Sign language recognition using data gloves

Abstract

Sign language is a method of communication primarily used by the hearing impaired and mute community. In this method, a letter is expressed by hand gestures. Meaningful words can be constructed by signaling multiple letters in a sequence. This is known as fingerspelling. For a non-sign-language speaker it is difficult to communicate with someone well-versed in sign language without assistance from professional interpreters. Therefore, it is worthwhile to develop a system which allows a non-sign-language speaker to understand the fingerspelling of a sign language. In this work a system has been developed to detect fingerspelling in American Sign Language (ASL) and Bengali Sign Language (BdSL) using data gloves. A data glove is just a glove which has a number of sensors attached to it. This study identifies a way to construct a suitable glove for both the languages. The methodologies employed can be used in resource-constrained environments. The system is capable of detecting both static and dynamic symbols in the alphabets. Furthermore this work presents a novel approach to perform continuous assessment of symbols from a stream of run-time data.