Asynchronous character input method and device based on EOG

Abstract

The invention discloses an asynchronous character input method and device based on EOG. The asynchronous character input method comprises the steps that a virtual keyboard is displayed on a display, wherein there are N keys on the keyboard, the virtual keyboard flickers round by round, and in each round, all the N keys flicker once randomly; a user follows the flicking of a target key to blink when conducting spelling, oculogram signals of the user are collected in real time, oculogram data is generated within a period after one key flickers is intercepted from the collected oculogram signals, the obtained oculogram data is used as an original feature vector of the flickering key, N original feature vectors are generated in each round, an oculogram identification method is called to conduct identification on the N original feature vectors, and one result from 0 to N is output, wherein 0 represents the condition where character inputting is not met, and 1 to N correspond to N keys on the virtual keyboard. According to the asynchronous character input method and device based on the EOG, it is guaranteed that blinking of the user can be normally detected, misjudgments of non-blinking signals can be effectively eliminated at the same time, and thus the accuracy rate of character input is increased.

Description

technical field [0001] The invention belongs to the technical field of character input relying on human biological signals, in particular to an EOG-based asynchronous character input method and device. Background technique [0002] In order to allow disabled people with motor dysfunction to interact with the outside world and improve their quality of life and self-care ability, a new type of human-machine interface that uses human bioelectric signals to communicate with computers or external devices has become the frontier of research in recent years. and hotspots. The bioelectrical signals of the human body mainly include electroencephalogram (EEG, Electroencephalogram), myoelectricity (EMG, Electromyography), electrocardiogram (ECG, Electrocardiogram), and eye electricity (EOG, Electrooculogram). The human-machine interface based on bioelectrical signals hardly requires the user to perform any significant movement, that is, it can realize the interaction with the outside ...

AI Technical Summary

Problems solved by technology

First of all, one of the specific embodiments of the patent discloses keyboard input. The keyboard area composed of 7 rows and 8 columns is highlighted in a row and column, and the highlighting time is 1 second each time. It is 15 seconds. Assuming that each blink can be detected successfully, it takes an average of 7.5 seconds to input a character (the row and column do not need to be completely flashed). If a certain blink detection is unsuccessful, the time will be increased by at least 7-8 seconds. So the speed of character input is relatively slow

Secondly, this public method cannot control false alarms very well. When the device is started but the user temporarily does not want to input characters, incorrect input will also occur due to involuntary blinking.

Finally, the disclosed method is too simple for the blink detection strategy, and only performs a threshold judgment on the signal within 30 milliseconds. It is easy to generate an eye wave of more than 120μV and be misjudged as a blink, and then generate a false input, but in fact, these eye waves are significantly different from blinking in waveform

Images