EEG Signal Preprocessing Method For Spatial Cognition Training In Virtual Environment

Electroencephalogram(EEG)signal is the spontaneous and rhythmic electrical activity of brain cell population collected by precise electronic devices such as EEG caps.With the recent development of EEG acquisition equipment,EEG signals have been widely used in many fields,including spatial cognition enhancement field.However,most EEG acquisition devices are non-invasive dry electrodes,which avoid the risk of device implantation to human health and make EEG acquisition easier and more convenient,but it is inevitably easy to introduce various kinds of noises in the collection process,such as the eye electricity,electrocardiogram,myoelectricity,etc.This will cause serious interference to the subsequent analysis of EEG signals and affect the results of EEG signal analysis.Therefore,it is necessary to preprocess the collected EEG original signal to remove noise.Firstly,preprocess the EEG signals in the spatial cognition field.For the problems that some data need to remove more artifacts and require fine preprocessing results,this study combines independent component analysis(ICA)and independent component(Independent component,IC)manual detection techniques to design a high-precision preprocessing method for spatial cognition data.Secondly,in view of the large amount of data,long processing time,cumbersome processing steps,and requiring a large amount of professional preprocessing knowledge,this study developed a set of automatic preprocessing procedures by combining ICA and IC automatic classification technology.It makes the EEG signal processing easier and convenient,and solves the problem of large amounts of data that are difficult to process manually.Finally,this research conducted experiments and analysis on the above content.In this experiment,the original data,the data preprocessed by traditional preprocessing methods,the data preprocessed by high-precision preprocessing methods,and the data preprocessed by fully automated preprocessing methods are analyzed and compared by using the actual collected spatial cognitive training data.The superiority of high-precision preprocessing method and the fully-automatic preprocessing method in spatial cognitive data preprocessing is verified.