| 改进相似性测度的高光谱影像FCM聚类分析 |
| |
| 引用本文： | 刘伟,冯伍法,任利华.改进相似性测度的高光谱影像FCM聚类分析[J].测绘工程,2008,17(5):29-32. |
| |
| 作者姓名： | 刘伟 冯伍法 任利华 |
| |
| 作者单位： | 信息工程大学,测绘学院,河南,郑州,450052;信息工程大学,测绘学院,河南,郑州,450052;68011部队,甘肃,兰州,730020 |
| |
| 摘 要： | 高光谱遥感将反映目标辐射属性的光谱信息与反映目标空间几何关系的图像信息有机地结合在一起，能够实现地面目标的精细分类识别。FCM是一种有效的聚类算法，但存在相似性测度模型单一、分类精度的提高受到限制等问题。文中结合高光谱影像的技术特点，综合考虑光谱曲线的形状、地物辐射亮度及其权重，提出可以更好描述光谱向量之间的相似性的距离测度，并将其引入到FCM聚类模型中。聚类分析试验结果表明：通过改进和优化相似性测度的FCM，可以显著提高高光谱影像聚类精度。
|
| 关 键 词： | 高光谱遥感 FCM聚类 相似性测度 |
Analysis on hyperspectral FCM clustering to improve comparability measure |
| |
| LIU Wei,FENG Wu-fa,REN Li-hua.Analysis on hyperspectral FCM clustering to improve comparability measure[J].Engineering of Surveying and Mapping,2008,17(5):29-32. |
| |
| Authors: | LIU Wei FENG Wu-fa REN Li-hua |
| |
| Institution: | LIU Wei, FENG Wu-fa, REN Li-hua (1. Institute of Surveying and Mapping, Information Engineering University, Zhengzhou 450052, China; 2. 68011 Troops, Lanzhou 730020,China) |
| |
| Abstract: | Hyperspectral remote sensing combines the spectrum information which reflects the radiation attribute with the image information which reflects the space geometry attribute of targets,and it can carry on fine identifications to targets on the ground.FCM is an effective clustering arithmetic,but its comparability measure is simplex.The precision is improved.This paper integrates the figure of spectrum with the radiation magnitude and their power,and brings out a better type of measure and introduces it into FCM model.The test result indicates that the precision of clustering to hyperspectral image is improved obviously through this method. |
| |
| Keywords: | hyperspectral remote sensing FCM clustering comparability measure |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
