A Novel Color Descriptor for Road-Sign Detection

Qieshi ZHANG; Sei-ichiro KAMATA

doi:10.1587/transfun.E96.A.971

A Novel Color Descriptor for Road-Sign Detection

Qieshi ZHANG, Sei-ichiro KAMATA

Full Text Views

0

Share
Cite this

Summary :

This paper presents a novel color descriptor based on the proposed Color Barycenter Hexagon (CBH) model for automatic Road-Sign (RS) detection. In the visual Driver Assistance System (DAS), RS detection is one of the most important factors. The system provides drivers with important information on driving safety. Different color combinations of RS indicate different functionalities; hence a robust color detector should be designed to address color changes in natural surroundings. The CBH model is constructed with barycenter distribution in the created color triangle, which represents RS colors in a more compact way. For detecting RS, the CBH model is used to segment color information at the initial step. Furthermore, a judgment process is applied to verify each RS candidate through the size, aspect ratio, and color ratio. Experimental results show that the proposed method is able to detect RS with robust, accurate performance and is invariant to light and scale in more complex surroundings.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E96-A No.5 pp.971-979

Publication Date: 2013/05/01

Publicized

Online ISSN: 1745-1337

DOI: 10.1587/transfun.E96.A.971

Type of Manuscript: PAPER

Category: Image

Cite this

Copy

Qieshi ZHANG, Sei-ichiro KAMATA, "A Novel Color Descriptor for Road-Sign Detection" in IEICE TRANSACTIONS on Fundamentals, vol. E96-A, no. 5, pp. 971-979, May 2013, doi: 10.1587/transfun.E96.A.971.
Abstract: This paper presents a novel color descriptor based on the proposed Color Barycenter Hexagon (CBH) model for automatic Road-Sign (RS) detection. In the visual Driver Assistance System (DAS), RS detection is one of the most important factors. The system provides drivers with important information on driving safety. Different color combinations of RS indicate different functionalities; hence a robust color detector should be designed to address color changes in natural surroundings. The CBH model is constructed with barycenter distribution in the created color triangle, which represents RS colors in a more compact way. For detecting RS, the CBH model is used to segment color information at the initial step. Furthermore, a judgment process is applied to verify each RS candidate through the size, aspect ratio, and color ratio. Experimental results show that the proposed method is able to detect RS with robust, accurate performance and is invariant to light and scale in more complex surroundings.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/transfun.E96.A.971/_p

Copy

@ARTICLE{e96-a_5_971,
author={Qieshi ZHANG, Sei-ichiro KAMATA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={A Novel Color Descriptor for Road-Sign Detection},
year={2013},
volume={E96-A},
number={5},
pages={971-979},
abstract={This paper presents a novel color descriptor based on the proposed Color Barycenter Hexagon (CBH) model for automatic Road-Sign (RS) detection. In the visual Driver Assistance System (DAS), RS detection is one of the most important factors. The system provides drivers with important information on driving safety. Different color combinations of RS indicate different functionalities; hence a robust color detector should be designed to address color changes in natural surroundings. The CBH model is constructed with barycenter distribution in the created color triangle, which represents RS colors in a more compact way. For detecting RS, the CBH model is used to segment color information at the initial step. Furthermore, a judgment process is applied to verify each RS candidate through the size, aspect ratio, and color ratio. Experimental results show that the proposed method is able to detect RS with robust, accurate performance and is invariant to light and scale in more complex surroundings.},
keywords={},
doi={10.1587/transfun.E96.A.971},
ISSN={1745-1337},
month={May},}

Copy

TY - JOUR
TI - A Novel Color Descriptor for Road-Sign Detection
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 971
EP - 979
AU - Qieshi ZHANG
AU - Sei-ichiro KAMATA
PY - 2013
DO - 10.1587/transfun.E96.A.971
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E96-A
IS - 5
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - May 2013
AB - This paper presents a novel color descriptor based on the proposed Color Barycenter Hexagon (CBH) model for automatic Road-Sign (RS) detection. In the visual Driver Assistance System (DAS), RS detection is one of the most important factors. The system provides drivers with important information on driving safety. Different color combinations of RS indicate different functionalities; hence a robust color detector should be designed to address color changes in natural surroundings. The CBH model is constructed with barycenter distribution in the created color triangle, which represents RS colors in a more compact way. For detecting RS, the CBH model is used to segment color information at the initial step. Furthermore, a judgment process is applied to verify each RS candidate through the size, aspect ratio, and color ratio. Experimental results show that the proposed method is able to detect RS with robust, accurate performance and is invariant to light and scale in more complex surroundings.
ER -