Classification of Emotional Stress and Physical Stress Using Electro-Optical Imaging Technology

Abstract

Emotional stress status is normally to be intertwined with physical stress information. It is meaningful to classify stress for real application such as homeland security and health. In this study, the classification algorithms are proposed based on electro-optical imaging system, such as thermal imaging (TI) system and multispectral imaging (MSI) system. Through the proposed model, the classification signals of ES and PS are successfully obtained. Experiments show that the classification result is encouraging, and the accuracy of the proposed algorithm is over 90%. This study can lead to a useful system for the stress classification and real applications. This research can lay a foundation for the application of stress recognition.