Unlock instant, AI-driven research and patent intelligence for your innovation.

Cross-correlation entropy based shared space mode empty domain feature extracting method

A technology of spatial pattern and extraction method, applied in the field of signal processing, which can solve the problems of spatial filter influence, outlier influence, inappropriate spatial filter, etc.

Active Publication Date: 2017-11-21
XI AN JIAOTONG UNIV
View PDF4 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This norm can amplify the negative effects of outliers in the EEG data and lead to inappropriate spatial filters
Not only the spatial filter is affected, but the features derived from it are also affected by outliers

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
  • Cross-correlation entropy based shared space mode empty domain feature extracting method
  • Cross-correlation entropy based shared space mode empty domain feature extracting method
  • Cross-correlation entropy based shared space mode empty domain feature extracting method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will be further described below in conjunction with the accompanying drawings.

[0036] The robust common spatial patterns (CSP) algorithm (CSP-CIM) based on cross-correlation entropy induced metric (correntropy induced metric, CIM) of the present invention is divided into three parts, data preprocessing, feature extraction and classification, The specific introduction is as follows:

[0037] Suppose there are two types of EEG data, represent a class, Represents another category, c is the number of channels for data acquisition, and l is the number of sampling points for each experiment. Assume that the two types of data have N x and N ytrials, then all EEG data can be expressed as with where m=l×N x , n=l×N y , is the total number of sample points of the two types of data. These EEG data need to be preprocessed, which is divided into three steps. Suppose a certain EEG data segment is First use a bandpass filter to filter to get Z ...

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 cross-correlation entropy based shared space mode empty domain feature extracting method. Through improvement of a traditional shared space mode algorithm, robustness of outliers is improved. Cross-correlation entropy measure can estimate L2, L1 and L0 norms in different dynamic zones and thus can be used for constructing a robust cost function. By using the cross-correlation entropy measure for replacing the L2 norm in a traditional algorithm cost function, a new robust algorithm having good effect on the outliers is obtained.

Description

【Technical field】 [0001] The invention belongs to the field of signal processing, and relates to a method for extracting common spatial pattern airspace features based on cross-correlation entropy. 【Background technique】 [0002] The brain-computer interface can convert brain signals into control instructions, helping severely paralyzed patients communicate with the outside world without going through the peripheral nervous system and muscles. Electroencephalography (EEG) is a widely used signal in brain-computer interface, which has the characteristics of convenient and non-invasive acquisition. In EEG-based BCIs, a key issue is how to classify EEG signals robustly and accurately. [0003] In order to extract effective separable features from raw signals, researchers have developed many algorithms. Among them, the shared spatial pattern algorithm is a very effective method for analyzing two types of multi-channel EEG data. The algorithm can find multiple spatial domain f...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62
CPCG06F18/21G06F18/213
Inventor 陈霸东董继尧郑南宁
Owner XI AN JIAOTONG UNIV