Real-time Emotion Recognition and Environmental Regulation Implemented by an Embedded System

• Main Authors: Wan-Hsin Hsiao, 蕭碗馨
• Other Authors: Kuang-Yow Lian
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/t6422v

碩士 === 國立臺北科技大學 === 電機工程系所 === 102 === The Autonomic Nervous System (ANS) of a human body is generally categorized into sympathetic and parasympathetic nerves, adjusting physiological functions as the two systems interact and antagonize each other. A real-time environmental regulation system, which can bring the user''s ANS to equilibrium, is designed. First of all, in order to measure various kinds of physiological signals using an embedded system, we employ skin conductivity, heart rate and skin temperature signals to extract characteristic points of the respective physiological signals and carry out the analysis. In terms of theoretical analysis, fuzzy decision tree classification algorithm is used to identify three different emotional states, namely, agitation, calm and depression. Lastly, environmental factors are regulated to calm the mind. When a suitable atmosphere is created through automatic lighting and scent control as suggested by principles of color psychology and aromatherapy, the ANS gets balanced. Here, a closed loop automatic regulating system is established.