计算物理 ›› 1997, Vol. 14 ›› Issue (S1): 490-492.

• 论文 • 上一篇    下一篇

数字化卫星云图自动识别的动态聚类算法

杨越奎, 张方友, 刘玉玲   

  1. 空军气象研究所, 北京 100085
  • 收稿日期:1997-02-16 修回日期:1997-04-25 出版日期:1997-12-25 发布日期:1997-12-25

AN DYNAMIC CLUSTER MODEL OF SATELLITE CLOUD PICTURE CLASSIFICATION

Yang Yuekui, ZhangFangyou, Liu Yuling   

  1. The Meteorological Research Institute of the Air Force, P. L. A., Beijing 100085
  • Received:1997-02-16 Revised:1997-04-25 Online:1997-12-25 Published:1997-12-25

摘要: 根据数字化气象卫星资料的特点,设计了对卫星云图进行自动分类识别的动态聚类算法。利用GMS-5的10.5~11.5μm和0.55~0.9μm两个通道的资料,采用迭代自组织的数据分析算法(ISODATA算法)对云图进行了分类试验。

关键词: 卫星云图, 自动分类, 算法

Abstract: On the basis of the characteristics of satellite cloud picture, this paper gives one automatic classification method of satellite cloud image:Iterative Self Organizing Data Analysis Techniques Algorithm (ISODATA). The satellite data used are from two channels of GMS 5. One channel is of 10.5-11.5μm, the other is of 0.55-0.9μm. The calculation shows:while the cloud image is classified as surface, high cloud, middle and low cloud, deep cloud, good results have been achieved.

Key words: satellite cloud picture, automatic classification, algorithm

中图分类号: