摘要
在面向分类的高光谱遥感数据降维过程中,考虑到高光谱遥感数据内在的非线性结构和传统流形学习非监督的特点,提出一种新的监督等距映射方法(S-Isomap)。方法基于类间距离大于类内距离的思想,首先利用KMEANS算法对原始数据进行聚类得到样本的初始类别标签,采用新距离搜寻数据点的K近邻,进而实施等距映射降维。实验证明了该方法优于传统Isomap。
Considering the intrinsic nonlinear structure of hyperspectral remote sensing data and the characteristic of unsupervision of traditional manifold learning, during the process of dimension reduction of classification-oriented hyperspectral remote sensing data, we propose a new method of supervised isometric mapping (S-Isonmap). The method is based on the idea that the between-class distance is greater than the within-class distance. First it obtains initial category labels of the samples by using KMEANS algorithm on primary data for clustering; then it searches the K-Nearest neighbour of the data points with new distances, and further executes the dimension reduction by Isomap. Experiments demonstrate that the presented method outperforms the traditional Isomap.
出处
《计算机应用与软件》
CSCD
北大核心
2012年第8期66-69,共4页
Computer Applications and Software
基金
国家自然科学基金项目(41071273)
高等学校博士学科点专项科研基金资助课题(20090095110002)
中央高校基本科研业务费专项资金项目(2010QNA21)
国土环境与灾害监测国家测绘局重点实验室开放基金资助项目(LEDM2011B07)
江苏高校优势学科建设工程资助项目(SA1102)
关键词
高光谱遥感
特征提取
KMEANS
监督等距映射
Hyperspectral remote sensing Feature extraction KMEANS Supervised isometric mapping
作者简介
钱进,硕士生,主研领域:摄影测量与遥感。
邓喀中,教授。
范洪冬,讲师。
刘冬,硕士生。