Generic placeholder image

International Journal of Sensors, Wireless Communications and Control

Editor-in-Chief

ISSN (Print): 2210-3279
ISSN (Online): 2210-3287

Research Article

An Approach Towards the Representation of Sign Language by Electromyography Signals with Fuzzy Implementation

Author(s): Paromita Das, Sudipta Paul*, Joydev Ghosh, Shilpi PalBhowmik, Biswarup Neogi and Ankur Ganguly

Volume 7, Issue 1, 2017

Page: [26 - 32] Pages: 7

DOI: 10.2174/2210327907666170222093839

Price: $65

Abstract

Background: The rapid developments of sign language recognition systems guarantee the participation of inarticulate people in every field of society. This invention implemented by Electromyography (EMG) sensor to recognize different intramuscular signals for precise recognition of the subjects. In this regards, the Finger and Wrist position with different muscles activity signals are captured by EMG sensor by Fuzzy Logic to develop Matrix model to represent spinal segment and palm muscle activity. This novel approach would provide improvement in the research work for the inarticulate people to design interpreter for less communication system.

Method: The proposed system detects the muscle signal pattern of a particular hand gestures using Electromyography (EMG) signal to developed sign language recognition system and analyzed with Fuzzy Logic. The gesture recognition systems work as following - gesture signal acquisition, processing of acquired gesture signal, descriptor extraction, and finally classification of descriptor to one of the probable gestures.

Results: In order to recognize some simple sign language implement artificial neural network along with fuzzy detection technique. The different EMG signals are analyzed and recognized by Artificial Neural Networks (ANN), and fuzzy detection is involved to obtain more accurate recognition. We consider the uncertainties involved at various stages to have the perfect recognition system i.e., defining image regions, finding features, establishing relations among them, and matching, so that it retains as much as possible information of the original input image for making a conclusion at the highest level. The main objective of this area is to reveal the technical aspects in recognition of particular sign language. The neural network is a strong enough tool to detect the sign expression according to our hand and finger movements. As per our requirement we implement the Neuroph Studio tool, and then a matrix can be arranged with the appropriate data.

Conclusion: This paper proposed a novel approach towards hand gesture recognition implementing data obtained from EMG sensors with promising recognition performance and high reliability. The role of Sign Language Recognition system is to ensure equality of opportunity and full participation of inarticulate people in the society.

Keywords: Electromyography, sign language, fuzzy logic, muscles, envelope, matrix.

Graphical Abstract

Rights & Permissions Print Cite
© 2024 Bentham Science Publishers | Privacy Policy