SIGN LANGUAGE TRANSLATION SYSTEM

SIGN LANGUAGE TRANSLATION SYSTEM

V. Anupama 1, Pasala Teja 2, Desetti Sai Madhulika 3, Nalam Hari Prasanth 4 , Lanka Nithin Kumar 5 , Khushi Kumari

1Associate Professor, [2-6] B. Tech Students, LIET

[1,2,3,4,5,6] Computer Science and Information Technology, Lendi Institute of Engineering and Technology,

Vizianagaram.

ABSTRACT

Normal individuals can readily engage and communicate with one another, but those with hearing  and  speech  impairments  have  difficulty  conversing with other hearing people  without  the  assistance  of  a  translator.  For  deaf and dumb individuals,  the  Indian  sign  language  is  a  communication  obstacle. People with hearing and speech impairments rely heavily on nonverbal communication, which includes hand gestures. This is why implementing a system that recognizes Indian sign language would have  a  tremendous  positive influence on the deaf and dumb. In this, a method for automatically recognizing finger printing in Indian Indian sign language is proposed.  In this case, the system is provided the  sign  in  the  form  of  gestures. On  the  input sign picture, multiple stages are conducted. To determine  the  shape  of  the sign, the first segmentation step is done based on skin color. After that, the discovered region is converted to a binary image. The binary image is then transformed using the Euclidean distance transformation. On the distance modified picture, row and column  projection  is  used.  Central  moments,  as well as HU’s moments, are done to extract features. SVM and GREYSCALE are used for classification.

Keywords: Indian sign language Recognition, Convolution Neural Network, Image Processing, Edge Detection, Hand Gesture Recognition