Skip to main navigation Skip to search Skip to main content

Video vehicle tracking based on modified GM(1,1) model

  • Chuanrong Wang*
  • , Guoyan Xu
  • , Feng Gao
  • , Jiangfeng Wang
  • *Corresponding author for this work
  • Beihang University

Research output: Contribution to journalArticlepeer-review

Abstract

A scheme is put forward of using GM(1,1) model to perform vehicle tracking with video. In view of the problem of traditional GM(1,1) model having good effect in short term, but having a large error in long-term tracking, the deviation of the predictive value from true value is taken as influencing factor, and GM(1,1) model is modified. The predicted value with improved GM(1,1) model is closer to truth one, and hence the long-term tracking effects can be assured. Finally vehicle tracking test verifies the effectiveness of improved alroithm, reducing the deviation of predicted value from the true one.

Original languageEnglish
Pages (from-to)347-350
Number of pages4
JournalQiche Gongcheng/Automotive Engineering
Volume32
Issue number4
StatePublished - Apr 2010

Keywords

  • GM(1,1) model
  • Prediction
  • Vehicle
  • Video tracking

Fingerprint

Dive into the research topics of 'Video vehicle tracking based on modified GM(1,1) model'. Together they form a unique fingerprint.

Cite this