Patient tracking for video/EEG monitoring based on change detection in DCT domain

Video/EEG monitoring is an important application for diagnosis of epilepsy patient. During the recording, the camera needs to be manually controlled to cover the patient. As the recording procedure is usually extensive in time, automatic tracking of the patient is a practical need. In this paper, we present a tracking method based on change detection techniques. Movements of the subject are detected and utilized to determine the subject´s location. The proposed change detection algorithm is carried out in DCT domain to handle video in a compressed format. Promising results have been obtained from simulation experiments.