Title :
A multi-class pattern recognition system for practical finger spelling translation
Author :
Hernandez-Rebollar, Jose L. ; Lindeman, Robert W. ; Kyriakopoulos, Nicholas
Author_Institution :
Dept. of ECE, George Washington Univ., DC, USA
Abstract :
The paper presents a portable system and method for recognizing the 26 hand shapes of the American Sign Language alphabet, using a novel glove-like device. Two additional signs, \´space\´, and \´enter\´ are added to the alphabet to allow the user to form words or phrases and send them to a speech synthesizer. Since the hand shape for a letter varies from one signer to another, this is a 28-class pattern recognition system. A three-level hierarchical classifier divides the problem into "dispatchers" and "recognizers." After reducing pattern dimension from ten to three, the projection of class distributions onto horizontal planes makes it possible to apply simple linear discrimination in 2D, and Bayes\´ Rule in those cases where classes had features with overlapped distributions. Twenty-one out of 26 letters were recognized with 100% accuracy; the worst case, letter U, achieved 78%.
Keywords :
gesture recognition; natural languages; speech synthesis; 28-class pattern recognition system; 2D simple linear discrimination; American Sign Language alphabet; Bayes rule; class distributions; glove-like device; hand shape recognition; horizontal planes; multi-class pattern recognition system; overlapped distributions; pattern dimension; portable system; speech synthesizer; three-level hierarchical classifier; Data gloves; Dictionaries; Fingers; Handicapped aids; Neural networks; Pattern recognition; Shape; Speech synthesis; Synthesizers; Virtual reality;
Conference_Titel :
Multimodal Interfaces, 2002. Proceedings. Fourth IEEE International Conference on
Print_ISBN :
0-7695-1834-6
DOI :
10.1109/ICMI.2002.1166990