SSVEP EEG classification method based on convolution neural model enhanced by EMD data

Abstract

The invention relates to a brain-computer interface, an artificial intelligence and an algorithm for pattern recognition, and more particularly, to an SSVEP EEG classification method based on a convolution neural model enhanced by EMD data, which can be applied to the fields of medical instruments, human-computer interaction, robot control and the like. At first, that method preprocess the original EEG data, After preprocessing, the EEG data are decomposed by using the empirical mode decomposition method, and a large amount of artificial EEG data are generated by mixing, which accord with thetime-frequency domain characteristics of the original EEG signal. The artificial EEG data and the original EEG data are combined and used for the parameter training of the neural network, so as to achieve the effect of using a small amount of EEG data to effectively train the network parameters. Finally, the complex Morlet wavelet transform is used to generate EEG tensor, and the original time-domain data is converted into aggregate time-domain data. The tensor dictionary of frequency-domain and spatial information is used as the input of neural network, and the EEG training set is classifiedby convolution neural network model.

Description

technical field [0001] The present invention relates to the algorithm of brain-computer interface, artificial intelligence and pattern recognition, more specifically, relates to the SSVEP EEG classification method based on the convolutional neural model enhanced by EMD data, which can be applied to the fields of medical equipment, human-computer interaction, robot control, etc. . Background technique [0002] In recent years, with the development of computer science and artificial intelligence technology, brain-computer interface technology (BCI), as a new way of human-computer interaction, has important application value in the field of rehabilitation science and control. BCI uses EEG signals to realize communication or control between the human brain and computers or other electronic devices, so it can help patients with physical disabilities restore their ability to communicate with the outside world to a certain extent. Steady-state visual evoked potential is a relative...

AI Technical Summary

Problems solved by technology

And at present, many researchers study EMD-based EEG data enhancement methods, but these studies are only used in the classification research of EEG data linear classifiers, and the EEG data enhancement method is only replacement, rather than generating new data for Expand the original training set

Images