上海交通大学学报

上海交通大学学报（自然版） ›› 2011, Vol. 45 ›› Issue (10): 1516-1520.

• 自动化技术、计算机技术 • 上一篇    下一篇

遗传关联向量机高光谱影像分类

董超1,2，田联房1，赵慧洁2   

  1. (1. 华南理工大学 自动化科学与工程学院, 广州 510641;2. 北京航空航天大学 仪器科学与光电工程学院, 北京 100191)
  • 收稿日期:2010-11-09 出版日期:2011-10-31 发布日期:2011-10-31
  • 基金资助:

    国家高技术研究发展计划（863）项目（2008AA121102）；中国博士后科学基金面上项目（20100480750）

Hyperspectral Image Classification by Genetic Relevance Vector Machine

 DONG  Chao-1, 2 , TIAN  Lian-Fang-1, ZHAO  Hui-Jie-2   

  • Received:2010-11-09 Online:2011-10-31 Published:2011-10-31

摘要： 基于高光谱影像临近波段相关性高, 直接在高维空间分类并非最优，并且使用交叉验证进行分类器参数寻优过程繁琐，提出了遗传关联向量机（GARVM）高光谱影像分类算法，使用遗传算法搜索面向关联向量机（RVM）的最优参数和特征子空间, 消除冗余信息, 简化参数优化过程.实验环节验证了GARVM算法的有效性，剔除约50%冗余波段后，总体分类精度提高3%, 对难分地物改进尤为明显, 其中混分最严重的2种大豆精度提高了8%.

关键词:  , 高光谱, 分类, 关联向量机, 遗传算法

Abstract: The adjacent bands of hyperspectral image are highly correlated. It is not optimum to classify the hyperspectral image in the high dimensional space. In addition, optimizing the parameter of classifier by the cross validation method is not a trivial task. Aiming at the two targets, the classification of the hyperspectral image with genetic relevance vector machine (GA-RVM) was proposed. GARVM searches the best parameter and feature space for relevance vector machine (RVM), to reduce the redundant information and simplify the parameter optimization procedure. GA-RVM was evaluated by several experiments. Nearly 50% of the bands are eliminated during the optimization, leading to a 3% increase in the overall accuracy. The improvements are obvious for the hard-to-separate classes. Two kinds of soybeans that have the most misclassifications acquire an 8% improvement in accuracy.

Key words: hyperspectral, classification, relevance vector machine, genetic algorithm

中图分类号: