Fatigue driving detection method and device

Abstract

The invention provides a fatigue driving detection method and device, and relates to the technical field of fatigue driving detection. The method comprises the steps of preprocessing a head image of adriver to obtain a grayscale image; calculating the distribution probability of the pixel values which are not zero in the grayscale image, and determining a segmentation threshold range of the grayscale image according to the distribution probability; segmenting the gray level image by adopting each segmentation threshold value in the segmentation threshold value range to obtain a target area and a background area, and respectively calculating the average relative entropy of the target area segmented by each segmentation threshold value and the total entropy of the average relative entropy of the background area; determining an optimal segmentation threshold value according to a maximum value in all total entropies and a segmentation threshold value range, carrying out threshold value segmentation on the gray level image by adopting an average relative entropy, and detecting a face region; separating the face region according to the optimal segmentation threshold, and reconstructinga face region image; calculating the blinking frequency and the yawing frequency to judge whether the driver is in fatigue driving or not. According to the invention, the detection efficiency of the face recognition can be greatly improved.

Description

technical field [0001] The invention relates to the technical field of fatigue driving detection technology, in particular to a fatigue driving detection method and device. Background technique [0002] Fatigue driving refers to the phenomenon that the driver's physiological function and psychological function are out of balance after driving the vehicle continuously for a long time, and the driving skill declines objectively. Fatigue driving can easily lead to major traffic safety accidents, so it is necessary to detect the driver's fatigue state. In the fatigue driving detection method, face area recognition is a very important part. [0003] At present, the recognition methods of face regions mainly include the whole face color image recognition method, neural network recognition method, recognition method based on illumination estimation model, maximum between-class variance method, iterative method, differential operator, cluster analysis, etc. The whole face color im...

AI Technical Summary

Problems solved by technology

The whole face color image recognition method includes pixel calculation in three dimensions of RGB, which has a large amount of calculation; although the neural network calculation recognition method has the function of self-learning, when the data is insufficient, all reasoning will be turned into numerical calculation, which will be lost. The main target information cannot work; the recognition method based on the illumination estimation model is still in the experimental research stage and can only be used in certain specific occasions

Although the methods of maximum inter-class variance method, iterative method, differential operator, and cluster analysis are relatively simple to calculate compared with the whole face color image recognition method, neural network recognition method, and recognition method based on the illumination estimation model, but The amount of calculation is still large, and the detection efficiency is not high

Images