计算机工程与应用 ›› 2015, Vol. 51 ›› Issue (15): 179-182.

• 图形图像处理 • 上一篇    下一篇

基于二叉划分树的多维尺度分析图像分类算法

焦斌亮1,2,范成龙1,2,王朝晖1,2   

  1. 1.燕山大学 信息科学与工程学院,河北 秦皇岛 066004
    2.河北省特种光纤与光纤传感重点实验室,河北 秦皇岛 066004
  • 出版日期:2015-08-01 发布日期:2015-08-14

Multidimensional scaling used for image classification based on binary partition trees

JIAO Binliang1,2, FAN Chenglong1,2, WANG Zhaohui1,2   

  1. 1.College of Information Science and Engineering, Yanshan University, Qinhuangdao, Hebei 066004, China
    2.The Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province, Qinhuangdao, Hebei 066004, China
  • Online:2015-08-01 Published:2015-08-14

摘要: 在二叉划分树的基础上提出一种应用多维尺度分析的合并准则。该算法对高光谱图像分析后建立区域模型,利用多维尺度分析各个区域模型的相似性,移除冗余信息对局部降维,对所得数据关联测量确定其关联性后,进行区域合并,形成二叉划分树的树形结构,利用修剪函数对所得二叉划分树进行修剪,完成分类。实验结果表明,该算法应用于高光谱图像分类具有较好的分类效果。

关键词: 高光谱图像分类, 区域模型, 二叉划分树, 多维尺度分析, 关联测量

Abstract: A merging criterion used multidimensional scaling on the basis of binary partition trees algorithm has been proposed. It builds regional models after an analysis on hyperspectral image, uses multidimensional scaling on the similarities of each regional model, reduces dimensions by removing redundant information, determines relevance by association measure on obtained data, and forms a BPT tree structure through regional merging. The pruning function is used to prune this tree structure on classification step. And the experimental conclusion demonstrates that it obtains a better effect for hyperspectral image classification.

Key words: hyperspectral image classification, regional model, binary partition trees, multidimensional scaling, association measure