The instant image processing application in the Intelligent Transportation System researchThe instant image processing application in the intelligent transportation system research / 即時影像處理應用在智慧型運輸系統之研究即時影像處理應用在智慧型運輸系統之研究即時影像處理應用在智慧型運輸系統之研究即時影像處理應用在智慧型運輸系統之研究

碩士 / 國立高雄應用科技大學 / 電機工程系 / 99 / According to the statistics of 99 year from Ministry of The Interior, Taiwan, ROC, the deaths in traffic accidents are about 3000 in a years. The injured people due to traffic accidents are around 260000, too. The 97% of traffic accidents are caused by erroneous driving behaviors. The traffic accidents not only damage lives but also crush families. These damages are also a heavy burden for the productivity of this country. Thus, since May 1998, the Institute of Transportation, Ministry of Transportation and Communication (MOTC) had started a project related o Advanced Vehicle Control and Safety Services, AVCSS, to develop the new technology of vehicle safety.
In this thesis, a Lane Departure Warning System (LDWS) is investigated. A Matlab-based simulator is developed to simulate the dynamics of vehicles in transportation. An image sequence captured by a video recorder installed on vehicle is manipulated through Hough Transform, RGB transform to derive the positions of lane line and vehicle tires. From these object positions, the lane departure of vehicles is obtained. IN this study, we find that the lane departure of vehicles is closely related to the tire rigidness. The experimental results show the vehicles with the high-rigidness tires will have lower lane departure .Besides, video recorders on vehicles is usable to detect the lane boundary and have a better performance in daytime. But more functions should be enhanced before the real deployments of LDWS applications to prevent vehicles from traffic accidents.

Identiferoai:union.ndltd.org:TW/099KUAS8442087
Date January 2011
CreatorsMao,Jui-Chih, 毛瑞之
ContributorsKau,Shih-Wei, Chen,Meng-Jen, 葛世偉, 陳盟仁
Source SetsNational Digital Library of Theses and Dissertations in Taiwan
Languagezh-TW
Detected LanguageEnglish
Type學位論文 ; thesis
Format75

Page generated in 0.0183 seconds