Electroencephalogram fatigue detection method based on sample and feature quality combined quantitative evaluation

Abstract

The invention provides an electroencephalogram fatigue detection method based on sample and feature quality combined quantitative evaluation. The method comprises the following steps that 1, carrying out electroencephalogram data acquisition by multiple testees under a simulated driving system; 2, carrying out preprocessing and feature extraction on all electroencephalogram data obtained in the step 1; 3, establishing a machine learning model to realize electroencephalogram fatigue detection based on sample and feature quality combined quantitative evaluation; 4, calculating a description factor v for measuring the quality of samples and a description factor theta of features; and 5, performing fatigue regression prediction on new electroencephalogram data of the testees. According to the method, after v and theta are embedded into a least square model, the obtained weight description factors for measuring the sample quality and the features provide an effective tool for executing electroencephalogram data sample selection and feature selection, the samples and features with relatively good quality are endowed with higher weights, and the fatigue conditions of the testees can be accurately obtained according to the electroencephalogram data.

Description

technical field [0001] The invention belongs to the technical field of EEG signal processing, and in particular relates to an EEG fatigue detection method for joint quantitative evaluation of samples and feature quality. Background technique [0002] With the increasing development of my country's traffic, it is of great significance to study more practical and objective driving fatigue detection to improve active traffic safety. Comprehensive fatigue detection methods at home and abroad are mainly divided into two types: subjective evaluation and objective detection. The subjective evaluation method is mainly to judge whether you are in a state of fatigue by recording subjective questionnaires such as the Pearson Fatigue Scale and the Stanford Sleep Scale. The field of objective detection is mainly divided into: detection based on vehicle behavior characteristics, detection based on driver characteristic behavior characteristics, and detection based on physiological electr...

AI Technical Summary

Problems solved by technology

However, due to the influence of cross-period acquisition and the position of the electrode cap in the current EEG acquisition process, the quality of the collected EEG data lacks reliability.

Images