Method and Apparatus for Driver Drowsiness Detection with optimized pre-processing for eyelid-closure classification using SVM

摘要

A method and apparatus for detecting driver's drowsiness through pre-processing optimized for eye blink classification using SVM are presented. The method for detecting driver's drowsiness through pre-processing optimized for eye blink classification using SVM proposed in the present invention includes the steps of detecting a driver's face image using a real-time camera or an offline video stream; Pre-processing using a gram-polynomial-based function, extracting features using a feature extractor for the preprocessed driver's face image, and learning through a classifier using the extracted features, and the state of the eyes and detecting drowsiness by determining the