A Generalized Classification Approach for Indian Sign Language Datasets Using Instance Based Classifiers

Abstract

Sign language is the primary means of communication in the deaf community. The importance of sign language is emphasized by growing research activities and various funding agencies which deal with the technical aspects of deaf people’s communication needs. Signing is primarily done with the hands. In technical terms the hands are used to describe the so-called manual parameters, which are pose, posture, position and motion. General view based sign language recognition systems extract these parameters by a single camera view because it seems to be user friendly and hardware complexity; however it needs a high accuracy classifier for classification and recognition purpose. The decision making of the system in this work employs statistical classifiers namely Navie bayes and K-NN to recognize the sign language isolated signs. Indian sign language datasets is used for the training and performance evaluation of the system.It compares the ability to deploy the K-NN and the Naïve Bayesian classifiers in solving the sign classification problem. The impact of such study may reflect the exploration for using such algorithms in other similar applications such as text classification and the development of automated systems.