Sign Language Problems

1521 Words7 Pages
HAND GESTURE RECOGNITION ALGORITHM IN ISL P.S.Neethu1,V.Rajakumari2,S.Umarenuka3 Assistant Professor, Department of Electronics and Communication , New Prince Shri Bhavani College of Engineering and Technology, Chennai, India1 Student, Department of Electronics and Communication, New Prince Shri Bhavani College of Engineering and Technology, Chennai, India2,3 ABSTRACT Sign Language is a natural language which deaf community uses for communication. Sign Language (SL) is a subset of gestures or signs made with fingers, hands, arms, eyes, and head, face etc. Each gesture in SL has a meaning assigned to it. Understanding SL is nothing but understanding the meaning of these gestures. There exists a problem in communication when…show more content…
It is a language which talk with help of hands movements, body parts and facial expression. Every country has its own developed SL. In India, this language is called as "Indian Sign Language (ISL)". Indian Sign language is commonly used sign language among deaf people in India. In different parts of India it has different signs but grammar is same throughout the country. Similarly in different states of India has different languages(for example in Maharashtra Marathi, In Gujarat Guajarati etc.., same way signs language has little difference in contradictory region. Spoken language is not connected to sign languages. sign language has its own grammar structure. There are hundreds of sign languages used by normal person who can hear but could not speak or one who has difficulties in speaking and by normal…show more content…
This can be attained by adopting an algorithm proposed in that uses HOG features of an image and classifies them using a (SVM)Support Vector Machine classifier. As the database for static signed gestures in the Indian Sign Language are very scarce online, the method is trained over a manually created dataset that comprises of a set of 10 images each for every signed alphabet or phrase. Following the segregation, the features are extracted from the set of training images for gestures separately. This is done by adding the HOG features of the test image to the respective feature matrices of the training data and Support vector Machine (SVM) classifier used for the classification of gesture with the highest value for correlation with the test image is recognized as the gesture in the test
Open Document