利用手機即時偵測車輛周圍的侵犯駕駛行為

A Phone-based System to Measure Driving Aggressiveness around Vehicles

謝宜軒(Yi-Hsuan Hsieh)

指導教授 ： 朱浩華

共同指導教授 ： 黃寶儀(Polly Huang)

國立臺灣大學/電機資訊學院/資訊工程學研究所/碩士(2013年)

https://doi.org/10.6342/NTU.2013.01161

摘要

「侵犯駕駛」事件對道路安全構成莫大的威脅，本篇論文提出一套手機應用系統來即時偵測並警示駕駛前後的「侵犯駕駛行為」以防止車禍的發生。本系統利用裝置於前方擋風玻璃的對外相機上的後置鏡頭以及裝於車後的後視相機作為感測器，並透過4-core的智慧型手機來處理影像即偵測駕駛周圍重要的4個「侵犯駕駛」事件。本論文設計、實作和描述了一個手機應用系統，實驗涵蓋19個不同的駕駛，總共駕駛226.3公里，共計376個外部 「侵犯駕駛」事件，實驗結果平均precision 為85% 而 recall 為84%。另外本論文亦展示了4-core的智慧型手機可以有效的平均只花175毫秒的處理時間來完成運算，因此可以有效的給予駕駛即時危險警示。

關鍵字

侵犯性駕駛 ； 手機系統 ； 近於即時系統 ； 影像處理

並列摘要

Aggressive driving has become one serious threat to road safety. This thesis pre-sents a near real-time phone-based system that detects and informs drivers to avoid car accidents caused by front and rear aggressive driving events. The rear camera of a smartphone and a rearview camera suite are deployed as vision sensors to capture front and rear images of a vehicle. The techniques of image processing are adopted on a smartphone to detect 4 significant aggressive driving events near our vehicles. We de-sign and implement the phone-based system, and conduct a real world experiment in-volving 19 different participates and 226.3 km of driving trips, and 376 aggressive driving events. The average precision and recall of identifying aggressive driving events are 85% and 84% respectively. We also show that the processing speed of our system is efficient on a 4-core smartphone with low 175 ms latency, so it enables instant feedback to drivers.

並列關鍵字

aggressive driving ； phone-based system ； near real-time system ； image process

參考文獻

3. Coroama, V. The Smart Tachograph – Individual Accounting of Traffic Costs and its Implications. Proc. Pervasive 2006, Springer (2006), pp. 135-152.

7. Kanhere, N.K., Birchfield, S.T. A Taxonomy and Analysis of Camera Calibration Methods For Traffic Monitoring Applications. Trans. Intell. Transport. Sys. 11, 2 (June 2010), 441-452.

8. Wang, T., Cardone, G., Corradi, A., Torresani, L., and Campbell, A.T. WalkSafe: a Pedestrian Safety App for Mobile Phone Users Who Walk and Talk While Crossing Roads. Proc. HotMobile 2012, ACM Press (2012), 5-6.

11. Chapter 3: Functional Classification.

12. Salvucci, D. D. and Liu, A. The Time Course of a Lane Change: Driver Control And Eye-Movement Behavior. Transportation Research Part F: Traffic Psychology and Behaviour 5, 2(2002), 123 – 132.