摘要
基于支持向量机（SVM）的迭代特征删除（SVM—RFE）法用于高光谱数据波段选择时，常用的非线性核函数训练时间长，并且每删除一个波段均需要重新训练SVM，总体效率低。研究表明在SVM分类中非线性核函数并不一定优于线性核函数。对比分析了两种核函数SVM在SVM—RFE中对分类结果的影响，并设计了两种提高SVM—RFE效率的策略：比率加速法和固定加速法。通过对AVIRIS高光谱数据实验得出：①SVM的分类精度随着冗余波段的增加而略微下降，即从分类精度上考虑SVM也需要特征选择；②相对于非线性核SVM—RFE，线性SVM—RFE选择出的最佳波段组合分类精度高1％～3％，训练时间极大减少；③两种效率优化策略均能提高特征选择效率，比率加速法在时间效率和分类精度上均优于固定加速法。
Supporting vector machine recursive feature elimination （SVM-RFE） has a low efficiency when it is ap plied to band selection for hyperspectral dada,since it usually uses a non-linear kernel and trains SVM every time after deleting a band. Recent research shows that SVM with non-linear kernel doesn＇t always perform better than linear one for SVM classification. Similarly,there is some uncertainty on which kernel is better in SVM RFE based band selection. This paper compares the classification results in SVM-RFE using two SVMs, then designs two opti mization strategies for accelerating the band selection process：the percentage accelerated method and the fixed ac- celerated method. Through an experiment on AVIRIS hyperspectral data,this paper found： ① Classification preci sion of SVM will slightly decrease with the increasing of redundant bands, which means SVM classification needs feature selection in terms of classification accuracy; ② The best band collection selected by SVM-RFE with linear SVM that has higher classification accuracy and less effective bands than that with non-linear SVM; ③ Both two optimization strategies improved the efficiency of the feature selection,and percentage eliminating performed beuer than fixed eliminating method in terms of computational efficiency and classification accuracy.
出处
《遥感技术与应用》
CSCD
北大核心
2013年第5期747-752,共6页
Remote Sensing Technology and Application
基金
国家863计划项目(2009AA122004)
中国博士后科学基金(20110490350)
关键词
支持向量机分类
Hughes现象
波段选择
迭代特征删除
Support Vector Machine Classification
Hughes phenomenon
Feature selection
Recursive Feature Elimination
