Low-resolution vehicle image recognition technology by frame composition of moving images

Yusuke Kanzawa, Hiroki Kobayashi, Takenao Ohkawa, Toshio Ito

Research output: Contribution to journalArticle

Abstract

Developing on-board automotive driver assistance systems intended to alert drivers about driving environments, and about possible collisions with other vehicles, has been attracting considerable attention. In particular, many researchers have suggested forward vehicle recognition technology using a camera installed in the vehicle. In forward vehicle recognition, however, it is difficult to detect the features of a vehicle from a distant vehicle image by conventional methods, because the image is too low-resolution (LR). This paper presents a vehicle image recognition technology for detecting the features of a distant vehicle by frame composition of moving images. To detect the vehicle features from a distant LR vehicle image, we use moving images obtained from a camera installed on the vehicle, and utilize super-resolution (SR) image reconstruction. SR image reconstruction consists of using signal processing techniques to obtain a high-resolution (or sequence) image from observed multiple LR images. The use of this technique on real road images shows the effectiveness of the proposed techniques.

Original languageEnglish
Pages (from-to)1-7
Number of pages7
JournalElectronics and Communications in Japan
Volume93
Issue number3
DOIs
Publication statusPublished - 2010 Mar
Externally publishedYes

Keywords

  • Computer vision
  • Frame composition
  • Super-resolution
  • Vehicle recognition

ASJC Scopus subject areas

  • Signal Processing
  • Physics and Astronomy(all)
  • Computer Networks and Communications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Low-resolution vehicle image recognition technology by frame composition of moving images'. Together they form a unique fingerprint.

  • Cite this