Binary tree of posterior probability support vector machines for hyperspectral image classification

Wang, Dongli; Zhou, Yan; Zheng, Jianguo

doi:doi:10.1117/1.3553800

GENERAL INFORMATION

You are not logged in Logged Out Log In

Journal of Applied Remote Sensing / Volume 5 / Research Papers

Previous Article | Next Article

Binary tree of posterior probability support vector machines for hyperspectral image classification

J. Appl. Remote Sens. 5, 053503 (Mar 11, 2011); http://dx.doi.org/10.1117/1.3553800

Alerts
- Alert Me When Cited
- Alert Me When Corrected
Tools
Share
- Email Abstract
- Connotea
- CiteULike
- del.icio.us
- BibSonomy
- Tweet this Article
- Add to Facebook

Abstract

References (27)

Dongli Wang and Yan Zhou

Xiangtan University, College of Information Engineering, Xiangtan 411105, China

Jianguo Zheng

Donghua University, Glorious Sun School, Shanghai 200051, China

The problem of hyperspectral remote sensing images classification is revisited by posterior probability support vector machines (PPSVMs). To address the multiclass classification problem, PPSVMs are extended using binary tree structure and boosting with the Fisher ratio as class separability measure. The class pair with larger Fisher ratio separability measure is separated at upper nodes of the binary tree to optimize the structure of the tree and improve the classification accuracy. Two approaches are proposed to select the class pair and construct the binary tree. One is the so-called some-against-rest binary tree of PPSVMs (SBT), in which some classes are separated from the remaining classes at each node considering the Fisher ratio separability measure. For the other approach, named one-against-rest binary tree of PPSVMs (OBT), only one class is separated from the remaining classes at each node. Both approaches need only to train n – 1 (n is the number of classes) binary PPSVM classifiers, while the average convergence performance of SBT and OBT are O(log₂n) and O[(n! − 1)/n], respectively. Experimental results show that both approaches obtain classification accuracy if not higher, at least comparable to other multiclass approaches, while using significantly fewer support vectors and reduced testing time.

History

Received Jan 03, 2010
Accepted Jan 21, 2011
Revised Jan 20, 2011
Published online Mar 11, 2011

Digital Object Identifier

http://dx.doi.org/10.1117/1.3553800

Citation

Dongli Wang, Yan Zhou and Jianguo Zheng, "Binary tree of posterior probability support vector machines for hyperspectral image classification", J. Appl. Remote Sens. 5, 053503 (Mar 11, 2011); http://dx.doi.org/10.1117/1.3553800