基於駕駛視線偵測之分心警示系統

本論文提出一套以駕駛視線偵測為基礎的分心警示系統，本系統即時偵測駕駛人視線，用以判定駕駛有無分心的駕駛行為。系統包括兩個階段：偵測階段與判定階段。偵測階段，一開始由自動式紅外線打光攝影機取得駕駛臉部區域，使用Haar-like特徵分類器偵測出人臉區域中的眼睛與嘴巴圖像，採用Canny邊緣偵測提取物件輪廓，接著利用Hough轉換對瞳孔進行定位，並提取內外眼角與左右嘴角特徵點。計算臉部方向的水平偏轉角度，與相對於駕駛臉部的視線方向，修訂以駕駛座前方為基準的視線方向。在判定階段，對前方視線為駕駛人主要任務之ㄧ的視野，分別設置九個角度計時器，當視線落在特定角度範圍內，則重置相對應的計時器，並由本文所提出的判定規則：五秒內未偵測到虹膜、凝視同一方向逾五秒則提出警告，用以判定駕駛人是否分心的狀態。當偵測出駕駛視線分心的危險行為，系統會以蜂鳴器提出警示聲，警告駕駛注意行車安全。

關鍵字

注意力偵測；視線偵測；分心駕駛；自動視覺偵測

並列摘要

In this thesis, a distraction alert system based on the detection of driver's gaze direction is proposed. This system recognize the driver's gaze direction in real-time and decide the distraction driving behavior. There are two stages in this system: detection stage and decision stage. The driver's face is captured by a camera with IR light in the beginning of detection stage. Haar-like feature classifier is used to find the eye and mouth in the face region. Contours extracted by Canny edge detection are used to locate the pupil by applying Hough transform. The corners of eye and mouth are extracted as the feature points to calculate the yaw angle of face and the gaze direction relative to face of driver. The information are used to find the final gaze direction according to the front of driver's seat. In the decision stage, nine timers are set to count the time that the gaze direction is not in the specific angle. The timer is reset when the driver is looking to the angle which is in the range of timer represented. When a timer counts 5 seconds, it is mean that the driver is not look at this angle and an alert signal is issued. The alert signal may remind driver to look this angle which has not looked for 5 seconds. Therefore, the proposed system will warn the driver to keep attention on the driving safety.

並列關鍵字

detected gazed direction ； distracted detection ； automatic visual detection ； distracted driving

參考文獻

[1] C. P. Papageorgiou, M. Oren, and T. Poggio, “A general framework for object detection,” IEEE Sixth International Conference on Computer Vision, pp. 555-562, 1998.

[2] J. Canny, “A Computational Approach to Edge Detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol8, pp.679-698, Nov. 1986.

[3] K. Guo, G. Yu, and Z. Li, “A New Rapid Algorithm for Driver Face Organ Detection,” IEEE International Automation and Logistics, pp. 1502-1506, Sep. 2008.

[4] K. Guo, G. Yu, and Z. Li, “An New Algorithm for Analyzing Driver's Attention State,” IEEE Intelligent Vehicles Symposium, pp. 21-23, 2009.

[5] L. Li, Y. Chen, and Z. Li, “Yawning Detection for Monitoring Driver Fatigue Based on Two Cameras,” IEEE International Conference on Intelligent Transportation Systems, pp. 1-6, 2009.

國際替代計量

基於駕駛視線偵測之分心警示系統

未授權

主題瀏覽