Xie Liping, Lu Chihua, Liu Zhien, Yan Lirong, Xu Tao
Hubei Key Laboratory of Advanced Technology for Automotive Components, Wuhan University of Technology, Wuhan, China.
Foshan Xianhu Laboratory of the Advanced Energy Science and Technology Guangdong Laboratory, Foshan, China.
Front Hum Neurosci. 2021 Oct 8;15:663049. doi: 10.3389/fnhum.2021.663049. eCollection 2021.
The research shows that subjective feelings of people, such as emotions and fatigue, can be objectively reflected by electroencephalography (EEG) physiological signals Thus, an evaluation method based on EEG, which is used to explore auditory brain cognition laws, is introduced in this study. The brain cognition laws are summarized by analyzing the EEG power topographic map under the stimulation of three kinds of automobile sound, namely, quality of comfort, powerfulness, and acceleration. Then, the EEG features of the subjects are classified through a machine learning algorithm, by which the recognition of diversified automobile sound is realized. In addition, the Kalman smoothing and minimal redundancy maximal relevance (mRMR) algorithm is used to improve the recognition accuracy. The results show that there are differences in the neural characteristics of diversified automobile sound quality, with a positive correlation between EEG energy and sound intensity. Furthermore, by using the Kalman smoothing and mRMR algorithm, recognition accuracy is improved, and the amount of calculation is reduced. The novel idea and method to explore the cognitive laws of automobile sound quality from the field of brain-computer interface technology are provided in this study.
研究表明，人的主观感受，如情绪和疲劳等，可通过脑电图（EEG）生理信号客观地反映出来。因此，本研究引入了一种基于EEG的评估方法，用于探索听觉脑认知规律。通过分析三种汽车声音（即舒适性、动力性和加速性）刺激下的EEG功率地形图，总结脑认知规律。然后，通过机器学习算法对受试者的EEG特征进行分类，从而实现对多样化汽车声音的识别。此外，采用卡尔曼平滑和最小冗余最大相关性（mRMR）算法提高识别准确率。结果表明，多样化汽车音质的神经特征存在差异，EEG能量与声音强度呈正相关。此外，通过使用卡尔曼平滑和mRMR算法，提高了识别准确率，减少了计算量。本研究提供了从脑机接口技术领域探索汽车音质认知规律的新思路和方法。
