This paper presents a robust approach to classify the driver state in a novel advanced driver assistance system named ShadeVision, which aims to improve the driving safety and comfort by avoiding the dazzling effect. Different from the existing algorithms for driver state classification, our proposed method is capable of identifying the "dazzled state" of the driver, which can function as a supplement for the regular dazzling detection algorithm, particularly at dawn and dusk. To this end, the driver's head is captured by a high-speed camera mounted in the vehicle. After segmenting the eye region, we design a new feature vector of the eye to classify the eye state in a single frame. Afterwards, the driver states are classified based on the eye states in successive frames. Finally, on-road tests are carried out to verify the effectiveness of the proposed algorithm.
Robust vision-based driver state classification in the dazzling avoidance system
2016-07-01
436570 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
ROBUST DAZZLING DETECTION IN A NOVEL VISUAL BASED DAZZLING AVOIDANCE SYSTEM
British Library Conference Proceedings | 2014
|Online Contents | 1997
|