基于DP算法的槽线自动判识  

Automatic Recognition of Meteorological Trough Lines Based on Douglas-Peucker Algorithm

作　　者：代曦[1] 李骞[1] 顾大权[1] 

机构地区：[1]解放军理工大学气象海洋学院,南京211101

出　　处：《气象科技》2016年第1期36-40,共5页Meteorological Science and Technology

基　　金：国家自然科学基金项目(41305138;41174164)资助

摘　　要：槽线自动识别是天气图自动化分析的重要内容,但目前的气象槽线分析主要通过人工实现。为了能够实现槽线准确快速的自动提取,本文从等压线数据着手,提出了一种基于Douglas-Peucker算法的气象槽线自动识别方法。该方法通过Douglas-Peucker算法从等压线中提取曲线特征点作为槽点,充分保留了等压线曲线的主要特征和拓扑关系;依据由等压线拓扑结构设定的规则连接槽点构成候选曲线,进行槽脊线区分得到槽线识别结果。试验表明该方法能够实现槽线的自动分析,分析结果较为准确。The automatic recognition of trough lines is essential in automatic synoptic maps analysis,while trough line analysis mainly depends on manual analysis presently.In order to extract trough lines rapidly and accurately with isoline data,a trough point extraction method based on the Douglas-Peucker algorithm is proposed.The algorithm extracts curve characteristic points as trough points through Douglas-Peucker algorithm,which retains the characters and topology of isolines mostly,then connects trough points to construct candidate curves with the related principles according to isoline topological structure,finally,remove ridge lines in the candidate curves.The results show that this algorithm can realize the automatic recognition of trough lines.

关 键 词：DP算法 槽线自动识别 天气分析 

分 类 号：P208[天文地球—地图制图学与地理信息工程]