Mind Stress Detection Using EEG Signal

Abstract

Study of world health organization shows stress could be a vital downside of this generation that affects each physical further because the psychological state of individuals. in line with analysis in space of stress detection has improved several techniques for watching the human brain and Body that detects Stress. the normal stress detection system relies on physiological signals and countenance techniques. This proposes a unique methodology that detects the strain victimization graph signals and reduces the strain by introducing the interventions into the system. Propose methodology delivered system that use SVM rule for divide the topics into completely different classes and to live stress to estimate the strain level. By Result generating throw system humans will take action for determinant best answer for stress management. System generates feedback from stress hormones. The collected information was then accustomed extract a group of options victimization separate riffle rework (DWT). The extracted options square measure manipulated to notice stress levels victimization hierarchical Support Vector Machine (SVM) classifier. For classifying "stressed" and "relaxed" states SVM are studied. Results have shown the potential of victimization graph signal to examine completely different levels of stress. This paper discusses the techniques associated transformations planned earlier in literature for extracting feature from a graph signal and classifying them.