High Performance Quadratic Classifier and the Application On PenDigits Recognition

Zhao, Z.J.; Sun, Jie; Ge, S.S.

doi:10.1109/CDC.2007.4434191

Access Status

Fulltext not available

Authors

Zhao, Z.J.

Sun, Jie

Ge, S.S.

Date

2007

Type

Conference Paper

Metadata

Show full item record

Citation

Zhao, Z.J. and Sun, J. and Ge, S.S. 2007. High Performance Quadratic Classifier and the Application On PenDigits Recognition. In: 46th IEEE Conference on Decision and Control, 12th Dec 2007, New Orleans.

Source Conference

46th IEEE Conference on Decision and Control

Faculty

Faculty of Science and Engineering

School

School of Elec Eng, Comp and Math Sci (EECMS)

URI

http://hdl.handle.net/20.500.11937/91440

Collection

Curtin Research Publications

Abstract

A nonconvex quadratic classifier is proposed for pattern recognition. The classifier is obtained by solving a second-order cone optimization problem on the training data set. Numerical results are presented to compare this classifier with the Gaussian classifier and k-NN classifiers. Regarding to the application of hand written digits recognition, the computational result shows that the proposed quadratic classifier always achieves highest correctness in the testing stage although it takes the longest computational time in the training stage. © 2007 IEEE.