5 November 2015 A new classification algorithm based on multi-kernel Support Vector Machine on infrared cloud background image
Author Affiliations +
Proceedings Volume 9795, Selected Papers of the Photoelectronic Technology Committee Conferences held June–July 2015; 97951I (2015) https://doi.org/10.1117/12.2216035
Event: Selected Proceedings of the Photoelectronic Technology Committee Conferences held June-July 2015, 2015, Hefei, Suzhou, and Harbin, China
Abstract
A new classification algorithm based on multi-kernel support vector machine (SVM) was proposed for classification problems on infrared cloud background image. The experimental results show that the method integrates the advantages of polynomial kernel functions, Gaussian radial kernel functions and multilayer perception kernel functions. Compared with the traditional single-kernel SVM classification method, the proposed method has better performance both in local interpolation and global extrapolation, and is more suitable for SVM classification problems when the training sample size is small. Experimental results show the superiority of the proposed algorithm..
© (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Tiebing Wang, Yiyu Zhou, Shenda Xu, Chuxiong Cheng, "A new classification algorithm based on multi-kernel Support Vector Machine on infrared cloud background image", Proc. SPIE 9795, Selected Papers of the Photoelectronic Technology Committee Conferences held June–July 2015, 97951I (5 November 2015); doi: 10.1117/12.2216035; https://doi.org/10.1117/12.2216035
PROCEEDINGS
6 PAGES


SHARE
RELATED CONTENT

The Space Thermal Signature Model: Principles And Applications
Proceedings of SPIE (September 16 1987)
Infrared Flight Simulation Using Computer Generated Imagery
Proceedings of SPIE (January 21 1985)
High-fidelity infrared scene simulation at Georgia Tech
Proceedings of SPIE (June 12 1996)
Synthetic IR Scene Generation
Proceedings of SPIE (August 24 1988)

Back to Top