Share Email Print
cover

Proceedings Paper

Hyper-spectral image segmentation using spectral clustering with covariance descriptors
Author(s): Olcay Kursun; Fethullah Karabiber; Cemalettin Koc; Abdullah Bal
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

Image segmentation is an important and difficult computer vision problem. Hyper-spectral images pose even more difficulty due to their high-dimensionality. Spectral clustering (SC) is a recently popular clustering/segmentation algorithm. In general, SC lifts the data to a high dimensional space, also known as the kernel trick, then derive eigenvectors in this new space, and finally using these new dimensions partition the data into clusters. We demonstrate that SC works efficiently when combined with covariance descriptors that can be used to assess pixelwise similarities rather than in the high-dimensional Euclidean space. We present the formulations and some preliminary results of the proposed hybrid image segmentation method for hyper-spectral images.

Paper Details

Date Published: 10 February 2009
PDF: 6 pages
Proc. SPIE 7245, Image Processing: Algorithms and Systems VII, 724512 (10 February 2009); doi: 10.1117/12.811132
Show Author Affiliations
Olcay Kursun, Bahcesehir Univ. (Turkey)
Fethullah Karabiber, Istanbul Univ. (Turkey)
Cemalettin Koc, Gebze Institute of Technology (Turkey)
Abdullah Bal, Yildiz Technical Univ. (Turkey)


Published in SPIE Proceedings Vol. 7245:
Image Processing: Algorithms and Systems VII
Nasser M. Nasrabadi; Jaakko T. Astola; Karen O. Egiazarian; Syed A. Rizvi, Editor(s)

© SPIE. Terms of Use
Back to Top