Machine Identification of Throat Aurora
-
摘要: 喉区极光是一种发生在电离层对流喉区附近的极光现象,是极光卵向低纬侧延伸出的南北向分立结构,其可能对应由磁鞘高速流与磁层顶作用引发的磁层顶重联过程.喉区极光研究对深入理解太阳风—磁层—电离层耦合过程具有重要意义.从长期观测所积累的大量全天空极光观测数据中准确高效识别出喉区极光结构,是开展喉区极光统计研究的基础.本文利用北极黄河站2003—2017年全天空成像仪的极光观测数据,建立了喉区极光图像标注数据集;基于密集连接卷积神经网络(DenseNet)对极光图像全局高维表征的自动学习,首次实现了喉区极光结构的机器识别.算法对喉区极光识别准确率达96%,且具有良好的泛化性能.研究表明基于深度学习的图像识别方法可用于从海量极光观测数据中自动识别喉区极光事件.Abstract: Throat aurora is an auroral form frequently observed nearby the ionospheric convection throat region. Extending from the equatorward edge of the dayside auroral oval, and thus appears to be a north-south aligned discrete auroral structure. It is suggested to be the projection of the magnetopause reconnection process caused by the interaction between magnetopause and high-speed jets in magnetosheath. The investigation on throat aurora is meaningful for understanding solar wind-magnetosphere coupling process. It is the fundamental work of statistic study that identifying accurately and efficiently these special auroral structures from plenty of ASI images. Through preprocessing and labeling ASI images from Yellow River Station during 2003 and 2017, the throat auroral dataset is established, and the feature space of the auroral data is explored by using DenseNet and classified simultaneously based on whether an observation image has a structure of throat aurora, through which machine identification of throat auroral is achieved for the first time. The accuracy of the identification model is close to 96% and it has good generalization performance.
-
Key words:
- Throat aurora /
- Image classification /
- DenseNet
-
[1] PARKER E N. Auroral phenomena[J]. Proc. IRE, 1959, 47(2):239-244 [2] XING Zanyang. Observational Characteristics of Dayside Poleward Moving Auroral Forms and its Generation Mechanisms[D]. Xi'an:Xidian University, 2013(邢赞扬. 日侧极向运动极光结构的观测特征及其产生机制研究[D]. 西安:西安电子科技大学, 2013) [3] HU Zejun, YANG Huigen, AI Yong. Multiple wavelengths observation of dayside auroras in visible range-A preliminary result of the first wintering aurora observation in Chinese arctic station at Ny-Alesund[J]. Chin. J. Polar Res., 2005, 17(2):107-114(胡泽骏, 杨惠根, 艾勇. 日侧极光卵的可见光多波段观测特征-中国北极黄河站首次极光观测初步分析[J]. 极地研究, 2005, 17(2):107-114) [4] FRE Y, HARALD U. Localized aurora beyond the auroral oval[J]. Rev. Geophys., 2007, 45(1):R-G1003 [5] HAN Desheng, HU Zejun, CHEN Xiangcai. Recent results obtained from dayside optical auroral observations at Yellow River Station[J]. Chin. J. Polar Res., 2018, 3(3):235-250(韩德胜, 胡泽骏, 陈相材. 基于北极黄河站观测的日侧极光研究新进展[J]. 极地研究, 2018, 3(3):235-250) [6] SYRJÄ SUO M T, DONOVAN E F. Diurnal auroral occurrence statistics obtained via machine vision[J]. Ann. Geophys. Eur. Geophys. Soc., 2004, 22(4):1103-1113 [7] HU Z J, YANG H, HUANG D, et al. Synoptic distribution of dayside aurora:multiple-wavelength all-sky observation at Yellow River Station in Ny-Ålesund, Svalbard[J]. J. Atmos. Sol.:Terr. Phys., 2009, 71(8/9):794-804 [8] HAN Bing, YANG Chen, GAO Xinbo. Aurora image classification based on LDA combining with saliency information[J]. J. Software, 2013, 24(11):2758-2766(韩冰, 杨辰, 高新波. 融合显著信息的极光图像分类[J]. 软件学报, 2013, 24(11):2758-2766) [9] HAN B, ZHAO X, TAO D, et al. Dayside aurora classification via BIFs-based sparse representation using manifold learning[J]. Int. J. Comput. Math., 2014, 91(11):2415-2426 [10] ZHONG Y, HUANG R, ZHAO J, et al. Aurora image classification based on multi-feature latent Dirichlet allocation[J]. Remote Sens., 2018, 10(2):233 [11] LI Y, JIANG N. An aurora image classification method based on compressive sensing and distributed WKNN[C]//Computer Software and Applications Conference (COMPSAC). Tokyo:IEEE, 2018:347-354 [12] CLAUSEN L B N, NICKISCH H. Automatic classification of auroral images from the Oslo Auroral THEMIS (OATH) data set using machine learning[J]. J. Geophys. Res.:Space Phys., 2018, 123(7):5640-5647 [13] YANG Q, TAO D, HAN D, et al. Extracting auroral key local structures from all-sky auroral images by artificial intelligence technique[J]. J. Geophys. Res.:Space Phys., 2019, 124(5):3512-3521 [14] HAN D S, CHEN X C, LIU J J, et al. An extensive survey of dayside diffuse aurora based on optical observations at Yellow River Station[J]. J. Geophys. Res.:Space Phys., 2015, 120(9):7447-7465 [15] HAN D S, NISHIMURA Y, LYONS L R, et al. Throat aurora:the ionospheric signature of magnetosheath particles penetrating into the magnetosphere[J]. Geophys. Res. Lett., 2016, 43(5):1819-1827 [16] HAN D S, LIU J J, CHEN X C, et al. Direct evidence for throat aurora being the ionospheric signature of magnetopause transient and reflecting localized magnetopause indentations[J]. J. Geophys. Res.:Space Phys., 2018, 123(4):2658-2667 [17] HAN D S, HIETALA H, CHEN X C, et al. Observational properties of dayside throat aurora and implications on the possible generation mechanisms[J]. J. Geophys. Res.:Space Phys., 2017, 122(2):1853-1870 [18] CHEN X C, HAN D S, LORENTZEN D A, et al. Dynamic properties of throat aurora revealed by simultaneous ground and satellite observations[J]. J. Geophys. Res.:Space Phys., 2017, 122(3):3469-3486 [19] Polar Research Institute of China, High Altitude Atmospheric and Space Physics Team, 630.0nm all-sky aurora image at China's Arctic Yellow River Station[DB/OL]. 2020, National Arctic and Antarctic Data Center. DOI:10.11856/NNS.D.2020.001.v0(中国极地研究中心高空大气与空间物理团队, 中国北极黄河站630.0nm全天空极光图像[DB/OL]. 2020, 国家极地科学数据中心. DOI:10.11856/NNS.D.2020.001.v0) [20] WANG Qian. Image Classification and Dynamic Process Analysis for Dayside Aurora on All-sky Image[D]. Xi'an:Xidian University, 2011(王倩. 日侧全天空极光图像分类及动态过程分析方法研究[D]. 西安:西安电子科技大学, 2011) [21] TONG Xin, ZOU Ziming, ZHONG Jia, et al. Labeled throat aurora image dataset of ASI at Yellow River Station.V1[DB/OL]. NSSDC Space Science Article Data Repository. http://hdl.pid21.cn/21.86116.7/01.99.00027(佟欣, 邹自明, 钟佳, 等. 黄河站ASI喉区极光识别标注数据集.V1[DB/OL]. NSSDC Space Science Article Data Repository. http://hdl.pid21.cn/21.86116.7/01.99.00027) [22] HUANG G, LIU Z, VAN DER MAATEN L, et al. Densely connected convolutional networks[R]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017:4700-4708 [23] LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553):436-444 [24] DENG J, DONG W, SOCHER R, et al. Imagenet:a large-scale hierarchical image database[J]. Proc. IEEE Comput. Vision Pattern Recog., 2009, 2:248-255 [25] FU R, GAO X, JIAN Y. Patchy aurora image segmentation based on ALBP and block threshold[C]//International Conference on Pattern Recognition. Istanbul:IEEE, 2010:3380-3383
点击查看大图
计量
- 文章访问数: 480
- HTML全文浏览量: 91
- PDF下载量: 30
- 被引次数: 0