A new system for driver drowsiness and distraction detection

摘要: Drowsiness especially in long distance journeys is a key factor traffic accidents. In this paper new module for automatic driver drowsiness detection based on visual information and Artificial Intelligence presented. The aim of system to locate, track analyze both the driver's face eyes compute index prevent Both eye performed by Haar-like features AdaBoost classifiers. order achieve better accuracy tracking, we propose method which combination object tracking. Proposed tracking method, also has capability self correction. After region found, Local Binary Pattern (LBP) employed extract characteristics. Using these features, an SVM classifier was trained perform state analysis. To evaluate effectiveness proposed drowsy person pictured, while his EEG signals were taken. video able 100% detecting blink 98.4%. Also can calculate orientation tilt using position valuable knowledge about concentration. Finally, make decision distraction driver. Experimental results show high each section makes reliable detection.