Emotion Recognition by Spatiotemporal Analysis of EEG Signals

In several security related and medical applications, it would be helpful to obtain objective information regarding a person's emotional state to ascertain a person's thoughts and intentions. This leads to the concept of a system, which would obtain information pertinent to a person's emotional state and process the information to identify the signal characteristics of a particular mindset to perform an activity.

EEGs associated with various emotional states are recorded, dominant features of EEGs are extracted and processed with machine training algorithms to classify and identify each mindset to perform an activity or to communicate with others.

Electroencephalograms (EEGs) can be used for implicit communications. A person can express his/her thoughts only by thinking. This research is to explore the possibility of employing EEGs into applications where the presence of noise is more prominent than the information content.

EEGs from different states of mind were recorded and processed to eliminate unrelated features, reduce dimensions while maximize information. The state of mind associated with each recording was recognized by an artificial neural network based technique.

Using EEGs to recognize human intentions or thoughts opens new field of study in brain computer interaction. The technology allows paralyzed patients to interact with other people or to control devices for human aid.

Decision Making