The Methodology Of Artifacts Detection And Removal Model And Application In Eye Closing And Opening Detection System

Within neuroscience computing the development of EEG (Electroencephalogram) based application has become more and more pressed for research. In this field, the artifacts removal has become a promising and challenging topic. The artifacts disturb analysis and feature extraction of EEG signals. Moreover, they are sometimes mistaken as EEG signals in study due to the overlap in frequency domain. Many studies have revealed certain weaknesses of the methods handling artifacts in previous EEG researches. Few studies have discussed combine artifacts detection and artifacts removal for the artifacts in EEG and design an eye closing and opening detection system with practical value.In the article, we present the artifact removal model for three artifacts, which influences EEG signals most, electrooculography (EOG), electromyography (EMG) and50/60Hz mains interference. We present an order that artifacts detection and artifacts removal should follow, and an appropriate alarm thresholds for artifacts detection. Moreover, different levels of noise alarm thresholds are used to determine the thresholds for noise removal methods. The proposed method greatly improves the EEG noise removal efficiency. Furthermore, An proposed method in removal of ocular artifacts from EEG is used which proved to be efficience and useful. In the eye closing and opening detection system, the performances of this method has been evaluated with20subjects and100trails and develop the alpha ratio threshold for detection and the individual difference and short-term stationary of alpha rhythm. The results show that the proposed method is more effective and has lower Time complexity comparing with traditional methods, which has several promising practical uses.