A framework for hand gesture recognition based on fusion of Flex, Contact and accelerometer sensor

This paper presents a framework for hand gesture recognition using information received from Flex, Contact and 3 axis accelerometer sensor. A final decision is made on the basis of matching the sensor values with the already stored values for each sign in the database. Experimental results for 26 American Sign Language Recognition demonstrate the functionality and effectiveness of the framework. An accuracy of 100% is achieved for recognizing the ASL alphabets.