Presentation 2014-02-18
Ground Truth Generation based on Spatio-temporal Properties of Lane Markers
Jun SHIWAKU, Hiroki TAKAHASHI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Automobile safeties have developed rapidly to prevent traffic accidents. Many lane detection systems are investigated. Those accuracy are evaluated based on Ground Truth. In this paper, an efficient Ground Truth generation method for reducing manual operation is proposed. First, time slice images are obtained from an in-vehicle video. Next, meanderings of the vehicle against a lane are canceled by minimizing SSD of adjacent rows in the nearest time-slice image of the vehicle. Then, lane markers in time slice images are extracted by snakes of active contour extracting method, and lane markers in all time slice images are extracted by propagating lane markers information from the bottom time slice image to the upper one. Ground Truth is generated with contour information of lane markers. Finally, accuracy evaluation of Ground Truth. As a result, an error of straight road is 1.4401pixel, curve road is 0.6551pixel.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Lane detection / Ground Truth / Snakes
Paper # ITS2013-51,IE2013-116
Date of Issue

Conference Information
Committee IE
Conference Date 2014/2/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Image Engineering (IE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Ground Truth Generation based on Spatio-temporal Properties of Lane Markers
Sub Title (in English)
Keyword(1) Lane detection
Keyword(2) Ground Truth
Keyword(3) Snakes
1st Author's Name Jun SHIWAKU
1st Author's Affiliation Graduate School of Informatics & Enginnering, the University of Electro-Communications()
2nd Author's Name Hiroki TAKAHASHI
2nd Author's Affiliation Graduate School of Informatics & Enginnering, the University of Electro-Communications
Date 2014-02-18
Paper # ITS2013-51,IE2013-116
Volume (vol) vol.113
Number (no) 434
Page pp.pp.-
#Pages 6
Date of Issue