Date Added: Jan 2012
The important problem of human psycho-physiological state recognition is studied. An integrated hardware - software setup has been developed to achieve automatic assessment of the affective status of a human. Several signal processing techniques are applied to the collected signals to extract the set of the most relevant in the physiological responses features. The support vector machine was used as a classifier. As a result 93.4% functional state prediction accuracy was obtained. Physiological data can provide useful information about human emotional or cognitive states and can help in the recognition of the level of cognitive load or of the presence of stress.