An intelligent wearable to aid speech impaired people by detection of specific hand gestures using flex sensors

Abstract

This paper presents a special method of communication for the speech impaired people and other patients using their hand gestures. A microcontroller based system has been developed for measuring fingers flexion to identify specific hand gestures using flex sensors. The system is made wearable and has been deployed on a glove. To measure the fingers flexion for a specific hand gesture five flex sensors are tailored for each finger on the glove. These flex sensors produce signals corresponding to different hand gestures and finger movements. Flex sensors simply change their resistance values according to the degree of bend angle made. They convert the change in bend to electrical resistance-the more the bend, the more is the resistance value. The system has been built around a widely used eight bit microcontroller ATMega328P that has a built-in 10-bit multiplexed channel Analog-to-Digital Converter for the signal conditioning purpose. The important issues while designing a glove is the calibration, the finger size, height and thickness varies from patient to patient and these inaccurate measurements of parameter leads to overlap of the sign.