electroencephalogram signal classification method based on a Gaussian Bernoulli convolution deep belief network

Abstract

The invention requests to protect an electroencephalogram classification signal method based on a Gaussian Bernoulli convolution deep belief network. According to the method, a blind source separationalgorithm based on the maximum negative entropy is adopted in the preprocessing stage to remove signal interference of motor imagery electroencephalogram; Selecting frequency and electrode parametersbased on the mutual information; combining the unsupervised learning of the Gaussian Bernoulli restricted Boltzmann machine with the convolutional neural network to carry out feature extraction and classification; A new convolutional deep belief network model based on a Gaussian Bernoulli restricted Boltzmann machine can extract meaningful characteristics from a full-size image through a generated convolutional filter. Compared with the prior art, the method has the advantages that a large number of negative weights are reduced, spatial information can be learned from adjacent image patches more effectively, the accuracy of electroencephalogram category judgment is obviously improved, and the accuracy of electroencephalogram category judgment is greatly improved.

Description

technical field [0001] The invention belongs to a method for classifying electroencephalogram signals, in particular to a method for classifying electroencephalogram signals based on a Gauss-Bernoulli convolution depth belief network. Background technique [0002] EEG signal recognition can be applied in fields such as medicine, neuroergonomics, smart environment, education and self-regulation, as well as security and authentication. At present, there are many methods for EEG signal recognition. The method based on support vector machine to classify EEG signals has improved the generalization ability, but the accuracy rate of classification is not high. With the development of deep learning, different types of deep learning methods have been applied in brain-computer interface classification, among which the convolutional neural network model combined with the stacked autoencoder EEG signal classification method in the brain-computer interface competition IV dataset 2b clas...

AI Technical Summary

Problems solved by technology

Gaussian-Bernoulli restricted Boltzmann machine of collaborative filtering is more suitable for the model of non-binary image data than the traditional restricted Boltzmann machine model, but its structure follows the characteristics of the restricted Boltzmann machine model, and there is also Fitting and computationally expensive problems

Images