Estimation of blind spot based on driver's view

Authors:
Kazuma Minoura;Noritaka Sekiyama;Toyohide Watanabe
Affiliations:
Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan;Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan;Nagoya University, Furo-cho, Chikusa-ku, Nagoya, Japan
Venue:
Proceedings of the 5th International Conference on Ubiquitous Information Management and Communication
Year:
2011

Citing 3
Cited 0

Foreground object detection from videos containing complex background

MULTIMEDIA '03 Proceedings of the eleventh ACM international conference on Multimedia
Maneuvering Aid for Large Vehicle using Omnidirectional Cameras

WACV '07 Proceedings of the Eighth IEEE Workshop on Applications of Computer Vision
Eliminating Blind Spots for Assisted Driving

IEEE Transactions on Intelligent Transportation Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Traffic accident is recognized as one of the most serious problems in our current society. Collisions at intersections enforce to be especially dangerous accident for human life. Some collisions are caused by driver's carelessness for finding out other vehicles in blind spots. In this paper, we propose a basic method to provide drivers with the warning information that vehicles are in blind spots. Our method estimates the blind spot, which can be generated from approximated shapes of other opposite vehicles with respect to driver's view position. In order to investigate our method successfully, we develop our prototype system of estimating blind spot and detecting moving objects by using the information about positions, shapes and moving directions of vehicles, derived from traffic monitoring movies. As a result, our prototype system could find out vehicles in blind spots for a particular driver with high accuracy.