Classification method of steady-state visual evoked potential based on empirical mode decomposition

Abstract

The present invention discloses a classification method of steady-state visual evoked potential based on empirical mode decomposition. Firstly, collected multichannel steady-state visual evoked potential (SSVEP) is subjected to multi-element empirical mode decomposition to be decomposed into several sub-signals on different frequency band ranges; then according to correlation coefficients betweenthe sub-signals of known classification label signals and template signals, classification suitability indexes corresponding to the sub-signals are calculated; then correlation coefficients between the sub-signals of unknown label signals and the template signals are calculated, the classification suitability indexes are used as a selection weight of sub-signal correlation coefficients, and correlation coefficients between original signals and the template signals are reconstructed; and finally, the steady-state visual evoked potential (SSVEP) is classified according to the reconstruction correlation coefficients between the original signals and the template signals. On a basis of improving a signal-to-noise ratio of the steady-state visual evoked potential (SSVEP), the classification method realizes a relatively high classification accuracy rate for the steady-state visual evoked potential (SSVEP).

Description

technical field [0001] The invention relates to the technical field of classification of steady-state visual evoked potential signals (electroencephalogram signals), in particular to a classification method of steady-state visual evoked potentials based on empirical mode decomposition. Background technique [0002] Brain-computer interface is a new type of human-computer interaction, which can control external devices by reading the brain neural activity information of the subject. In actual use, cortical electroencephalogram (EEG) is used as the main signal source in the brain-computer interface control system due to its advantages of high time resolution and convenient signal extraction. The brain-computer interface of event-related EEG signals such as motor imagery and P300 is the basic mode of the brain-computer interface. After continuous exploration and research, steady-state visual evoked potential signal (SSVEP) has become the most widely used EEG signal in the brai...

AI Technical Summary

Problems solved by technology

[0004] (1) The canonical correlation analysis algorithm regards the steady-state visual evoked potential signal as a single EEG signal, and will lose effective feature information in the process of filtering and other preprocessing operations, resulting in limited classification accuracy

[0005] (2) In many improved algorithms, there is a lack of prior knowledge on the selection and weight of the sub-signals of the steady-state visual evoked potential signal after frequency division, and a lot of factors will be added in the process of decomposing and reconstructing the sub-signals. Noise signal, unable to effectively achieve the purpose of improving the signal-to-noise ratio of the steady-state visual evoked potential signal

Images