Motor imagery electroencephalogram signal classification method based on iterative learning
The invention discloses a motor imagery electroencephalogram signal classification method based on iterative learning, and the method comprises the steps: carrying out the offline training of a patient, obtaining a motor imagery electroencephalogram signal data set with a label, carrying out the fea...
Saved in:
Main Authors | , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
09.09.2022
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a motor imagery electroencephalogram signal classification method based on iterative learning, and the method comprises the steps: carrying out the offline training of a patient, obtaining a motor imagery electroencephalogram signal data set with a label, carrying out the feature extraction of the preprocessed electroencephalogram signal data through a spatial filtering technology, carrying out the binary classification of a feature vector through a support vector machine, and carrying out the classification of a motor imagery electroencephalogram signal. Obtaining a basic classification model; performing principal component analysis on the offline data set, establishing an exponentially weighted moving average model, and calculating upper and lower limits of the model; collecting online training data of a patient, judging whether the principal component of the group of data is located at the upper limit and the lower limit of the model or not, and if the requirement is met, integratin |
---|---|
Bibliography: | Application Number: CN202210760799 |