Mental Fatigue Detection Based on Multi-Inter-Domain Optical Flow Characteristics

The expression of chronic fatigue for the human body has the characteristics of long duration and small inter-frame displacement, which makes it difficult to identify quickly and accurately. According to the calculation method of phase, energy and feature matching, the gradient-based optical flow method is simple and the better results are obtained. Therefore, the global optical flow technique based on gradient is studied by using global optical flow technique. Continuous multi-frame between the local area of the calculation, to obtain a dense optical flow, to be separated from the frame of the two images of the movement changes, to solve the long duration and inter-frame small displacement problem. The optical flow characteristics are extracted by calculating the point motion of the eye area and the mouth region. First, the optical value of the optical flow is used to extract the model of the threshold value, and then the time of the yawning is extracted. Then, Closure rate and yaw frequency. Through the experiment, the above method can accurately identify the fatigue expression.