From hyperplanes to large margin classifiers: Applications to SAR ATR

被引:5
作者
Zhao, Q [1 ]
Principe, JC [1 ]
Xu, DX [1 ]
机构
[1] Univ Florida, Computat Neuroengn Lab, Dept Elect & Comp Engn, Gainesville, FL 32611 USA
来源
AUTOMATIC TARGET RECOGNITION IX | 1999年 / 3718卷
关键词
structural risk minimization; SAR/ATR; support vector machines; hyperplanes;
D O I
10.1117/12.359940
中图分类号
V [航空、航天];
学科分类号
08 ; 0825 ;
摘要
In this paper, the structural risk minimization (SRM) criterion is employed to train a large margin classifier, the support vector machine (SVM). Its relative performance is compared with traditional classifiers employing hyperplanes against a realistic difficult problem, the synthetic aperture radar(SAR) automatic target recognition (ATR). In most pattern recognition applications, the task is to perform classification into a fixed number of classes. However, in some practical cases, such as ATR, one also needs to carry out a reliable pattern rejection. Experimental results showed that the SVM with the Gaussian kernels performs well in target recognition. Moreover, the SVM is able to form a local or "bounded" decision region that presents better rejection to confusers.
引用
收藏
页码:101 / 109
页数:9
相关论文
共 14 条
[1]   NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].
AKAIKE, H .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723
[2]  
Cherkassky V, 1997, IEEE Trans Neural Netw, V8, P1564, DOI 10.1109/TNN.1997.641482
[3]  
FRIESS T, 1998, MACH LEARN P 15 INT
[4]  
GIROSI F, 1997, 1606 MIT AI LAB
[5]  
Haykin S., 1994, NEURAL NETWORKS COMP
[6]  
Nilsson N. J., 1965, Series in Systems Science
[7]  
PRINCIPE J, 1998, P IM UND WORKSH MONT, P833
[8]   MODELING BY SHORTEST DATA DESCRIPTION [J].
RISSANEN, J .
AUTOMATICA, 1978, 14 (05) :465-471
[9]  
Tikhonov A.N., 1977, Solutions of Ill-posed problems
[10]  
VELTEN V, 1998, STANDARD SAR ATR EVA