Recognition of Finger Spelling of American Sign Language with Artificial Neural Network Using Position/Orientation Sensors and Data Glove

Abstract

An American Sign Language (ASL) finger spelling and an alphabet gesture recognition system was designed with ANN and constructed in order to translate the ASL alphabet into the corresponding printed and sounded English letters. The system uses a sensory Cyberglove and a Flock of Birds 3-D motion tracker to extract the gestures. The finger joint angle data obtained from strain gauges in the sensory glove define the hand shape while the data from the tracker describes the trajectory and orientation. The data flow from these devices is controlled by a motion trigger. Then, data is processed by an alphabet recognition network to generate the words and names. Our goal is to establish an ASL finger spelling system using these devices in real time. We trained and tested our system for ASL alphabet, names and word spelling. Our test results show that the accuracy of recognition is 96%.