Human-computer interaction behavior recognition method in office scene based on EOG signals

Abstract

The invention discloses a method for recognizing human-computer interaction behavior in office scenes based on EOG signals, comprising the following steps: firstly collecting EOG signals under different behavior states, and preprocessing the collected EOG signals; Detection, identifying the starting point and ending point of the effective behavior state in the filtered EOG signal; after obtaining the effective EOG signal, the signal is extracted by wavelet transform, and the extracted features are analyzed by linear discriminant analysis method. The parameters are reduced in dimension; then the features after the dimension reduction are classified by the support vector machine, and the correct behavior is identified; finally, the identified results are used to control the application software on the mobile phone. The invention has high recognition accuracy, good robustness and great application potential, can better utilize eye movement information to control external equipment, enriches the diversity of eye movement control categories, and has strong expansibility, high recognition accuracy and good prospects. advantage.

Description

technical field [0001] The invention relates to the technical field of bioelectrical signal application, in particular to a method for recognizing human-computer interaction behavior in an office scene based on an eye-electrical signal. Background technique [0002] With the rapid development of modern science and technology, human-computer interaction technology has gradually become a current research hotspot. Human-computer interaction is a bridge of communication between humans and computers, making the relationship between humans and machines more natural and harmonious. In daily life, the functions of human-computer interaction are mainly completed by external devices that can input and output such as mouse, keyboard, touchpad, etc. However, for patients with physical disabilities or disabled people, they cannot achieve these mainstream interaction methods through normal limb operations, and thereby improve their quality of life. If we can design a system that does not...

AI Technical Summary

Problems solved by technology

At the present stage, the human-computer interaction system based on oculograph is mainly used to control external devices by simply scanning up, down, left, and right of the eyeball and blinking. There are relatively few types of such control methods, and the diversity of functions is greatly restricted limitations, so we hope to expand the control category to control external devices to meet people's daily needs

Images