Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Canonical correlation analysis-based electroencephalogram alpha wave detection and identification method

A technology of canonical correlation analysis and recognition method, applied in the field of EEG alpha wave detection and recognition based on canonical correlation analysis, can solve problems such as not suitable for fast detection, low detection efficiency, long signal duration, etc., to reduce the amount of calculation, Effects of rapid detection recognition, improved speed and accuracy

Active Publication Date: 2018-08-10
XIDIAN UNIV
View PDF9 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method can detect and utilize α-wave signals more effectively, but the disadvantage of this method is that the detection time needs at least 2 seconds, and the detection efficiency is low
[0006] To sum up, at this stage, the method of energy analysis is mainly used for α wave detection. Only when the signal energy continues to exceed a certain threshold can the judgment be made. This type of method requires a long signal duration and is not suitable for rapid detection.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Canonical correlation analysis-based electroencephalogram alpha wave detection and identification method
  • Canonical correlation analysis-based electroencephalogram alpha wave detection and identification method
  • Canonical correlation analysis-based electroencephalogram alpha wave detection and identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] At present, the method of energy analysis is mainly used in the detection and identification of EEG α waves. When the amplitude of EEG signals exceeds a threshold within a period of time, the signal will be considered as an α wave signal. This type of method requires a long duration of the signal, which is not suitable for a brain-computer interface system that requires fast detection and recognition. The present invention specifically studies a method for detecting and identifying EEG alpha waves based on canonical correlation analysis, see figure 1 , including the following steps:

[0031] (1) Input original EEG signals, including EEG α-wave signals and EEG non-α-wave signals.

[0032] (2) Preprocessing the original EEG signal: select the data of the four EEG signal channels Cz, CPz, Pz, and POz, perform a band-pass filter of 0.5 Hz to 30 Hz on the selected data, and then segment the data, The time length of each segment of data is t seconds, and after segmentation,...

Embodiment 2

[0048] The method for detecting and identifying EEG alpha waves based on typical correlation analysis is the same as in embodiment 1, the selection of different frequency sets described in step (3) and the reference signals corresponding to the construction of different frequencies, specifically including as follows:

[0049] (3.1) Select different frequency sets F ref : Frequency selection is carried out with a certain step size δ as an interval within 8Hz~13Hz, and N frequencies are obtained, that is, N=(13-8) / δ+1=5 / δ+1, and N frequency combinations form different frequency sets F ref .

[0050] In this example, the step size δ is selected as 0.2Hz, then N=(13-8) / 0.2+1=5 / 0.2+1=26, so different frequency sets F ref Consists of 26 frequencies.

[0051] (3.2) Construct reference signals corresponding to different frequencies: for different frequency sets F ref frequency f in i , construct f i Corresponding sine-cosine reference signal The construction method is as follow...

Embodiment 3

[0056] The EEG alpha wave detection and identification method based on canonical correlation analysis is the same as embodiment 1~2, the feature frequency selection described in step (5), see figure 2 ,Special attention figure 2 The last three steps in the process include the following:

[0057] (5.1) Calculation of different frequency sets F ref middle frequency f i Corresponding classification accuracy Extract the frequency f from the set of correlation coefficients Ω i Corresponding correlation coefficient, take a part of the correlation coefficient as the training sample, and another part of the correlation coefficient as the test sample, use the support vector machine (SVM) to analyze the frequency f i The corresponding correlation coefficients are trained and tested to obtain the frequency f i Corresponding classification accuracy

[0058] In this example, the frequency set F ref The number of intermediate frequencies is 26, f i Corresponding classification ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a canonical correlation analysis-based electroencephalogram alpha wave detection and identification method, and solves the problem in quick and efficient detection and identification of an electroencephalogram alpha wave signal. The method comprises the steps of inputting an electroencephalogram signal; preprocessing the electroencephalogram signal to obtain training and test data; selecting different frequency sets, and establishing reference signals corresponding to different frequencies; calculating correlation coefficients of the training data and the different frequency reference signals through canonical correlation analysis, thereby forming a correlation coefficient set; performing feature frequency selection to obtain a feature frequency set; selecting the correlation coefficients corresponding to the feature frequency set from the correlation coefficient set to form a training feature set; training a classifier by using the training feature set; and calculating a test data feature set; and performing classified identification by using the classifier, thereby finishing electroencephalogram alpha wave detection and identification. Through the featurefrequency selection, quick detection of the electroencephalogram alpha wave signal is realized; and the method is high in detection speed, high in accuracy and stable in work, and is used for signal detection in a brain computer interface system of the electroencephalogram alpha wave signal.

Description

technical field [0001] The invention belongs to the technical field of cognitive neuroscience, and mainly relates to the feature extraction and recognition detection of alpha waves in electroencephalogram signals, in particular to a method for detecting and identifying electroencephalogram alpha waves based on canonical correlation analysis. The invention can be used for rapid detection of the electroencephalogram signal α, and judges whether a segment of the electroencephalogram signal contains an α wave. Background technique [0002] Alpha wave signal is a kind of spontaneous EEG rhythm, also known as alpha rhythm. α rhythm is an electrical activity corresponding to the idle rhythm of the visual cortex in the cerebral cortex. Usually, the energy of the α wave signal is mainly concentrated in the frequency band of 8-13Hz. The intensity of the α wave signal that can be detected on the surface of the human scalp is very weak. The voltage amplitude fluctuates approximately in...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/00
CPCG06V40/10G06F2218/02G06F2218/12
Inventor 张昕王晓甜石光明王英迪李甫齐飞王永杰
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com