Proceedings of The 5th International Conference on Innovation in Science and Technology
Hand Gesture Classification based on Inaudible Sound using Convolutional Neural Network
Jinhyuck Kim, Jeongung Kim and Sunwoong Choi
Recognizing and classifying the gesture of a user has become important for an increase in the use of wearable devices. This study propose a method for classifying hand gestures by creating inaudible sound using a smartphone and reflected sound signal. The proposed method converts the sound data, which has been reflected and recorded, into an image using short-time Fourier transform (STFT), and the obtained data are applied to a convolutional neural network (CNN) model to classify hand gestures. The results showed classification accuracy for 6 hand gestures with an average of 92.17%.Furthermore, it is confirmed that the proposed method has a higher classification accuracy than other machine learning classification algorithms.