DYNAMIC SIGN LANGUAGE INTERPRETER USING DEEP LEARNING


Siddhesh Garud1, JayeshPatel2, Mayur Wagh3, Assistant Prof. Satish Kuchiwale4
Students, Dept. of Computer Engineering, Smt. Indira Gandhi College of Engineering, Navi Mumbai, Maharashtra, India.
Professor, Dept. of Computer Engineering, Smt. Indira Gandhi College of Engineering, Navi Mumbai, Maharashtra, India.

Abstract: Sign language recognition has emerged as one of the important areas of research in computer vision. As the community of speech and hearing impaired people has depended on sign language as a communication medium. New techniques have been developed from the past decade still now, to counter the problem of building a communication bridge between normal people and speech and hearing-impaired people. Sign gestures can be classified as static and dynamic. Static gesture recognition is easier than dynamic gesture recognition but both recognition systems are important for the human community This paper focuses on different techniques used for sign language recognition.
Keywords: Sign Language Recognition, Computer Vision, ISL.
I. INTRODUCTION
Sign language is the most natural ways of exchanging information among deaf and dumb people. It has been observed that deaf people are facing difficulty interacting with other people. The goal of sign language recognition is to provide an efficient and accurate system to convert sign language into text so that communication between deaf and normal people can be more efficient. Sign language consists of vocabulary of signs in exactly the same way as spoken language consists of a vocabulary of words. Indian sign language (ISL) is sign language used in India. ISL involves both static and dynamic gestures, single as well as double handed gestures, in addition to this the hands involved in gesturing may have complex motion. Some signs include facial expressions too. Because of these difficulties less research work has been done in ISL recognition system. A thorough literature survey covering almost all the aspects of the SLR is a primary step to build a ISL recognition system.

Any non-verbal communication like the motion of hands, facial expression and other body parts is a form of gesture. Gesture recognition enables devices to understand human actions. Normal people do not want to learn sign language. That’s why this community becomes isolated from others, they cannot express themselves. So if computer can be programmed in such a way that it can translate sign language to some speech or text format, the difference between the normal people and the deaf and dumb community can be minimized and then communication between Deaf and Dumb community will be easy. Now there is a requirement for such a type of system which can recognize and translate sign language into text or speech format. Almost all gestures have already assigned meaning and grammar is used to create a meaningful sentence from a set of recognized gestures. ISL gestures are mostly made up of two hands. So it is quite difficult to recognize. Continuous ISL or Continuous Sign language is a sequence of gestures that generate a meaningful sentence. Continuous ISL includes dynamic gesture recognition.

II. APPROACHES
There are basically two main approaches to solve sign language recognition problems.

● Static gesture

Static gesture recognition aims to identify particular kinds of posture which remain still for a period of time in the videos. Static gestures are single images which involve no time frame. We use the adaboost algorithm when training cascade classifiers, which can promote systems to be robust and realtime and can be used to identify static gestures..
● Dynamic gesture
In sign language recognition systems, gestures cannot be judged at the time they are detected, since the dynamic gestures should be taken into consideration. Namely, we do not judge gestures only by one image but just record the position of the hand because it may be one fragment in a series of images which contains a dynamic gesture such as a wave. Dynamic gestures are single images which involves multiple time in frame




dynamic-sign-language-interpreter-using-deep-learning
FREE DOWNLOAD
@ engpaper.com published paper
PUBLICATION PROCEDURE WITH US ENGPAPER.COM
ENGPAPER.COM PUBLISHED PAPERS