https://scholars.lib.ntu.edu.tw/handle/123456789/393016
標題: | A Vision-Based Hierarchical Framework for Autonomous Front-Vehicle Taillights Detection and Signal Recognition | 作者: | Cui, Z. Yang, S.-W. HSIN-MU TSAI |
公開日期: | 2015 | 卷: | 2015-October | 起(迄)頁: | 931-937 | 來源出版物: | IEEE Conference on Intelligent Transportation Systems | 摘要: | Automatically recognizing rear light signals of front vehicles can significantly improve driving safety by automatic alarm and taking actions proactively to prevent rear-end collisions and accidents. Much previous research only focuses on detecting brake signals at night. In this paper, we present the design and implementation of a robust hierarchical framework for detecting taillights of vehicles and estimating alert signals (turning and braking) in the daytime. The three-layer structure of the vision-based framework can obviously reduce both false positives and false negatives of taillight detection. Comparing to other existing work addressing nighttime detection, the proposed method is capable of recognizing taillight signals under different illumination circumstances. By carrying out contrast experiments with existing state-of-the-art methods, the results show the high detection rate of the framework in different weather conditions during the daytime. © 2015 IEEE. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-84950299080&doi=10.1109%2fITSC.2015.156&partnerID=40&md5=bf0b3f9de1758238f328c0158c5d5053 http://scholars.lib.ntu.edu.tw/handle/123456789/393016 |
DOI: | 10.1109/ITSC.2015.156 | SDG/關鍵字: | Accidents; Intelligent systems; Intelligent vehicle highway systems; Signal processing; Transportation; Vehicles; Contrast experiment; Design and implementations; High detection rate; Rear-end collisions; Signal recognition; State-of-the-art methods; Taillight detection; Three-layer structures; Signal detection |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。