计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (1): 180-182.DOI: 10.3778/j.issn.1002-8331.2010.01.053

• 图形、图像、模式识别 • 上一篇    下一篇

FUZZY-ARTMAP混合像元分解及其应用

潘建平   

  1. 重庆交通大学 土木建筑学院,重庆 400074
  • 收稿日期:2009-01-08 修回日期:2009-03-17 出版日期:2010-01-01 发布日期:2010-01-01
  • 通讯作者: 潘建平

Pixel unmixing and its application based on FUZZY-ARTMAP

PAN Jian-ping   

  1. School of Civil Engineering & Architecture,Chongqing Jiaotong University,Chongqing 400074,China
  • Received:2009-01-08 Revised:2009-03-17 Online:2010-01-01 Published:2010-01-01
  • Contact: PAN Jian-ping

摘要: 对于定量遥感来说,混合像元是制约它深入发展的难点之一。针对中高分辨率图像ETM+,采用FUZZY ARTMAP神经网络进行混合像元分解,获得像元中不同地物的含量,进而获得像元的植被覆盖度。该方法在某地区ETM+图像的应用表明,试验结果能满足应用需求。

关键词: 混合像元分解, 神经网络, 植被覆盖度, 模糊ARTMAP神经网络

Abstract: Subpixel is a difficulty for the development of quantitative remote sensing.The paper decomposes subpixel with fuzzy-artmap for the ETM+ image,and gets vegetation fraction for every pixel.The test shows that result is satisfactory for application.

Key words: pixel unmixing, neural network, vegetation fraction, FUZZY-ARTMAP

中图分类号: