Indexed by:Conference Paper

Date of Publication:2009-10-17

Included Journals:Scopus、CPCI-S、EI

Page Number:4495-4499

Key Words:Stress detection; fuzzy c-means

Abstract:A method based on fuzzy c-means (FCM) clustering algorithm is proposed to detect stress continuously in this study. The method calculates the exact stress value of each period and achieves a continuous stress curve. Biomedical signals used in this study were collected from drivers in a driving experience, and appropriate features are selected to form multi-dimensional feature-vectors. By using FCM algorithm, these feature-vectors are clustered to several clusters. Stress value of each period is calculated based on the membership degree between feature-vectors and clusters. The experience results by using signals acquired from some drivers' driving experiences show that the method may distinguish stress of different driving periods clearly, and the stress curve may give a direct-viewing of change of stress.