An approach for moving object recognition based on BPR and CI

  • Authors:

  • Li Wang;Lida Xu;Renjing Liu;Hai Hong Wang

  • Affiliations:

  • School of Economics and Management, Beihang University, Beijing, China 100083;College of Economics and Management, Beijing Jiaotong University, Beijing, China 100044 and Department of Information Technology and Decision Sciences, Old Dominion University, Norfolk, USA 23529 ...;School of Management, Xian Jiaotong University, Xian, China 710049;College of Information Science and Technology, Qingdao University of Science and Technology, Qingdao, China 266100

  • Venue:
  • Information Systems Frontiers
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

A recognition and classification method of multiple moving objects in traffic based on the combination of the Biomimetic Pattern Recognition (BPR) and Choquet Integral (CI) is proposed. The recognition process consists of three stages. At the first stage, vehicles and pedestrians are detected in video images and the area, the shape and the velocity features are obtained by classical methods. At the second stage, BPR is used to classify the Zernike moments extracted at the first stage. At the last stage, CI is then adopted for multi-features fusion based on the output of BPR, and the area and the velocity features obtained at the first stage to improve the recognition accuracy. Experiment results show that this approach is efficient.