Sign Language Recognition Based on EMG Signals through a Hibrid Intelligent System

Abstract

. Non-verbal communication is an important part of everyday interactions and human-computer interaction. Vision techniques and instrumented gloves for sign language recognition are commonly used, but these are often expensive and considered invasive to the user. This research proposes the recognition of words from the American Sign Language (ASL) using the SCEPTRE database acquired by two Myoelectrical bracelets. Computational intelligence techniques were used to optimize the number of attributes using Principal Component Analysis (PCA) and a classifier based on Neural Networks (NN). The results suggest that it is possible to reduce the attributes using PCA without significantly losing the quality in classification. This allows faster processing, a convenient feature for classifiers for real-time SL recognition.