A Human-Computer Interaction System Utilizing Inertial Measurement Unit and Convolutional Neural Network

This paper acquaints a head gesture and eye blink controlled human-computer interaction device to obstruct carpal tunnel syndrome (CTS) and making the computer mouse accessible for the differently able individuals. Traditional computer mouse puts a user's hand for too long in a pronation posture which increases inflammation in the wrist and hand. To solve this problem, in this paper as an alternative to the customary computer mouse is introduced. The system uses three-axis accelerometer and gyroscope to detect the head gesture and translate it into the position of the mouse cursor on the computer screen. The complementary filtering technique obtains a smooth and accurate output from the inertial measurement unit (EMU). To perform the left and right clicking event, the user's need to shut the left and right eye for a moment while opening another eye. The system usages the deep convolutional neural network (CNN) to classify the individual openness and closeness of both eyes. The implementation details along with the performance evaluation results are delineated in this paper show that the system has extensive control over its performance.