SSVEP-based classification method, system and device and readable storage medium

Abstract

The invention provides an SSVEP-based classification method, system and device and a readable storage medium. The classification method comprises the following steps: providing stimulation of different frequencies for a testee, wherein each frequency corresponds to one classification; acquiring SSVEP signals of the testee in real time; calculating a time node according to a preset rule, when the collection time reaches the time node, analyzing the SSVEP signals collected at the time node, and obtaining a probability value that the testee responds to different stimulation frequencies; judging whether the maximum value of the probability value reaches a preset probability threshold value or not; if yes, outputting the maximum frequency value and the corresponding classification; or else, returning and continuing to acquire the SSVEP signals of the testee. Therefore, when the signal-to-noise ratio is high due to strong frequency stimulation or high induction degree, the classification result with high accuracy can be output in a short time, and when the signal-to-noise ratio is low due to weak frequency stimulation or low induction degree, the accuracy of the output classification result can be ensured by acquiring the SSVEP signals for a longer time.

Description

technical field [0001] The present application relates to the field of artificial intelligence computing, and in particular, to a classification method, system, device and readable storage medium based on SSVEP. Background technique [0002] When the human eye looks at a certain frequency of flickering stimuli, it will induce SSVEP (Steady-State Visual Evoked Potentials, steady-state visual evoked potentials) of the same frequency. SSVEP can be used to determine which stimuli the subject is looking at among stimuli of different frequencies. , so as to achieve target selection. [0003] Most of the current research directly uses a fixed-length sliding window to intercept the signal, calculates the characteristics of the signal in the window, and classifies it according to the signal characteristics. Each category corresponds to a stimulus target with a response frequency. [0004] Since the signal-to-noise ratio of the SSVEP signal is affected by various factors, different s...

AI Technical Summary

Problems solved by technology

In short, using a fixed-length sliding window, the window length will affect the classification accuracy and real-time performance. If the window length is too small, the analysis accuracy and accuracy will decrease. If the window length is too large, the delay will be greater. In some scenarios that require real-time feedback poor experience

Images