Development of Indian Sign Language Recognition System BY DEEPANAIR.V.S B.S.ANANTHALEKSHMI GAYATHRIMOHAN Guided By DR.DEVARAJ

Introduction A sign language is a language which uses visually transmitted sign patterns to convey meaning by simultaneously combining hand shapes, orientation and movement of the hands, arms or body, and facial expressions to express/communicate with each others. Sign language is commonly used by the physically impaired people who cannot speak and hear. India is diversified in culture, language and religion. Since there is a large diversity among Indian languages, literature survey reports the non-existence of standard form of Indian Sign Language (ISL) gestures. ISL alphabets are derived from British Sign Language (BSL) and French Sign Language (FSL).

INTRODUCTION Indian sign language uses both hands to represent each alphabet and gesture. Sign language recognition is a multidisciplinary research area involving Pattern recognition, computer vision and natural language processing. They are different from spoken languages: the structure of a spoken language makes use of words sequentially, whereas a sign language makes use of several body movements in parallel.

Literature review (Shape, Texture and Local Movement Hand Gesture Features for Indian Sign Language Recognition 1J. Rekha, 2J. Bhattacharya and 3S. Majumder,1Scientist, 3Scientist & Head, Surface Robotics Laboratory, Central Mechanical Engineering Research Institute (CMERI-CSIR) Durgapur, India) They proposed an approach which addresses local-global ambiguity identification,inter-class variability enhancement for each hand gesture. They analyzed the shape, texture and finger features of each hand are extracted using Principle Curvature Based Region (PCBR) detector, Wavelet Packet Decomposition (WPD-2) and complexity defects algorithms respectively for hand posture recognition process. Their Experimental results are compared with the conventional and existing algorithms to prove the better efficiency of the proposed approach.

Literature review(cont.) (Real-time Sign Language Recognition based on Neural Network Architecture Priyanka Mekala1, Ying Gao2, Jeffrey Fan1, Asad Davari Dept. of Electrical and Computer Eng., Florida International University, FL, U.S.A;Electrical Eng.) In real-time, it is highly essential to have an autonomous translator that can process the images and recognize the signs very fast at the speed of streaming images. A image acquisition process is subjected to many environmental concerns such as the position of the video camera, environmental conditions like lighting sensitivity, background condition and camera used. the real time system proposed using the neural networks identification and tracking to translate the sign language to a text format.

Issues in sign Language Recognition Indian sign language uses both hands to represent each alphabet and gesture. Sign language recognition is a multidisciplinary research area involving Pattern recognition, computer vision and natural language processing. Few research works has been carried out in ISL recognition and interpretation using image processing/vision techniques. But those are only initial work tried with simple image processing techniques and are not dealt with real time data.

Objectives of the project To develop an automatic sign language recognition system with the help of image processing and computer vision techniques. To use natural image sequences, without the signer having to wear data gloves or colored gloves, and to be able to recognize hundreds of signs. The motivation for this work is to provide a real time interface so that signers can easily and quickly communicate with non-signers. To efficiently and accurately recognize signed words, from indian Sign Language, using a minimal number of training examples.

System description Real-time processing: -The translator is sufficiently fast to capture images of signer, process the images and display the sign translation on the computer screen. A camera sensor is needed in order to capture the features/ gestures of the signer. Development of the sign recognition system.

Proposed System Architecture

SYSTEM DESCRIPTION Feature Extraction for ISL Alphabet Recognition Two well known feature extraction methods are applied on hand image to obtain the shape and texture information. They are Principal Curvature Based Region detector (PCBR) and 2-D Wavelet Packet Decomposition (WPD). Features represent a particular object in a well defined manner. Sometimes one or more features are required to define an object.

System Description NEURAL NETWORK The performance of the recognition system is evaluated by testing its ability to classify signs for both training and testing set of data. The effect of the number of inputs to the neural network is considered.

Sign database

HARDWARE AND SOFTWARE REQUIREMENTS The system will be implemented in MATALAB. Standard PC (1.5 GHz AMD processor, 128 MB of RAM running under windows 2000.) WEB-CAM-1.3 is used for image capturing.

Plan of action 1.Study about the automatic sign language recognition system 2.Block diagram representation of the proposed ISL recognition 3.Collection of ISL database. 4.Simulation of ISL recognition system using MATLAB. 5.Testing the developed sign language recognition system 6.Creation of GUI design for user interface. 7.Real time implementation of ISL recognition system.

References 1.”Real-time Sign Language Recognition based on Neural Network Architecture” Priyanka Mekala1, Ying Gao2, Jeffrey Fan1, Asad Davari3 Dept. of Electrical and Computer Eng., Florida International University, FL, U.S.A.Electrical Eng. Dept., University of Wisconsin, Platteville, WI, U.S.A. 2. Shape, Texture and Local Movement Hand Gesture Features for Indian Sign Language Recognition 1J. Rekha, 2J. Bhattacharya and 3S. Majumder 1Scientist, 3Scientist & Head, Surface Robotics Laboratory, Central Mechanical Engineering Research Institute (CMERI-CSIR)Durgapur, India. 3. Hand Modeling and Tracking for Video-Based Sign Language Recognition by Robust PrincipalComponent Analysis,Wei Du? and Justus Piater University of Liege, Department of Electrical Engineering and Computer Science Monteore Institute, B28, B-4000 Liege, Belgium