Academic
Publications
Automated Traffic Sign Detection for Modern Driver Assistance Systems

Automated Traffic Sign Detection for Modern Driver Assistance Systems,Alexander REITERER,Taher HASSAN,Naser EL-SHEIMY

Automated Traffic Sign Detection for Modern Driver Assistance Systems  
BibTex | RIS | RefWorks Download
SUMMARY Modern Driver Assistance Systems (DAS) are required to assist, guide, and control vehicles on highways and city streets based on GPS, INS and map matching. They play an important role in the navigation of modern vehicles. Although a GPS-navigation system can be updated in view of the modifications of the roads, it does not include exhaustive information about the traffic signalization. It would be useful to signal to a driver at least some important traffic signs. This paper presents the basic concept of a new approach for the automated detection of traffic signs to be incorporated in DASs. The developed procedure is based on the well known Scale Invariant Feature Transform (SIFT) algorithm. The results of extensive testing on real data sets shows that the presented approach detects over 70% of traffic signs correctly.
Published in 2010.
Cumulative Annual