Abstract | ||
---|---|---|
We describe a system that controls whether the headlights of a vehicle are in the highbeam or lowbeam state based on input from a forward looking video camera. The core of the system relies on conventional computer vision techniques, albeit with a sophisticated spot finder front-end. Despite this architecture we are able to use an automated supervised learning technique to tune the system to yield high performance. Using a customer-imposed metric we present both in-car and off-line results from our system along with several competitors, and investigate the system's performance under different weather conditions. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/IVS.2011.5940492 | Intelligent Vehicles Symposium |
Keywords | Field | DocType |
computer vision,control engineering computing,intelligent control,learning (artificial intelligence),lighting control,robust control,traffic engineering computing,video cameras,automated supervised learning technique,computer vision techniques,customer-imposed metric,forward looking video camera,highbeam state light,lowbeam state light,robust intelligent vehicle headlight controller,sophisticated spot finder front- end | Intelligent control,Computer vision,Architecture,Control theory,Supervised learning,Artificial intelligence,Engineering,Video camera,Robust control | Conference |
ISSN | ISBN | Citations |
1931-0587 | 978-1-4577-0890-9 | 2 |
PageRank | References | Authors |
0.53 | 7 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jonathan H. Connell | 1 | 712 | 60.10 |
Benjamin W. Herta | 2 | 2 | 0.53 |
Sharath Pankanti | 3 | 3542 | 292.65 |
Holger Hess | 4 | 2 | 0.53 |
Sebastian Pliefke | 5 | 2 | 0.53 |