Real-time cooperative image processing for interactive environment understanding

A real-time cooperative image processing method that enables interactive understanding of the environment is proposed for use in intelligent teleoperation systems. An object recognition process for interactive environmental map building is developed by combining simple independent recognition agents with a cooperation mechanism to achieve real-time data acquisition in a changing environment. The cooperation mechanism is obtained by designing each independent agent to work recursively while referring to a common object model. This method provides a flexible recognition process that interactively produces useful results, even from complicated environment scenes. The proposed method is shown to work successfully.<<ETX>>

[1]  K. Kamejima,et al.  Image based operation for human-robot interaction , 1990, IEEE Control Systems Magazine.

[2]  Ikuo Takeuchi,et al.  Image based operation: a human-robot interaction architecture for intelligent manufacturing , 1989, 15th Annual Conference of IEEE Industrial Electronics Society.

[3]  Ikuo Takeuchi,et al.  Knowledge representation for image based robot operation , 1988, Proceedings of the International Workshop on Artificial Intelligence for Industrial Applications.

[4]  Ikuo Takeuchi,et al.  Dynamic work space model matching for interactive robot operation , 1989, International Workshop on Industrial Applications of Machine Intelligence and Vision,.

[5]  Kohji Kamejima,et al.  Sensory and a priori information integration for object pointing in interactive robot operation , 1993, Proceedings of 1993 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS '93).