RT info:eu-repo/semantics/article
T1 Perception advances in outdoor vehicle detection for automatic cruise control
A1 Álvarez, S.
A1 Sotelo Vázquez, Miguel Ángel
A1 Ocaña Miguel, Manuel
A1 Fernández Llorca, David
A1 Parra Alonso, Ignacio
A1 Bergasa Pascual, Luis Miguel
K1 Vision
K1 Vehicle detection
K1 Automatic cruise control
K1 SVM (Support Vector Machine)
K1 Intelligent transportation systems
K1 Electrónica
K1 Electronics
AB This paper describes a vehicle detection system basedon support vector machine (SVM) and monocular vision.The final goal is to provide vehicle-to-vehicle time gapfor automatic cruise control (ACC) applications in theframework of intelligent transportation systems (ITS). Thechallenge is to use a single camera as input, in order toachieve a low cost final system that meets the requirementsneeded to undertake serial production in automotive industry.The basic feature of the detected objects are first located inthe image using vision and then combined with a SVMbased classifier. An intelligent learning approach is proposedin order to better deal with objects variability, illuminationconditions, partial occlusions and rotations. A large databasecontaining thousands of object examples extracted from realroad scenes has been created for learning purposes. Theclassifier is trained using SVM in order to be able to classifyvehicles, including trucks. In addition, the vehicle detectionsystem described in this paper provides early detection ofpassing cars and assigns lane to target vehicles. In the paper,we present and discuss the results achieved up to date in realtraffic conditions.
PB Cambridge University Press
SN 0263-5747
YR 2010
FD 2010-09-04
LK http://hdl.handle.net/10017/45887
UL http://hdl.handle.net/10017/45887
LA eng
NO Ministerio de Educación y Ciencia
DS MINDS@UW
RD 29-abr-2024