Title :
Hybrid AI system for geometric pattern recognition
Author :
Fernando, Christopher G. ; Munasinghe, Ranjith
Author_Institution :
Dept. of Ind. Technol., West Virginia Univ. Inst. of Technol., Montgomery, WV, USA
Abstract :
The research area of hybrid and neural processing has been actively developing. Hybrid neural systems are computational systems which are based mainly on artificial neural networks but also allow a symbolic interpretation or interaction with symbolic classical artificial intelligence. In this paper we describe a hybrid AI system developed for 2D object recognition. The 2D object recognition system was developed as the initial step for developing a 3D object recognition system for an unmanned aerial vehicle (UAV).
Keywords :
artificial intelligence; computer vision; neural nets; object recognition; remotely operated vehicles; 2D object recognition; artificial intelligence; artificial neural networks; computational systems; geometric pattern recognition; hybrid AI system; hybrid neural systems; symbolic interpretation; unmanned aerial vehicle; Artificial intelligence; Artificial neural networks; Biological neural networks; Computer networks; Expert systems; Humans; Neurons; Object recognition; Pattern recognition; Robustness;
Conference_Titel :
System Theory, 2004. Proceedings of the Thirty-Sixth Southeastern Symposium on
Print_ISBN :
0-7803-8281-1
DOI :
10.1109/SSST.2004.1295633
