A fatigue detection method based on eye movement
A technology of fatigue detection and eye movement, which is applied in the field of fatigue detection, can solve the problems of not being able to accurately detect whether it is in a fatigue state or not, and achieve the effect of reducing performance requirements and accurately expressing fatigue
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0041] The present invention provides a method for detecting fatigue based on eye movement, the steps comprising:
[0042] S1. Establish an initial classification model, including two states of wakefulness and fatigue;
[0043] It is understandable that the initial classification model was established by conducting a mental rotation eye movement experiment and an event scene experiment on the subjects.
[0044] Specifically, mental rotational eye-tracking experiments such as figure 1 As shown, a three-factor within-subject repeated-measures experimental design of 2 (figure pair type) × 5 (rotation angle) × 2 (fatigue state) was adopted, and one of the factors was the figure pair type, which was divided into plane pairs (one of the figures in the plane Rotate up a certain number of degrees), mirror pair (two different graphics) two levels. Another factor is the rotation angle of the graphic, which is divided into clockwise 0°, 45°, 90°, 135°, and 180°. The fatigue state is d...
Embodiment 2
[0090] The difference between this embodiment and Embodiment 1 is that in the step S3, the new sequence of eye movements is constructed, such as figure 2 and image 3 As shown in the figure, the collected blink event sequence is converted, and the blink frequency input sequence, the blink average duration sequence and the blink duration sequence are established respectively,
[0091] Among them, the blink frequency input sequence is constructed with a sampling interval of 6s.
[0092] Among them, the average blink duration sequence defines the sampling interval of 6s.
[0093] Among them, the blink duration sequence defines the sampling interval of 6s.
Embodiment 3
[0095] The difference between this embodiment and Embodiment 1 is that in the step S4, the blink frequency input sequence, the blink average duration sequence, and the blink duration sequence are respectively subjected to Fourier transform, and then the Fourier transform of each frequency point in the low frequency range is calculated. Amplitude and.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


