Abstract

A basic method for recognizing sign language has been developed with thermal-image processing. In this method, a pair of traces of two hands, which corresponds to a word in Japanese sign language, is input into a computer as sequential coordinates with thermal-image processing. Then the word is recognized by finding the most similar pair of traces in the database to that expressing the word. Forty-seven Japanese words in sign language were recognized with 71% accuracy.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call