A NEW SEGMENTATION TECHNIQUE FOR CLASSIFICATION OF MOVING VEHICLES

In this paper, the authors present a new image processing technique for classifying moving vehicles. The technique is based on color motion segmentation and split-merge segmentation approaches. The segmentation process is handled in two steps and does not require processing the whole image, thus saving computation time. An adaptive threshold is used to automatically choose the threshold value for split-merge method. In addition, the authors classify the vehicle into four categories based on contour feature. Experimental results show the high potential for this new technique.

• Supplemental Notes:
  • Publication Date: 2000 IEEE Service Center, Piscataway NJ
• Corporate Authors:

  Kyoto Daigaku

  ,    

  Nanyang Technological University

  School of Electrical and Electronic Engineering
  ,   Singapore 

  Saitama Daigaku

  ,    

  Tohoku Daigaku

  ,    

  Chonbuk National University

  ,    

  Anritsu Denki Kabushiki Kaisha

  ,    

  NTT Network Innovation Laboratories

  ,    

  Yuseisho Tsushin Sogo Kenkyujo

  ,    

  Instituto Superior de Engenharia do Porto

  ,    

  Fukuyama Daigaku

  ,    
• Authors:
  • Zhang, Chunrui
  • SIYAL, M Y
• Conference:
  • IEEE Vehicular Technology Conference (51st : 2000). VTC2000-Spring ... proceedings
  • Location: Tokyo , Japan
• Publication Date: 2000

Language

• English

Media Info

• Pagination: p. 323-326
• Serial:
  • Volume: 1

Subject/Index Terms

• TRT Terms: Automatic vehicle classification; Image processing

Filing Info

• Accession Number: 00801332
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: PATH
• Created Date: Nov 7 2000 12:00AM