Fuzzy Model for Parameterized Sign Language
Sumaira Kausar, Samabia Tehsin, Muhammad Younus Javed · 2016
Sign languages use visual pattern to communicate rather than acoustic patterns that are communication mode in verbal communication.Sign languages being the most structured form of gestures can be considered as the benchmark for systems of gesture recognition.SLR has got its applicability in the areas of appliances control, robot control, interactive learning, industrial machine control, virtual reality, games, simulations etc. apart from its significance for hearing impaired community.The paper aims to present a systematic, robust, reliable, and consistent system for static Pakistani Sign Language (PSL) recognition.The paper is based on empirical evaluation of different classification techniques for SLR.This pragmatic approach leads to a Fuzzy Model (FM) that has shown very high accuracy rate for PSL recognition.Sign languages have inherent uncertainty, so SLR systems demand a classification method that give due consideration to this aspect of uncertainty.This is the reason for selecting fuzzy inference for the proposed SLR system and experimental analysis has proven its suitability for SLR.The meticulous statistical analysis performed for proposed PSL-FM has shown very promising results.