计算机工程与应用 ›› 2019, Vol. 55 ›› Issue (1): 226-232.DOI: 10.3778/j.issn.1002-8331.1709-0243

• 工程与应用 • 上一篇    下一篇

基于曲率值追踪的气象槽线自动分析算法

黄  岩,李  骞,范  茵   

  1. 国防科技大学 气象海洋学院,南京 211100
  • 出版日期:2019-01-01 发布日期:2019-01-07

School of Meteorology & Oceanography, National University of Defense Technology, Nanjing 211100, China

HUANG Yang, LI Qian, FAN Yin   

  1. School of Meteorology & Oceanography, National University of Defense Technology, Nanjing 211100, China
  • Online:2019-01-01 Published:2019-01-07

摘要: 针对传统槽线自动分析算法中存在的处理效率较低,且分析准确度不高的问题,提出一种基于曲率值追踪的槽线自动分析算法。算法利用位势高度场格点数据,直接计算格点处曲率值,并根据曲率值及相对位置对格点进行分类判别,以提取候选槽点;在候选槽点中筛选合理槽点进行追踪连接;最终通过后处理步骤,得到槽线自动分析结果。实验结果表明,该算法可有效实现气象槽线的自动分析,在分析准确度和运行速度方面较之前算法均有所改进。

关键词: 槽线, 自动分析, 位势高度场, 格点分类, 曲率值追踪

Abstract: To solve the problems occurred in traditional trough lines analysis algorithms such as low efficiency and accuracy, an automatic algorithm of trough lines analysis is proposed. The algorithm obtaines trough lines using geopotential height field data through some steps. It is firstly to calculate curvature of contour lines where grid points locate, the grid points are classified into candidate and non-candidate trough points according to the curvature and relative?positions. Then, the eligible trough points identified among candidate trough points are tracked and connected. Finally, the analysis results of trough lines are obtained after post-processing steps, which mainly includes noise eliminating and curve smoothing. The experimental results show that the algorithm can recognize trough lines effectively, which is faster and more accurate than some other previous methods.

Key words: trough lines, automatic analysis, geopotential height field, classification of grid points, tracking according to curvature