Smart Wearable Sign Language Interpretation System with Flex and Pressure Sensors

Sign language is a way for deaf and mute people to communicate with others in the society. This study is proposed to develop a smart sign language interpretation system to translate hand and fingers gestures into text and speech. The smart system utilized five flex sensors, two pressure sensors, and a 3-axis motion sensor to detect the American sign language alphabets. The whole system consisted of two modules: a wearable module with sensors and processing unit, and a mobile application module. Sensors data are analyzed by a built-in support vector machine classifier in the processing unit and the resulted text is sent to the mobile application through low energy Bluetooth transmission. The Android-based mobile application converts the received text into speech. Experiments results revealed that the characters recognition accuracy rate to 98.2 %. Current study outperformed existing method such as vision-based which are highly affected by the angles and background factors.