Luca Iocchi,et al.  Robust Color Segmentation Through Adaptive Color Distribution Transformation , 2006, RoboCup.

[6]  A. Graser,et al.  Robust Color Object Recognition for a Service Robotic Task in the System FRIEND II , 2007, 2007 IEEE 10th International Conference on Rehabilitation Robotics.

[7]  Danica Kragic,et al.  Vision for robotic object manipulation in domestic settings , 2005, Robotics Auton. Syst..

[8]  Jean-Baptiste de la Rivière,et al.  Image-based analysis for model-based tracking , 2005 .