Earticle

초록

영어

In order to meet the increase of vehicle detection items and detailed testing contents, the detection methods of automobile headlight feature points were studied. The classification ways of automobile headlights were described and a detection system of headlight feature points was designed and realized by using digital image processing technologies according to the characteristics of headlight shapes. The front end image of automobile was analyzed, the front license plate was located precisely, the headlight detection areas were pre-located, and the headlight feature points were extracted based on modified Hough transform. The experimental results showed that the regional centers of circular headlights were extracted and the corners of rectangular and alien shaped headlights were detected. This method could automatically detect the feature points of automobile headlights and lay good foundation for next vehicle testing and fault diagnosis.

목차

Abstract
 1. Introduction
 2. Headlight Feature Points Detection Based on Shape Classification
  2.1. Headlight Classification
  2.2. Detection Scheme of Headlight Feature Points
 3. Implementation of Headlight Feature Points Detection System
  3.1. Basic Idea of Hough Transform
  3.2. Location of Front License Plate
  3.3. Feature Point Detection of Circular Headlights
  3.4. Feature Point Detection of Corner Type
  3.5. Experiment Results
 4. Conclusions
 Acknowledgments
 References

키워드

저자

• Peijiang Chen [ College of Mechanical and Electronic Engineering, Nanjing Forestry University, Jiangsu, 210037, China, School of Automobile, Linyi University, Shandong, 276000, China ]
• Yongjun Min [ College of Mechanical and Electronic Engineering, Nanjing Forestry University, Jiangsu, 210037, China ] Corresponding Author

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Control and Automation

• 간기

  월간

• pISSN

  2005-4297

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605