计算机科学 ›› 2015, Vol. 42 ›› Issue (5): 47-50.doi: 10.11896/j.issn.1002-137X.2015.05.009

• 2014' 数据挖掘会议 • 上一篇    下一篇

基于监督学习的日冕暗化检测与提取算法

田红梅,彭 博,李天瑞,谢宗霞   

  1. 西南交通大学信息科学与技术学院 成都610031,西南交通大学信息科学与技术学院 成都610031,西南交通大学信息科学与技术学院 成都610031,天津大学软件学院 天津300072
  • 出版日期:2018-11-14 发布日期:2018-11-14
  • 基金资助:
    本文受国家自然科学基金(61202190,61175047,61105054),中央高校基本科研业务费专项资金(2682013CX055)资助

Coronal Dimming Detecting and Extracting Algorithm Based on Supervised Learning

TIAN Hong-mei, PENG Bo, LI Tian-rui and XIE Zong-xia   

  • Online:2018-11-14 Published:2018-11-14

摘要: 日冕物质抛射(CME)是空间灾害天气的重要驱动源,而日冕暗化(dimming)被认为是CME初发的主要表征,对理解和预测CME具有重要作用。基于极紫外成像望远镜(EIT)和大气成像仪(AIA)的观测数据,实现了图像中日冕暗化现象的检测与提取。通过分析差分图中与暗化现象相关的图像统计特征,采用Adaboost分类算法检测暗化现象的发生,进而分割出日冕暗化区域。实验表明,提出的算法较现有算法能更准确有效地检测和提取日冕暗化区域,为分析日冕暗化特性提供了研究基础。

关键词: 日冕物质抛射,日冕暗化,Adaboost分类,图像分割

Abstract: Coronal mass ejections (CMEs),which release huge quantities of matter and electromagnetic radiation into space above the sun’s surface,are considered as one of the driven sources of space weather.Coronal dimming is now viewed as the important characteristic of CME.Dimming can help understand,predict and locate the occurrence of CME.Based on the observed data from extreme ultra-violet imaging telescope (EIT) and atmospheric imaging assembly (AIA),this paper implemented the coronal dimming detection and extraction.By analyzing the statistical characteristics of the difference images related to dimming,we applied Adaboost classification algorithm into dimming detection,and then segmented the coronal dimming region.The experiment results show that the proposed algorithm can effectively detect and extract the coronal dimming areas.Our work establishes the basis for analysis of coronal dimming features.

Key words: Coronal mass ejections,Coronal dimming,Adaboost classification,Image segmentation

[1] Harrison R A,Lyons M.A spectroscopic study of coronal dimming associated with a coronal mass ejection[J].Astronomy and Astrophysics,2000(358):1097-1108
[2] Hudson H S,Lemen J R,Webb D F.Coronal X-Ray Dimming in Two Limb Flares[C]∥Proceedings of a Yohkoh Conference.1996:379-382
[3] Gopalswamy N,Hanaoka Y,Hudson H S.Structure and dynami-cs of the corona surrounding an eruptive prominence[J].Advances in Space Research,2000,25(9):1851-1854
[4] Hudson H S,Lemen J R,Cyr O C S,et al.X-ray coronal changes during halo CMEs[J].Geophysical Research Letters,1998,14(25):2481-2484
[5] Delaboudinière J P,Artzner G E,Brunaud J,et al.EIT:Ex-treme-ultraviolet Imaging Telescope for the SOHO mission[J].Solar Physics,1995,162(1/2):291-312
[6] Podladchikova O,Berghmans D.Automated Detection of EitWaves And Dimmings[J].Solar Physics,2005,228(1/2):265-284
[7] Attrill G D R,Wills-Davey M J.Automatic Detection and Extraction of Coronal Dimmings from SDO/AIA Data[J].Solar Physics,2010,262(2):461-480
[8] Attrill G,Nakwacki M S,Harra L K,et al.Using the Evolution of Coronal Dimming Regions to Probe the Global Magnetic Field Topology[J].Solar Physics,2006,238(1):117-139
[9] Reinard A A,Biesecker D A.Coronal Mass Ejection-AssociatedCoronal Dimmings[J].The Astrophysical Journal,2008,674(1):576
[10] Krista L D,Reinard A.Study of the Recurring Dimming Region Detected at AR 11305 Using the Coronal Dimming Tracker (CoDiT) [J].The Astrophysical Journal,2013,762(2):91
[11] Ayinala M,Parhi K K.Low complexity algorithm for seizureprediction using Adaboost[C]∥Proceedings of the Engineering in Medicine and Biology Society (EMBC).San Diego,2012:1061-1064
[12] Cao J,Kwong S,Wang R.A noise-detection based AdaBoost algorithm for mislabeled data[J].Pattern Recognition,2012,45(12):4451-4465
[13] Lan R S,Jiang Y,Ding L G,et al.Automated flare predictionusing the AdaBoost algorithm[J].Research in Astronomy and Astrophysics,2012,12(9):1191-1196
[14] Freeland S L,Handy B N.Data Analysis with the SolarSoft System[J].Solar Physics,1998,182(2):497-500
[15] Grady L.Multilabel random walker image segmentation using prior models[C]∥Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Volume 1,San Diego,2005:763-770

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!