Multi-spectral image analysis and NURBS-based 3D visualization of land maps

  • Authors:

  • Dilshan Jayathilaka;Ravinda G. N. Meegama

  • Affiliations:

  • Centre for Computer Studies, Sabaragamuwa University, Srilanka;Centre for Computer Studies, Sabaragamuwa University, Srilanka

  • Venue:
  • ACOS'06 Proceedings of the 5th WSEAS international conference on Applied computer science
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a novel technique to visualize in 3D objects extracted from multi-spectral satellite images. At the beginning, a thematic map of the original image is obtained by assessing the quality and correcting geometric errors. This map is then classified to identify the features of interest. A Non-Uniform Rational B-Spline (NURBS) surface is generated from this extracted feature in order to visualize the object in 3D. The results indicate an overall classification accuracy of eighty percent.