Generation of Training Data by Degradation Models for Traffic Sign Symbol Recognition

Hiroyuki ISHIDA; Tomokazu TAKAHASHI; Ichiro IDE; Yoshito MEKADA; Hiroshi MURASE

doi:10.1093/ietisy/e90-d.8.1134

Generation of Training Data by Degradation Models for Traffic Sign Symbol Recognition

Hiroyuki ISHIDA, Tomokazu TAKAHASHI, Ichiro IDE, Yoshito MEKADA, Hiroshi MURASE

Full Text Views

0

Share
Cite this

Summary :

We present a novel training method for recognizing traffic sign symbols. The symbol images captured by a car-mounted camera suffer from various forms of image degradation. To cope with degradations, similarly degraded images should be used as training data. Our method artificially generates such training data from original templates of traffic sign symbols. Degradation models and a GA-based algorithm that simulates actual captured images are established. The proposed method enables us to obtain training data of all categories without exhaustively collecting them. Experimental results show the effectiveness of the proposed method for traffic sign symbol recognition.

Publication: IEICE TRANSACTIONS on Information Vol.E90-D No.8 pp.1134-1141

Publication Date: 2007/08/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1093/ietisy/e90-d.8.1134

Type of Manuscript: Special Section PAPER (Special Section on Image Recognition and Understanding)

Category

Cite this

Copy

Hiroyuki ISHIDA, Tomokazu TAKAHASHI, Ichiro IDE, Yoshito MEKADA, Hiroshi MURASE, "Generation of Training Data by Degradation Models for Traffic Sign Symbol Recognition" in IEICE TRANSACTIONS on Information, vol. E90-D, no. 8, pp. 1134-1141, August 2007, doi: 10.1093/ietisy/e90-d.8.1134.
Abstract: We present a novel training method for recognizing traffic sign symbols. The symbol images captured by a car-mounted camera suffer from various forms of image degradation. To cope with degradations, similarly degraded images should be used as training data. Our method artificially generates such training data from original templates of traffic sign symbols. Degradation models and a GA-based algorithm that simulates actual captured images are established. The proposed method enables us to obtain training data of all categories without exhaustively collecting them. Experimental results show the effectiveness of the proposed method for traffic sign symbol recognition.
URL: https://globals.ieice.org/en_transactions/information/10.1093/ietisy/e90-d.8.1134/_p

Copy

@ARTICLE{e90-d_8_1134,
author={Hiroyuki ISHIDA, Tomokazu TAKAHASHI, Ichiro IDE, Yoshito MEKADA, Hiroshi MURASE, },
journal={IEICE TRANSACTIONS on Information},
title={Generation of Training Data by Degradation Models for Traffic Sign Symbol Recognition},
year={2007},
volume={E90-D},
number={8},
pages={1134-1141},
abstract={We present a novel training method for recognizing traffic sign symbols. The symbol images captured by a car-mounted camera suffer from various forms of image degradation. To cope with degradations, similarly degraded images should be used as training data. Our method artificially generates such training data from original templates of traffic sign symbols. Degradation models and a GA-based algorithm that simulates actual captured images are established. The proposed method enables us to obtain training data of all categories without exhaustively collecting them. Experimental results show the effectiveness of the proposed method for traffic sign symbol recognition.},
keywords={},
doi={10.1093/ietisy/e90-d.8.1134},
ISSN={1745-1361},
month={August},}

Copy

TY - JOUR
TI - Generation of Training Data by Degradation Models for Traffic Sign Symbol Recognition
T2 - IEICE TRANSACTIONS on Information
SP - 1134
EP - 1141
AU - Hiroyuki ISHIDA
AU - Tomokazu TAKAHASHI
AU - Ichiro IDE
AU - Yoshito MEKADA
AU - Hiroshi MURASE
PY - 2007
DO - 10.1093/ietisy/e90-d.8.1134
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E90-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2007
AB - We present a novel training method for recognizing traffic sign symbols. The symbol images captured by a car-mounted camera suffer from various forms of image degradation. To cope with degradations, similarly degraded images should be used as training data. Our method artificially generates such training data from original templates of traffic sign symbols. Degradation models and a GA-based algorithm that simulates actual captured images are established. The proposed method enables us to obtain training data of all categories without exhaustively collecting them. Experimental results show the effectiveness of the proposed method for traffic sign symbol recognition.
ER -