云导风的初步探究  

The Preliminary Research on Cloud Motion Wind

作　　者：杨平吕[1] 刘祥余[1] 郭付才[1] 彭新伟[2] 黄思训[1] 项杰[1] 

机构地区：[1]解放军理工大学气象海洋学院,江苏南京211101 [2]南京军区气象水文中心,江苏南京210016

出　　处：《数学的实践与认识》2013年第15期163-174,共12页Mathematics in Practice and Theory

摘　　要：针对云导风反演算法进行了深入研究.通过分析卫星云图的灰度特性和纹理特征,提出了基于最大灰度和梯度交叉相关系数的云导风反演方法.依据最大信息熵准则自适应调整窗口大小,根据NCEP数值预报产品指定风矢高度.实验结果表明,方法有效提高了反演的义精度和效率.The cloud motion wind inverse method is explored in this paper. By analyzing the gray-scale and texture features of satellite images, a new cloud motion wind algorithm based on maximum cross-correlation coefficient of gray-scale and gradient information is proposed. The size of match window is changed adaptively based on the maximum entropy criterion. And the wind vector height is specified by the NCEP numerical prediction products. Experimental results show that the proposed method effectively improves the accuracy and efficiency.

关 键 词：云导风 最大交叉相关系数 信息熵 双线性插值 

分 类 号：P426.5[天文地球—大气科学及气象学] TP391.41[自动化与计算机技术—计算机应用技术]