2D Line Matching Using Geometric and Intensity Data

  • Authors:
  • Dong-Min Woo;Dong-Chul Park;Seung-Soo Han;Seunghwa Beack

  • Affiliations:
  • -;-;-;-

  • Venue:
  • AICI '09 Proceedings of the 2009 International Conference on Artificial Intelligence and Computational Intelligence - Volume 03
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a stereo matching method of 2D line segments for the detection of 3D line segment. Stereo line matching is a difficult task, since it relies on incomplete 2D line data. The proposed matching method is based on geometric and intensity information of 2D line segment. A set of line attributes is used to match a pair of 2D lines. Line attributes consist of intensity information and geometric information including length, orientation, and endpoint position. Multi-threshold technique has been employed in the implementation of the proposed method so that flexible and efficient matching capability can be attained. Aerial and satellite images are used for the experiments of the proposed method. The experimental results indicate that the proposed method generates accurate 3D line segments from aerial and satellite images.