Multi-state recognition method based on forehead single-lead electroencephalogram signals

Abstract

The invention relates to a multi-state recognition method based on forehead single-lead electroencephalogram signals. The method is suitable for electroencephalogram field. According to the technical scheme, the multi-state recognition method based on the forehead single-lead electroencephalogram signals is characterized by comprising the steps that the forehead lobe electroencephalogram signals with drifting artifacts and high-frequency artifacts filtered out are obtained; solving an extreme point of the electroencephalogram signal, and when a minimum-maximum-minimum pair is detected in sequence, recording the position and amplitude of the maximum and the width between the front and rear minimum to form a feature point; carrying out rough classification on the feature points by adopting a trained classification model; when the rough classification detects that the feature points are blinking or head movement artifacts, a template matching algorithm is adopted to accurately distinguish the head movement signals from the eye electric signals. The forehead single-lead electrode is used for collecting the electroencephalogram signals at the same time, the electro-oculogram signals can be accurately recognized from the electroencephalogram signals, and the wearing comfort of the device in the detection process is improved.

Description

technical field [0001] The invention relates to a multi-state recognition method based on forehead single-lead EEG signals. Applicable to the field of EEG. Background technique [0002] EEG signals have high temporal resolution and precision, and can accurately reflect human physiological states in real time. They have been widely promoted in fatigue detection, state recognition, and human-computer interaction. EEG signals are generated by the discharge of neurons in the brain and are extremely weak. They are easily interfered by various noises during the acquisition process, such as low-frequency noise, high-frequency noise, power frequency noise, eye movement artifacts, and myoelectric artifacts. In most EEG studies, such artifacts are useless compared to the target EEG signal, and many processing algorithms have been developed to reduce the artifacts, and better results have been obtained. However, the artifact itself is also a characterization of human behavior. For ex...

AI Technical Summary

Problems solved by technology

[0003] In existing studies, such as [Ma, J., et al., A novel EOG / EEG hybrid human–machine interface adopting eye movements and ERPs: Application to robot control.], the author developed a set of The control system has achieved relatively good results, but due to the use of vertical eye electricity, the eye electricity electrodes need to be placed directly below the eyes, which affects a certain degree of aesthetics and wearing comfort

Images