首页 | 本学科首页   官方微博 | 高级检索  
     检索      

高光谱影像分类及组合分类器的应用
引用本文:宫大鹏,李之歆,韩建峰,杨哲海,张雅争.高光谱影像分类及组合分类器的应用[J].海洋测绘,2006,26(5):20-23.
作者姓名:宫大鹏  李之歆  韩建峰  杨哲海  张雅争
作者单位:1. 65015部队,辽宁,大连,116023
2. 65015部队,辽宁,大连,116023;解放军信息工程大学,测绘学院,河南,郑州,450052
3. 海军大连舰艇学院,辽宁,大连,116018
基金项目:国家863计划(2001AA135160)
摘    要:高光谱影像包含有丰富的地物类别信息,但由于受分类方法等因素的限制,这些信息并不能得到充分的挖掘。组合分类器由于可以实现各分类结果信息的互补,因而总有着较各单分类器效果更好的分类性能。将组合分类方法引入到高光谱影像的分类实践中,并对组合的关键问题一组合规则进行探讨,提出了混合组合规则的概念。实验表明,利用该组合分类方法对高光谱影像进行分类,可以得到良好的处理结果。

关 键 词:高光谱  分类  分类器  组合
文章编号:1671-3044(2006)05-0020-04
收稿时间:2006-04-06
修稿时间:2006-07-05

Hyperspectral Classification and the Utilization of Combined Classifiers
GONG Da-peng,LI Zhi-xin,HAN Jian-feng,YANG Zhe-hai,ZHANG Ya-zheng.Hyperspectral Classification and the Utilization of Combined Classifiers[J].Hydrographic Surveying and Charting,2006,26(5):20-23.
Authors:GONG Da-peng  LI Zhi-xin  HAN Jian-feng  YANG Zhe-hai  ZHANG Ya-zheng
Institution:1. 65015 Troops, Dalian, Liaoning, 116023 ; 2. Institute of Surveying and Mapping, Information Engineering University, Zhengzhou, Henan,450052 ; 3. Dalian Naval Academy, Dalian, Liaoning, 116018
Abstract:Hyperspectral images can provide more information about the earth's surface due to the higher spectral resolution.Unfortunately the classification performance will be deteriorated with the liniations of the classifiers.Nevertheless,combined classifiers,which utilize the comlementary information contained in the classification results,usually leads to much better classification performance.In this paper the combined classifiers were introduced into the hyperspectral classification.Also a new hybrid combination rule was proposed.Experiment demonstrated that the proposed combined classifiers are superior in hyperspectral classification.
Keywords:hyperspectral  classification  classifier  combination
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号