首页> 中文期刊> 《智能系统学报》 >基于关联分析的气象云图识别问题研究

基于关联分析的气象云图识别问题研究

         

摘要

Important steps in nephogram recognition are selection and tracking of the target clouds. A new scheme of relational analysis is developed to improve tracking of tracers. The tracers are selected based on the clouds’ attrib-ute and attribute evaluation value. A concept lattice can be established. However, if the selected tracers are consid-ered as object set,s then characteristics of clouds as attribute sets, and the relationship among the same attributes of different tracers at different moments as the binary relation are efficient. So, tracers can be tracked according to the number of the intension of concept lattice nodes and the similarity of extension. The results showed that the charac-teristics of tracers based on the selection method are useful for tracking tracers based on relational analysis.%气象云图识别中的一项很重要的工作是示踪云选取及追踪。提出了关联分析的识别方法,旨在提高示踪云识别。针对云块的属性及属性评估值进行示踪云选取;将选取的示踪云作为对象集,云块特征作为属性集,不同时刻示踪云关于属性的关系作为二元关系来建立概念格,借助概念格节点的内涵个数及外延属性的相似度进行示踪云追踪。实验结果表明,此选取方法得到的示踪云特征明显,有利于示踪云的追踪。

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号