ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper In-vehicle Camera System to Detect Approaching Vehicles from Rear Side
Cited - time in scopus Share share facebook twitter linkedin kakaostory
Authors
Masashi Yoshida, Mitsuru Ambai, Shinji Ozawa, Seong Ik Cho
Issue Date
2007-01
Citation
Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV) 2007, pp.427-430
Language
English
Type
Conference Paper
Abstract
In recent days, various in-vehicle camera systems have been proposed to detect dangerous vehicles. In this paper, we focus on the vehicles on the side lane and propose the in-vehicle camera system to detect approaching vehicles from a rear side direction. We use single gray-scale camera that has a wide field of view. This makes it possible for the input image to contain the rear area of the side lane from the position immediately lateral to the camera. The key feature of our algorithm is that we use only the vertical and horizontal edge projection to detect and track a vehicle. We can apply our algorithm to various types of vehicle, because we do not use any shape model of vehicles. Also, thankful that a projection is the one-dimensional signal, our algorithm runs quickly. We tested image sequences that include various types of vehicle and obtained an excellent result.