Study on Detection and Localization Algorithm of Traffic Signs from Natural Scenes

Automatic detection and location of traffic signs is an important part of intelligent transportation, especially for unmanned vehicle technology research. For the morphological feature of China road traffic signs, we propose a traffic sign detection method based on color segmentation and shape analy...

Full description

Saved in:
Bibliographic Details
Published in:Advances in Mechanical Engineering Vol. 2014; pp. 319 - 324
Main Authors: Han, Xian-Zhong, Chen, Chen, Wang, Ke-Jian, Yuan, Yingchun, Song, Yulong
Format: Journal Article
Language:English
Published: London, England Hindawi Limiteds 01.01.2014
SAGE Publications
Sage Publications Ltd. (UK)
Sage Publications Ltd
SAGE Publishing
Subjects:
Algorithms
Analysis
Automation
Color
Mechanical engineering
Methods
Optical illusions
Pattern recognition (Computers)
Position (location)
Retinex (algorithm)
Road signs
Segmentation
Signs
Street signs
Traffic accidents & safety
Traffic signs
China
ISSN:1687-8132, 1687-8140, 1687-8140, 1687-8132
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Automatic detection and location of traffic signs is an important part of intelligent transportation, especially for unmanned vehicle technology research. For the morphological feature of China road traffic signs, we propose a traffic sign detection method based on color segmentation and shape analysis. Firstly, in order to solve the problems of traffic signs color cast, distortion, and cross-color in natural scenes, the images are processed by white balance, Retinex color enhancement, and affine transformation. Then, the type of traffic signs is discriminated and detected, according to the color and shape characteristics of traffic signs. The experimental results show that this method can effectively detect and recognize traffic signs.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1687-8132
1687-8140
1687-8140
1687-8132
DOI:10.1155/2014/318635