Enhanced segmentation of EOG signals

Extraction of certain events from a signal is a very common problem in the field of medical signal processing. This paper presents a combined method to extract events from electro-oculogram (EOG) signal that are characterized by a band of frequencies. The method uses a combination of mean weighted instantaneous frequency and Pan-Tompkins algorithm. The goal is to improve specifically the identification of the start and end of the detected event. The algorithm is tested and applied on the Electro-oculogram signals and the results are presented.