A Classification Method for Continuous Imaging EEG Signals

Abstract

The invention belongs to the field of mode identification and discloses a classification method for continuous imagining brain electrical signals. Firstly, supposing that conversion points exist among different imagining tasks and the Euclidean distance between samples in the conversion points is larger than the distance between the samples in the conversion points, the conversion points are detected by setting the distance threshold value between the samples; secondly, considering that signals will be polluted by noise due to distraction, fatigue or other factors when the same task is imagined continuously, a sample purification concept is added to the strategy, part of samples are screened out from all the samples corresponding to the tasks by setting the range of the distance between the samples, and most of the types of the part of samples are fed back to serve as the type of all the samples of the task. The classification method for the continuous imagining brain electrical signals takes the relationship between the adjacent samples into full account, improves the identification rate of all the samples, and is quite suitable for carrying out off-line analysis on the continuous brain electrical signals.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and relates to a classification method for continuous imagining electroencephalogram signals. Background technique [0002] The research on Brain Computer Interface (BCI) began with the development of EEG. In recent decades, with the development of technology research such as signal processing and machine learning, BCI research has gradually become a hot topic. Motor imagery EEG signal is a common research in the field of BCI. By collecting and analyzing the EEG signal when people imagine moving a certain part of the body or performing certain thinking activities, it can identify the state of people's brain and then control external devices. BCI technology not only provides a new diagnostic method for patients with brain diseases, but more importantly, it realizes a new way for people to communicate with the outside world. [0003] In BCI research, the identification of EEG signals that reflec...

AI Technical Summary

Problems solved by technology

[0004] The output of the classifier model trained by the training data set is the prediction result of a single test sample. For the data set with a certain connection between adjacent samples, this conventional classification method ignores the connection between samples, resulting in the classification The result only depends on the prediction of the classifier, but cannot get a higher recognition rate

Images