Electroencephalogram channel optimization method based on sparse learning and domain adversarial network

Abstract

The invention discloses an electroencephalogram channel optimization method based on combination of deep learning and sparse learning. The method comprises the steps that firstly, a model driving experiment is used for collecting a data set, samples in a source domain and a target domain tend to be balanced through data expansion, sparse learning and domain adversarial learning are conducted on the balanced samples, and the design initial intention is to minimize loss values of a label predictor and a discriminator at the same time; based on the purpose, an objective function is designed, themodel has the feature selection capability by adding L21norm, in addition, GAN is used, and the robustness and generalization capability of the model are improved to a certain extent. And finally, inthe experimental evaluation stage, on one hand, the performance of the method is independently evaluated, and on the other hand, the performance is compared with other channel optimization algorithm items, and unique advantages are obtained. On the other hand, on the premise that the accuracy is guaranteed, the number of channels can be effectively reduced, and therefore the burden and expenditureof the system are reduced.

Description

technical field [0001] The present invention relates to the field of brain-computer interface and EEG data, in particular to an EEG channel optimization method based on the combination of deep learning and sparse learning. Background technique [0002] Brain-computer interface (BCI) based on EEG is more and more widely used in the fields of mental state detection, emotion recognition, and auxiliary epilepsy treatment, and the research on fatigue driving based on EEG has attracted more and more attention. But so far, there is a contradiction in the convenience of signal analysis and acquisition of EEG signal acquisition devices, that is, the more acquisition channels, the easier the analysis, but the more inconvenient to wear, that is to say, the EEG signal acquisition channel The more, the easier it is to analyze, but the less convenient it is to wear. Therefore, how to reduce and optimize EEG electrodes to obtain high-quality EEG signals for analysis has attracted widespre...

AI Technical Summary

Problems solved by technology

[0005] The present invention solves two problems by applying sparse learning based on deep learning and domain confrontation network to the cross-subject EEG channel optimization: first, multi-channel will greatly increase the energy consumption and complexity of the system; second, The second is that the traditional channel optimization channel method performed poorly in cross-subject experiments

Images