Earticle

초록

영어

A coarse-to-fine traffic sign classification algorithm is proposed. The task for traffic sign classification is to analyze the detected regions and determine the class of the sign in the region. By analyzing existing traffic sign classification algorithms, the major problem affecting the classification accuracy is pointed out. Based on this analysis, a coarse-to-fine classification algorithm is proposed. The algorithm first classifies traffic signs into several super classes, then performs class-specific shape adjustment, and finally gets the fine classification result. Experimental results show that the proposed algorithm outperforms other existing algorithms in classification accuracy, and is robust to many adverse situations.

목차

Abstract
 1. Introduction
 2. Traffic Sign Classification Algorithm Based on Support Vector Machine
  2.1. The GTSRB Data Set
  2.2. The Basic Ideas and the Whole Structure of Algorithm
  2.3. Prohibition Sign Shape Corrections Based on Mirror Symmetry Algorithm
  2.4. Correction of Warning Sign Based on Triangular Shape Detection
  2.5. HOG Color Feature Extractions
 3. Experiment Design and Discussion
  3.1. High Class Classifications
  3.2. Prohibition Sign Classifications
  3.3. Warning Sign Classification
  3.4. Classification of Indicator, Ending the Prohibition Sign and Other Signs
  3.5. Analysis of Results and Comparison
 4. Conclusion
 References

키워드

저자

• Ye Sun [ College of computer, Changchun Normal University, Jilin 130032, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Hybrid Information Technology

• 간기

  격월간

• pISSN

  1738-9968

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605