안면근육 표면근전도 신호기반 특징 및 근육 선별을 통한 단모음인식

초록

In this paper, we propose monophthong recognition method using feature and muscle selection based on facial surface EMG signals. We observed that EMG signal patterns may vary according to Korean monophthong pronunciation. The proposed method can recognize Korean monophthong and improve recognition accuracy by selecting muscle and feature rather than by using all of them. Experimental results show that the recognition accuracy of the proposed method using best muscle and best, second features set is better than that using all of muscles and features. The improved average accuracies are 29.13% in kNN, 34.52% in LDA.