Title :
Hybrid intelligent perception system
Author :
Rao, Nageswara S V ; Glover, Charles W.
Author_Institution :
Dept. of Comput. Sci., Old Dominion Univ., Norfolk, VA, USA
Abstract :
A hybrid system capable of recognizing and learning two-dimensional patterns that can be used in imaging applications and robotic systems is described. The hybrid system comprises symbolic and neuromorphic modules which are complementary in character. The symbolic modules are efficient in handling precise objects but are susceptible to noise, whereas the neural networks are robust in the presence of noise and limited in handling issues such as global connectivity. A prototype system is being built to recognize control panels by using a vision system. This system recognizes and classifies control panels, and also learns new control panel models and variations in the existing ones
Keywords :
computer vision; computerised pattern recognition; computerised picture processing; knowledge based systems; learning systems; mobile robots; neural nets; computerised pattern recognition; control panel recognition; global connectivity; hybrid system; imaging applications; intelligent perception system; learning systems; mobile robots; neural networks; neuromorphic modules; pattern learning; robotic systems; symbolic modules; vision system; Control systems; Hybrid intelligent systems; Image recognition; Intelligent robots; Machine vision; Neural networks; Neuromorphics; Noise robustness; Pattern recognition; Prototypes;
Conference_Titel :
Southeastcon '91., IEEE Proceedings of
Conference_Location :
Williamsburg, VA
Print_ISBN :
0-7803-0033-5
DOI :
10.1109/SECON.1991.147876
