sEMG-based lip shapes recognition

摘要

With the development of intelligent technology, the need for more effective and practical human-computer interaction becomes more and more urgent. The objective of this paper is to use the facial surface electromyography (sEMG) for lip shapes recognition. We recorded the sEMG signals of eight mouth movements from five subjects, used various feature extraction methods to do classification and used a feature selection method. . Results show that the best classification performance was obtained from Support Vector Machines and the efficient number of features is five. Our research would be a potential method to implentate a robust human computer interface further. High recognition accuracy means that we can send commands to control external devices by collecting sEMG signals and predicting their expressed actions to achieve effective human-computer interaction.