Share Email Print

Proceedings Paper

Zenith angle-based method for pattern recognition of landform elements using feature vector matching
Author(s): Yanlan Wu; Yongqiong Liu; Hai Hu; Chuanyong Yang
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Pattern recognition of landform elements provides fundamental information for landscape research such as landscape evaluation and hazard prediction. Totally different from the existing methods where surface geometrical forms are commonly described by local curvatures, this paper uses zenith angle as a basis for pattern recognition in topography. One property of zenith angle is a regional morphmetrical variable, which has potential to overcoming the lack of a consideration of the center point a part of the regional terrain in curvature-based methods. Moreover, the zenith angles are converted and stored in the form of a feature vector so that a feature vector matching approach can be applied to implement the pattern recognition. The proposed method has been implemented and applied to a 10 m cell size DEM of GISMAP Terrain (issued by Hokkaido-chizu in Japan) as a test case. Through visual observations, the transparent composite map of the classification map and the shaded relief of DEM shows that the recognized patterns match the relief well. The topography profiles also reveals that the results of pattern recognition are compatible with the concave-convex property of the terrain shape. Moreover, the 3D maps of a local terrain data randomly taken from the DEM for each of the feature matching cases show that the recognized pattern of the central point is matched well with the characteristics of the surrounding topography.

Paper Details

Date Published: 16 October 2009
PDF: 10 pages
Proc. SPIE 7492, International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining, 74923H (16 October 2009); doi: 10.1117/12.838663
Show Author Affiliations
Yanlan Wu, Wuhan Univ. (China)
Yongqiong Liu, Wuhan Univ. (China)
Hai Hu, Wuhan Univ. (China)
Chuanyong Yang, Foshan Urban Planning Design and Surveying Research Institute (China)

Published in SPIE Proceedings Vol. 7492:
International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining
Yaolin Liu; Xinming Tang, Editor(s)

© SPIE. Terms of Use
Back to Top