Electroencephalogram signal classification method and device and storage medium

Abstract

The invention relates to an electroencephalogram signal classification method and device and a storage medium, and belongs to the technical field of computers. The method comprises the steps of obtaining n target electroencephalogram signals to be classified; based on an adaptive neighborhood search algorithm, determining a neighborhood of each target electroencephalogram signal in the n target electroencephalogram signals; based on a local tangent space arrangement algorithm, mapping the high-dimensional data of each target electroencephalogram signal to the same low-dimensional subspace by using the neighborhood of each target electroencephalogram signal to obtain data features after dimension reduction; using a clustering algorithm to classify the data features after dimension reduction to obtain classification results of the n target electroencephalogram signals. The problems that classification boundary drift is caused by signal attenuation, and electroencephalogram signals are wrongly classified can be solved; local linearity can be measured through an adaptive neighborhood search algorithm to expand the neighborhood, class boundary points drifting due to attenuation are correctly classified, and the accuracy of electroencephalogram signal classification is improved.

Description

【Technical field】 [0001] The present application relates to a method, a device and a storage medium for classifying electroencephalogram signals, which belong to the technical field of computers. 【Background technique】 [0002] EEG signals are non-linear, non-stationary time-series signals that can be detected by sensors on electrodes on the scalp, and these signals are the external manifestations of neuron membrane potential. By classifying EEG signals, recognition of actions or diseases can be achieved. [0003] Since the EEG signal will be attenuated from generation to collection due to factors such as network, communication, equipment, environment and other factors, as well as the state of the acquisition object, the signal attenuation can reduce the density of the original EEG signal, resulting in different classifications of EEG signals. The boundary of the electrical signal is blurred, which has a bad influence on the feature extraction and classification of the sign...

AI Technical Summary

Problems solved by technology

[0004] The application provides an EEG signal classification method, device and storage medium, which can solve the problem of classification boundary drift caused by signal attenuation and EEG signals being misclassified

Images