The Research On Driver Fatigue Detection Based On Facial Feature Recognition

With the development of China’s transportation, traffic accident occurs more and more, and fatigue driving is becoming an important reason. Therefore, the research on the method of fatigue drive detection in order to protect the safety of the driver and pedestrian or promote the construction of intelligent transportation systems has very important significance.This paper first researches the facial detection method based Haar-like feature and AdaBoost algorithm. It extracts the Haar-like feature in the sample image, and then chooses the optimal Haar-like feature through training. Then designs weak classifiers and all the weak classifiers were combined into a strong classifier by the Votes Committee mode. At last, many strong classifier were combined into a more complex system using cascade algorithm for face detection. After facial detection using AdaBoost algorithm, this paper pointed out the problem of the algorithm through analyzing the experimental result and optimize this method through adding to the Haar-like feature.After facial detection, in order to reduce the image detection effect from the changes of the driver’s position in the driver fatigue detection, a driver fatigue detection method is proposed based on Muti-scale Local Binary Pattern Histogram Fourier (MLBP-HF) and Support Vector Machine (SVM). The method includes two processions which are training and recognition. During training, we firstly extract the features of the driver’s face fatigue images which captured from videos, that is calculate the blocked image and get the MLBP histogram (MLBPH) using different scale of the uniform local binary pattern (LBP) operators, then combine them and use Discrete Fourier Transform to get the MLBP-HF features. At last, we input these features data to the SVM and train them to get its model and the parameters. During recognition, we calculate the MLBP-HF of the testing image samples, then input these MLBP-HF features to the trained SVM to detect fatigue.35figures,13tables,89references.