Sign Language to English Translator using Palm and Finger Tracking

Abstract

It is estimated that, worldwide, 1 in 1000 people are deaf. Thus considering the world population which is of more than 6.8 billion, 0.1% = 6.8 million may be a rough estimate for the total number of deaf people in the world. Deaf people use gestures and their hands to express their ideas to another. Most of the normal people do not understand this language of the deaf which is called the sign language. Thus when a normal person and deaf person try to communicate with each other both will not be able to perform a fully understandable communication. There is a language barrier which exists between the worlds of deaf and non-deaf and it is the main reason that these two worlds still unable to merge to one world. This project provides a solution to reduce this barrier, thus the objective of the proposed project is to identify the palm and finger based signs, gestures made using sign language and translate it to English thus that a normal person will be able to understand what is meant using the sign language. This translator tool implemented in this project uses statistical database for recognizing the signs and due to that it can be used as a verification tool to practice sign language thus that the sign language can be learn without instructor. Idea of this thesis is to prove the importance of building a statistical database store the features of the signs and then to map the features stored in the database to the signs.