Publication

Abstract : Sign language (SL) is a mode of communication between normal and deaf-dumb people in which the vision-based methodology is used. The proposed work is to design and develop a sign language recognition system to recognize gestures from Indian sign language (ISL). The signs are captured by using a webcam. These signs are preprocessed for feature extraction using the Fisher algorithm. The obtained features are classified by using the Suppor Vector Machine (SVM) algorithm. After comparing all the features the signs captured with the testing database are calculated for sign recognition. Finally, recognized gestures can convert signs used by deaf-dumb people to a word and speech format. This system provides a better understanding between speech-impaired people and those who may not know sign language. B considering all the possibilities of hand gestures our system provide a high recognition rate when compared with those of recent contribution.