Tracking Road Centerlines from High Resolution Remote Sensing Images by Least Squares Correlation Matching

Feature extraction from remotely sensed images is an important component of geographic information systems (GIS). This article describes a method of tracking road centerlines from high resolution remote sensing images by least squares correlation matching. The authors report on the use of a semi-automatic algorithm for tracking road centerlines from satellite images at 1 m resolution. The study assumes that road centerlines are visible in the image and that among points on road centerlines similarity transformation holds. The authors contend that their least squares correlation matching algorithm works by defining a template around a user-given input point (which must lie on a road centerline) and then by matching the template against the image along the orientation of the road under consideration. Once matching succeeds, a new match proceeds by shifting a matched target window further along road orientation. By repeating the process above, one can obtain a series of points, which lie on a road centerline successively. The authors report on the use of this method with an Ikonos image over Seoul, Korea.

  • Availability:
  • Authors:
    • Kim, Taejung
    • Park, Seung-Ran
    • Kim, Moon-Gyu
    • Jeong, Soo
    • Kim, Kyung-Ok
  • Publication Date: 2004-12


  • English

Media Info

  • Media Type: Print
  • Features: Figures; Photos; References;
  • Pagination: pp 1417-1422
  • Serial:

Subject/Index Terms

Filing Info

  • Accession Number: 01002038
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jul 18 2005 7:17AM