基于全天空图像和紫外图像的极光事件检测与分析
|
摘要
极光是沿磁力线运动的高能带电粒子沉降到极区电离层高度时激发大气粒子后产生的发光现象。作为极区日地物理过程（特别是磁层-电离层相互作用）最具代表性的表现形式，人们通过极光形态及其演化的系统观测可以获得磁层和日地空间电磁活动的大量信息，有助于深入研究太阳活动对地球的影响方式与程度，对了解空间天气过程的变化规律具有重要意义。
极光研究是空间物理研究领域的重要方向之一，极光的综合观测也成为世界各国极地科学考察活动的重要科考项目。光学成像是应用最早、使用最广泛的极光观测手段，其优点在于能够得到极光的二维图像，并且可以实现对极光的空间运动特性进行连续观测。目前常用的光学观测包括基于地面的全天空成像仪（All-sky imager, ASI）拍摄的ASI图像和基于卫星的紫外成像仪（Ultra-violetimager, UVI）拍摄的UVI图像。随着极光系统观测的开展，海量的极光图像数据在逐年累加，如何高效分析和利用这些数据成为各国极光物理研究人员亟待解决的一个问题。
本文针对传统人工分析费时费力以及少量事件分析得出的结论普适性不强等问题，研究海量极光图像的自动分析方法。基于我国北极黄河站的高分辨率ASI观测图像，重点分析了极光形态分类和极向运动事件检测两个典型问题；基于Polar卫星的全域视角UVI图像，重点研究了亚暴起始时刻检测和极光卵边界位置建模两个热点问题。本文提出了多种极光图像的表征、分类和检测方法，取得的主要研究成果有：
(1)考虑到极光特征随时间动态演变的特点，提出了一种基于隐马尔可夫模型（HMM）的极光动态特征描述和分类方法。该方法首先采用局部二值模式(LBP)算子提取ASI极光图像的空间纹理信息，然后采用HMM对所获得的LBP特征序列进行建模表征，兼顾了极光的空时二维信息。由于极光演化过程速度多变，极光序列的长度有长有短，为此本文引入仿射归一化对数似然函数，解决了不等长图像序列的HMM建模问题。利用该方法对北极黄河站2003-2009年的越冬ASI观测数据进行了极光形态的有监督分类和分布规律统计，结果表明，相比基于静态单帧ASI图像的分析方法，基于图像序列的分析方法具有更高的分类准确率和更好的鲁棒性。
(2)极向运动是一种典型的极光事件，本文提出了一种基于ASI图像序列的极向运动极光事件自动识别方法。该方法首先采用块匹配算法、方向编码机制及直方图统计技术对相邻帧间的极光运动情况进行估计，然后采用HMM对极光运动特征向量进行建模和相似度表征；最后采用支持向量机（SVM）进行二分类，找出极向运动事件。识别过程中，由于相比一般的极光运动模式，极向运动只是少数情况，本文通过采用元学习的方法训练不同测度下的SVM模型，很好的解决了数据不平衡问题。本文提出的方法在北极黄河站2003-2004年的越冬数据上进行了验证，实验结果表明自动识别方法得出的极向运动发生规律与已有的人工统计结果基本一致。
(3)准确界定亚暴起始时刻是理解亚暴相关问题的关键。在目前的观测手段中，UVI图像是公认的研究亚暴现象的最佳手段，但是现有的研究都是建立在基于人工标定亚暴起始点的基础上。本文提出从UVI极光图像中自动检测亚暴起始点。模糊c均值聚类方法被用来提取UVI图像中的亮斑区域，然后根据极光点亮后亮斑强度和面积随时间的变化情况来界定是不是亚暴事件。本文的方法在1996年冬季三个月的Polar卫星数据进行了实验，通过与已有文献中给出的人工标记进行对比，实验结果表明本文的方法能有效检测出具有亚暴特征的极光事件。
(4)极光卵的边界位置信息携带有磁层过程的丰富信息。与传统的研究极光卵边界随某一个综合地磁参数变化不同，本文提出从太阳风等离子体和行星际磁场这个极光产生的源头来考虑极光卵边界的位置变化。本文首先基于模糊c均值聚类方法自动获取了1996-1999期间三个冬季大量的Polar卫星UVI极光卵边界位置数据，并基于OMNI数据获得了太阳风行星际条件参数；然后利用多元回归分析方法建立了极光卵边界位置变化与太阳风行星际条件之间的预测模型。本文采用交叉验证方法进行数值实验，将模型预测值与真实观测数据之间的平均绝对值误差作为模型评判准则，结果表明，本文建立的预测模型平均绝对误差在1.65地磁纬度左右，可以用于空间天气研究中对极光卵边界位置的预测。本文受国家自然科学基金（60872154,41031064）和海洋公益性行业科研专项（201005017-04）等项目资助。
An aurora is a natural light display in the sky particularly in the high latituderegions, caused by the collision of energetic charged particles with atoms in the highlatitude atmosphere. The spatial structure and temporal evolution of auroral luminosityare ascribed to the cumulative effects of the solar wind–magnetosphere interaction andthe physics of the magnetosphere–ionosphere interaction. Therefore, much informationof the magnetosphere and the ionosphere in solar-terrestrial space can be obtained aftersystematic observation and deep analysis to auroras, which helps us to study the wayand extent of the sun activities affecting the earth and it is very important to learnabout the behave of space weather.
Auroral research has been one of the important topics in space physics worldwide,and the comprehensive observation to aurora has become significant in polar adventure.Optical imaging is the earliest applied and most popular auroral observation tools, withthe merit that it can obtain the2-D auroral images and can continually observe auroralspacial motion characteristics. At present, the frequently-used optical observationsinclude auroral images obtained by all-sky imagers (ASI) and ultra-violet imagers(UVI). With the systematic observation to aurora, massive auroral images have beenincreasing each year, so how to efficiently utilize the huge auroral data is an urgentproblem to be solved for auroral researchers in different countries.
Traditional auroral image analysis has always been made manually, which is notonly difficult and time-consuming, but also the conclusions drawn from the normalcase study are lack of universality and block the popularity. To address this problem,we try to automatically analyze the massive auroral data. Based on the high resolutionASI images observed in Yellow River Station, we focus on auroral morphologyclassification and poleward moving auroral events detection, and based on the globalUVI images observed by Polar satellite, we aim at substorm onset detection andmodeling auroral oval boundary locations. These are significant topics in auroralresearch and we try to analyze them from a perspective of computer learning, which isa fresh attempt in auroral research. A few methods suited for the representation,classification and detection of auroral image data have been proposed in this paper. Insummary, the author’s major contributions are outlined as follows:
1) Because the naturally occurring auroras are a dynamically evolving process,we propose to study auroral morphology classification based on ASI imagesequence. The uniform local binary patterns (uLBP) are employed to describe the2-D space structures of ASI images, and then the extracted uLBPsequences are modeled by hidden Markov models (HMM). Thus this methodcan characterize both temporal and spatial information of auroras meanwhile.The velocity of auroral evolution is various, and the length of auroralsequences is different. We present an affine log-likelihood normalizationtechnique to manage the sequences with different lengths. The proposedmethod is used in the automatic recognition of four primary categories of ASIimages. Compared to the frame-based method, the supervised classificationresults of our method achieve higher accuracies and lower rejection rates, andthe occurrence distributions of the four auroral categories further illustrate thevalidity of the proposed method on auroral representation and classification.
2) We present an automatic method to recognize the poleward moving aurorasfrom all-sky image sequences. A simplified block matching algorithmcombined with an orientation coding scheme and histogram statistics strategywas utilized to estimate the auroral motion between interlaced images. Anall-sky image sequence was first modeled by HMM models and thenrepresented by HMM similarities. The imbalanced classification problem, i.e.,non-poleward moving auroral events far outnumbering poleward movingauroral events, was addressed by the metric-driven biased Support VectorMachine (SVM). The experimental results show that the occurrence rules ofthe poleward moving auroral events calculated by the proposed automaticmethod are the same with the existing manual statistical results.
3) Substorm research largely depends on the precise definition and the timingaccuracy of the substorm onset used in various observations. At present, theglobal UVI images are widely accepted as the best tool from which to learnsubstorm. However, the existing studies are all based on the manualrecognition of substorm onsets. We propose to automatically identify auroralsubstorm onset timing from UVI images. We first transformed the originalUVI images into the MLT-MLAT rectangular coordinate system, and thebright bulge was determined with the spatial fuzzy c-means method. Theproposed technique was tested using Polar UVI observations acquired fromthree months of the winter season in1996-1997. The identified substorm onsetresults were compared with the available manual statistical report, and theexperimental results demonstrate the proposed technique can efficientlyrecognize those auroral events with substorm features.
4) The location of auroral oval boundary gives us valuable information aboutmagnetosphere processes. Different from traditional study to investigate thevariation of auroral oval boundaries with a certain magnetic index, we proposeto consider their variation from the source, i.e., the relationship betweenauroral boundaries and solar wind plasma and the interplanetary magneticfield. Auroral oval boundaries were automatically identified from UVI imagesof three winters. Based on the big data, at each one-hour MLT sector, westatistically analyze the response of oval boundaries to various solar-windmagnetic field conditions by using multivariate regression technique. Thepredictions were compared to the actual magnetic latitude with the MeanAbsolute Deviation (MAD) as an evaluator. The average MAD is about1.65magnetic degrees. The high fitness between predictions and observationsdemonstrates that the presented models can be used for prediction of auroraloval.
极光研究是空间物理研究领域的重要方向之一，极光的综合观测也成为世界各国极地科学考察活动的重要科考项目。光学成像是应用最早、使用最广泛的极光观测手段，其优点在于能够得到极光的二维图像，并且可以实现对极光的空间运动特性进行连续观测。目前常用的光学观测包括基于地面的全天空成像仪（All-sky imager, ASI）拍摄的ASI图像和基于卫星的紫外成像仪（Ultra-violetimager, UVI）拍摄的UVI图像。随着极光系统观测的开展，海量的极光图像数据在逐年累加，如何高效分析和利用这些数据成为各国极光物理研究人员亟待解决的一个问题。
本文针对传统人工分析费时费力以及少量事件分析得出的结论普适性不强等问题，研究海量极光图像的自动分析方法。基于我国北极黄河站的高分辨率ASI观测图像，重点分析了极光形态分类和极向运动事件检测两个典型问题；基于Polar卫星的全域视角UVI图像，重点研究了亚暴起始时刻检测和极光卵边界位置建模两个热点问题。本文提出了多种极光图像的表征、分类和检测方法，取得的主要研究成果有：
(1)考虑到极光特征随时间动态演变的特点，提出了一种基于隐马尔可夫模型（HMM）的极光动态特征描述和分类方法。该方法首先采用局部二值模式(LBP)算子提取ASI极光图像的空间纹理信息，然后采用HMM对所获得的LBP特征序列进行建模表征，兼顾了极光的空时二维信息。由于极光演化过程速度多变，极光序列的长度有长有短，为此本文引入仿射归一化对数似然函数，解决了不等长图像序列的HMM建模问题。利用该方法对北极黄河站2003-2009年的越冬ASI观测数据进行了极光形态的有监督分类和分布规律统计，结果表明，相比基于静态单帧ASI图像的分析方法，基于图像序列的分析方法具有更高的分类准确率和更好的鲁棒性。
(2)极向运动是一种典型的极光事件，本文提出了一种基于ASI图像序列的极向运动极光事件自动识别方法。该方法首先采用块匹配算法、方向编码机制及直方图统计技术对相邻帧间的极光运动情况进行估计，然后采用HMM对极光运动特征向量进行建模和相似度表征；最后采用支持向量机（SVM）进行二分类，找出极向运动事件。识别过程中，由于相比一般的极光运动模式，极向运动只是少数情况，本文通过采用元学习的方法训练不同测度下的SVM模型，很好的解决了数据不平衡问题。本文提出的方法在北极黄河站2003-2004年的越冬数据上进行了验证，实验结果表明自动识别方法得出的极向运动发生规律与已有的人工统计结果基本一致。
(3)准确界定亚暴起始时刻是理解亚暴相关问题的关键。在目前的观测手段中，UVI图像是公认的研究亚暴现象的最佳手段，但是现有的研究都是建立在基于人工标定亚暴起始点的基础上。本文提出从UVI极光图像中自动检测亚暴起始点。模糊c均值聚类方法被用来提取UVI图像中的亮斑区域，然后根据极光点亮后亮斑强度和面积随时间的变化情况来界定是不是亚暴事件。本文的方法在1996年冬季三个月的Polar卫星数据进行了实验，通过与已有文献中给出的人工标记进行对比，实验结果表明本文的方法能有效检测出具有亚暴特征的极光事件。
(4)极光卵的边界位置信息携带有磁层过程的丰富信息。与传统的研究极光卵边界随某一个综合地磁参数变化不同，本文提出从太阳风等离子体和行星际磁场这个极光产生的源头来考虑极光卵边界的位置变化。本文首先基于模糊c均值聚类方法自动获取了1996-1999期间三个冬季大量的Polar卫星UVI极光卵边界位置数据，并基于OMNI数据获得了太阳风行星际条件参数；然后利用多元回归分析方法建立了极光卵边界位置变化与太阳风行星际条件之间的预测模型。本文采用交叉验证方法进行数值实验，将模型预测值与真实观测数据之间的平均绝对值误差作为模型评判准则，结果表明，本文建立的预测模型平均绝对误差在1.65地磁纬度左右，可以用于空间天气研究中对极光卵边界位置的预测。本文受国家自然科学基金（60872154,41031064）和海洋公益性行业科研专项（201005017-04）等项目资助。
An aurora is a natural light display in the sky particularly in the high latituderegions, caused by the collision of energetic charged particles with atoms in the highlatitude atmosphere. The spatial structure and temporal evolution of auroral luminosityare ascribed to the cumulative effects of the solar wind–magnetosphere interaction andthe physics of the magnetosphere–ionosphere interaction. Therefore, much informationof the magnetosphere and the ionosphere in solar-terrestrial space can be obtained aftersystematic observation and deep analysis to auroras, which helps us to study the wayand extent of the sun activities affecting the earth and it is very important to learnabout the behave of space weather.
Auroral research has been one of the important topics in space physics worldwide,and the comprehensive observation to aurora has become significant in polar adventure.Optical imaging is the earliest applied and most popular auroral observation tools, withthe merit that it can obtain the2-D auroral images and can continually observe auroralspacial motion characteristics. At present, the frequently-used optical observationsinclude auroral images obtained by all-sky imagers (ASI) and ultra-violet imagers(UVI). With the systematic observation to aurora, massive auroral images have beenincreasing each year, so how to efficiently utilize the huge auroral data is an urgentproblem to be solved for auroral researchers in different countries.
Traditional auroral image analysis has always been made manually, which is notonly difficult and time-consuming, but also the conclusions drawn from the normalcase study are lack of universality and block the popularity. To address this problem,we try to automatically analyze the massive auroral data. Based on the high resolutionASI images observed in Yellow River Station, we focus on auroral morphologyclassification and poleward moving auroral events detection, and based on the globalUVI images observed by Polar satellite, we aim at substorm onset detection andmodeling auroral oval boundary locations. These are significant topics in auroralresearch and we try to analyze them from a perspective of computer learning, which isa fresh attempt in auroral research. A few methods suited for the representation,classification and detection of auroral image data have been proposed in this paper. Insummary, the author’s major contributions are outlined as follows:
1) Because the naturally occurring auroras are a dynamically evolving process,we propose to study auroral morphology classification based on ASI imagesequence. The uniform local binary patterns (uLBP) are employed to describe the2-D space structures of ASI images, and then the extracted uLBPsequences are modeled by hidden Markov models (HMM). Thus this methodcan characterize both temporal and spatial information of auroras meanwhile.The velocity of auroral evolution is various, and the length of auroralsequences is different. We present an affine log-likelihood normalizationtechnique to manage the sequences with different lengths. The proposedmethod is used in the automatic recognition of four primary categories of ASIimages. Compared to the frame-based method, the supervised classificationresults of our method achieve higher accuracies and lower rejection rates, andthe occurrence distributions of the four auroral categories further illustrate thevalidity of the proposed method on auroral representation and classification.
2) We present an automatic method to recognize the poleward moving aurorasfrom all-sky image sequences. A simplified block matching algorithmcombined with an orientation coding scheme and histogram statistics strategywas utilized to estimate the auroral motion between interlaced images. Anall-sky image sequence was first modeled by HMM models and thenrepresented by HMM similarities. The imbalanced classification problem, i.e.,non-poleward moving auroral events far outnumbering poleward movingauroral events, was addressed by the metric-driven biased Support VectorMachine (SVM). The experimental results show that the occurrence rules ofthe poleward moving auroral events calculated by the proposed automaticmethod are the same with the existing manual statistical results.
3) Substorm research largely depends on the precise definition and the timingaccuracy of the substorm onset used in various observations. At present, theglobal UVI images are widely accepted as the best tool from which to learnsubstorm. However, the existing studies are all based on the manualrecognition of substorm onsets. We propose to automatically identify auroralsubstorm onset timing from UVI images. We first transformed the originalUVI images into the MLT-MLAT rectangular coordinate system, and thebright bulge was determined with the spatial fuzzy c-means method. Theproposed technique was tested using Polar UVI observations acquired fromthree months of the winter season in1996-1997. The identified substorm onsetresults were compared with the available manual statistical report, and theexperimental results demonstrate the proposed technique can efficientlyrecognize those auroral events with substorm features.
4) The location of auroral oval boundary gives us valuable information aboutmagnetosphere processes. Different from traditional study to investigate thevariation of auroral oval boundaries with a certain magnetic index, we proposeto consider their variation from the source, i.e., the relationship betweenauroral boundaries and solar wind plasma and the interplanetary magneticfield. Auroral oval boundaries were automatically identified from UVI imagesof three winters. Based on the big data, at each one-hour MLT sector, westatistically analyze the response of oval boundaries to various solar-windmagnetic field conditions by using multivariate regression technique. Thepredictions were compared to the actual magnetic latitude with the MeanAbsolute Deviation (MAD) as an evaluator. The average MAD is about1.65magnetic degrees. The high fitness between predictions and observationsdemonstrates that the presented models can be used for prediction of auroraloval.
引文
[1]刘瑞源,杨惠根.中国极区高空大气物理学观测研究进展.极地研究.2011,23(4),241-258.
[2]胡泽骏,杨惠根,艾勇,等.日侧极光卵的可见光多波段观测特征.极地研究.2005,17(2),107-114.
[3] http://sd-www.jhuapl.edu/Aurora/nightb/nightb.html.
[4] Liou K. Polar Ultraviolet Imager observation of auroral breakup. Journal ofGeophysical Research: Space Physics (1978–2012).2010,115(A12).
[5] Torr M, Torr D, Zukic M, et al. A far ultraviolet imager for the internationalsolar-terrestrial physics mission. Space science reviews.1995,71(1-4),329-383.
[6] Yang H, Sato N, Makita K, et al. Synoptic observations of auroras along thepostnoon oval: A survey with all-sky TV observations at Zhongshan, Antarctica.Journal of Atmospheric and Solar-Terrestrial Physics.2000,62(9),787-797.
[7] Sandholt PE, Carlson HC, Egeland A. Dayside and polar cap aurora: KluwerAcademic Pub;2002.
[8]胡红桥,刘瑞源,王敬芳,等.南极中山站极光形态的统计特征.极地研究.1999,11(1),8-18.
[9] Hu Z-J, Yang H, Huang D, et al. Synoptic distribution of dayside aurora:Multiple-wavelength all-sky observation at Yellow River Station inNy-lesund, Svalbard. Journal of Atmospheric and Solar-Terrestrial Physics.2009,71(8),794-804.
[10] Wang Q, Liang J, Hu Z-J, et al. Spatial texture based automatic classification ofdayside aurora in all-sky images. Journal of Atmospheric and Solar-TerrestrialPhysics.2010,72(5),498-508.
[11] Dungey JW. Interplanetary magnetic field and the auroral zones. PhysicalReview Letters.1961,6(2),47-48.
[12] Lemaire J, Roth M. Penetration of solar wind plasma elements into themagnetosphere. Journal of Atmospheric and Terrestrial Physics.1978,40(3),331-335.
[13] Wei C, Lee L, Belle‐Hamer L. A simulation study of the vortex structure inthe low‐latitude boundary layer. Journal of Geophysical Research: SpacePhysics (1978–2012).1990,95(A12),20793-20807.
[14] Xing Z, Yang H, Han D, et al. Poleward moving auroral forms (PMAFs)observed at the Yellow River Station: A statistical study of its dependence onthe solar wind conditions. Journal of Atmospheric and Solar-Terrestrial Physics.2012.
[15] Feldstein YI, Starkov G. Dynamics of auroral belt and polar geomagneticdisturbances. Planetary and Space Science.1967,15(2),209-229.
[16] Frank LA, Craven JD, Burch J, et al. Polar views of the Earth's aurora withDynamics Explorer. Geophysical research letters.1982,9(9),1001-1004.
[17] Akasofu S-I. The development of the auroral substorm. Planetary and SpaceScience.1964,12(4),273-282.
[18] Akasofu SI, Kan J. Dayside and nightside auroral arc systems. Geophysicalresearch letters.1980,7(10),753-756.
[19] Frey H, Mende S, Angelopoulos V, et al. Substorm onset observations byIMAGE‐FUV. Journal of Geophysical Research: Space Physics (1978–2012).2004,109(A10).
[20] Murphree J, Cogger L, Anger C, et al. Large scale6300,5577,3914dayside auroral morphology. Geophysical research letters.1980,7(4),239-242.
[21] Lui A, Venkatesan D, Murphree J. Auroral bright spots on the dayside oval.Journal of Geophysical Research: Space Physics (1978–2012).1989,94(A5),5515-5522.
[22] Liou K, Newell P, Meng CI, et al. Synoptic auroral distribution: A survey usingPolar ultraviolet imagery. Journal of Geophysical Research: Space Physics(1978–2012).1997,102(A12),27197-27205.
[23] Samadani R, Mihovilovic D, Clauer C, et al. Evaluation of an elastic curvetechnique for finding the auroral oval from satellite images automatically. IEEETransactions on Geoscience and Remote Sensing.1990,28,590-597.
[24] Daida J, Samadani R, Vesecky J. Adaptive Image Segmentation Applied ToExtracting The Auroral Oval From Satellite Images.1990. IEEE. p1649-1650.
[25] Samadani R, Mihovilovic D, Clauer CR, et al. A Computer Vision System forAutomatically Finding the Auroral Oval from Satellite Images. ImageProcessing Algorithms and Techniques, Proceedings of SPIE.1990,1244,68-75.
[26] Mihovilovic DA, Samadani R. Minimization-pruning algorithm for findingelliptical boundaries in images with nonconstant background and with missingdata.1992. International Society for Optics and Photonics. p546-557.
[27] Hung C, Germany G. K-means and iterative selection algorithms in imagesegmentation.2003.
[28] Li X, Ramachandran R, He M, et al. Comparing different thresholdingalgorithms for segmenting auroras.2004. IEEE. p594-601.
[29] Germany G, Parks G, Ranganath H, et al. Analysis of auroral morphology:Substorm precursor and onset on January10,1997. Geophysical researchletters.1998,25(15),3043-3046.
[30] Cao C, Newman T, Germany G. A Shape-Based Technique for Aurora OvalSegmentation From UVI Images.2005. p1285.
[31] Newman CCTS, Germany GA, Spann JF. A New Method for Auroral OvalDetection in FUV Imagery.2000.
[32] Cao C, Newman TS, Germany GA. New shape-based auroral ovalsegmentation driven by LLS-RHT. Pattern Recognition.2009,42(5),607-618.
[33] WANG Q, MENG Q, HU Z, et al. Extraction of auroral oval boundaries fromUVI images: A new FLICM clustering-based method and its evaluation.
[34] http://www.ava.fmi.fi/~syrjasuo/MV/.
[35] Syrj suo M, Kauristie K, Pulkkinen T. A search engine for auroral forms.Advances in Space Research.2001,28(11),1611-1616.
[36] Syrjasuo MT, Donovan EF, Cogger LL. Content-based retrieval of auroralimages-thousands of irregular shapes.2004. p224-228.
[37] Syrjasuo MT, Donovan EF. Using relevance feedback in retrieving auroralimages.2005.
[38] Kauristie K, Syrj suo M, Amm O, et al. A statistical study of evening sectorarcs and electrojets. Advances in Space Research.2001,28(11),1605-1610.
[39] Syrjasuo M, Donovan E, Qin X. Automatic classification of auroral images insubstorm studies.2007. p309-313.
[40] Syrjasuo M, Partamies N. Numeric image features for detection of aurora.Geoscience and Remote Sensing Letters, IEEE.2012,9(2),176-179.
[41] Baranoski GV, Rokne JG, Shirley P, et al. Simulating the aurora. The Journal ofVisualization and Computer Animation.2003,14(1),43-59.
[42] Baranoski GV, Wan J, Rokne JG, et al. Simulating the dynamics of auroralphenomena. ACM Transactions on Graphics (TOG).2005,24(1),37-59.
[43] Seviour R, Kosch M, Honary F. Identification of clouds and aurorae in opticaldata images. New Journal of Physics.2003,5(1),6.
[44] Kim SK, Ranganath HS. Content-Based Retrieval of Aurora Images Based onthe Hierarchical Representation.2010. Springer. p249-260.
[45] Biradar C, SB P. An Innovative Approach for Aurora Recognition. InternationalJournal of Engineering.2012,1(7).
[46] Whiter D, Gustavsson B, Partamies N, et al. A new automatic method forestimating the peak auroral emission height from all-sky camera images.Geoscientific Instrumentation, Methods and Data Systems Discussions.2012,2,893-924.
[47] Fu R, Li J, Gao X, et al. Automatic aurora images classification algorithmbased on separated texture.2009. IEEE. p1331-1335.
[48] Han S, Wu Z, Wu G, et al. Automatic Classification of Ultraviolet AuroraImages Based on Texture and Shape Features.2011. IEEE. p527-532.
[49] Gao X, Fu R, Li X, et al. Aurora image segmentation by combining patch andtexture thresholding. Computer Vision and Image Understanding.2011,115(3),390-402.
[50] Blixt EM, Semeter J, Ivchenko N. Optical flow analysis of the aurora borealis.Geoscience and Remote Sensing Letters, IEEE.2006,3(1),159-163.
[51] Syrj suo MT. Auroral monitoring network: From all-sky camera system toautomated image analysis: Finnish Meteorological Institute;2001.
[52] Feldstein Y, Elphinstone R. Aurorae and the large-scale structure of themagnetosphere. Journal of geomagnetism and geoelectricity.1992,44(12),1159-1174.
[53] Kullen A, Brittnacher M, Cumnock J, et al. Solar wind dependence of theoccurrence and motion of polar auroral arcs: A statistical study. Journal ofgeophysical research.2002,107(A11),1362.
[54] Knudsen D, Donovan E, Cogger L, et al. Width and structure of mesoscaleoptical auroral arcs. Geophysical research letters.2001,28(4),705-708.
[55] Thorolfsson A, Cerisier J-C, Lockwood M, et al. Simultaneous optical andradar signatures of poleward-moving auroral forms.2000. Springer. p1054-1066.
[56] Denig W, Burke W, Maynard N, et al. Ionospheric signatures of daysidemagnetopause transients: A case study using satellite and ground measurements.Journal of geophysical research.1993,98(A4),5969-5980.
[57] Milan S, Lester M, Cowley S, et al. Meridian-scanning photometer, coherentHF radar, and magnetometer observations of the cusp: a case study.1999.Springer. p159-172.
[58] Farrugia C, Sandholt P, Torbert R, et al. Temporal and spatial aspects of thecusp inferred from local and global ground‐and space‐based observations ina case study. Journal of Geophysical Research: Space Physics (1978–2012).2004,109(A4).
[59] Sandholt PE, Farrugia CJ, Moen J, et al. A classification of dayside auroralforms and activities as a function of interplanetary magnetic field orientation.Journal of Geophysical Research: Space Physics (1978–2012).1998,103(A10),23325-23345.
[60] Safargaleev V, Tagirov V, Osipenko S. Multiple poleward propagating auroralevents in the15-16MLT sector. Sodankyl Geophys Observatory Publ.2003,92,15-18.
[61] Rabiner LR. A tutorial on hidden Markov models and selected applications inspeech recognition. Proceedings of the IEEE.1989,77(2),257-286.
[62] Benítez M, Ramírez J, Segura JC, et al. Continuous HMM-based seismic-eventclassification at Deception Island, Antarctica. Geoscience and Remote Sensing,IEEE Transactions on.2007,45(1),138-146.
[63] Chen C, Liang J, Zhao H, et al. Gait recognition using Hidden Markov model.Advances in Natural Computation: Springer;2006. p399-407.
[64] Shafiei MM, Rabiee HR. A new online signature verification algorithm usingvariable length segmentation and hidden Markov models.2003. IEEE. p443-446.
[65] Van BL, Garcia-Salicetti S, Dorizzi B. On using the Viterbi path along withHMM likelihood information for online signature verification. Systems, Man,and Cybernetics, Part B: Cybernetics, IEEE Transactions on.2007,37(5),1237-1247.
[66] Hughey R, Krogh A. Hidden Markov models for sequence analysis: extensionand analysis of the basic method. Computer applications in the biosciences:CABIOS.1996,12(2),95-107.
[67] Mitra S, Acharya T. Gesture recognition: A survey. Systems, Man, andCybernetics, Part C: Applications and Reviews, IEEE Transactions on.2007,37(3),311-324.
[68] Yang M-H, Kriegman DJ, Ahuja N. Detecting faces in images: A survey.Pattern Analysis and Machine Intelligence, IEEE Transactions on.2002,24(1),34-58.
[69] Missaoui O, Frigui H, Gader P. Land-mine detection with ground-penetratingradar using multistream discrete hidden Markov models. Geoscience andRemote Sensing, IEEE Transactions on.2011,49(6),2080-2099.
[70] García-García D, Hernández EP, Diaz de Maria F. A new distance measure formodel-based sequence clustering. Pattern Analysis and Machine Intelligence,IEEE Transactions on.2009,31(7),1325-1331.
[71] Duda RO, Hart PE, Stork DG. Pattern classifcation. New York: John Wiley,Section.2001,10, l.
[72] Ortega-Garcia J, Fiérrez-Aguilar J, Martin-Rello J, et al. Complete signalmodeling and score normalization for function-based dynamic signatureverification.2003. Springer. p658-667.
[73] Bicego M, Murino V, Figueiredo MA. Similarity-based classification ofsequences using hidden Markov models. Pattern Recognition.2004,37(12),2281-2291.
[74] Porikli F. Clustering variable length sequences by eigenvector decompositionusing HMM. Structural, Syntactic, and Statistical Pattern Recognition: Springer;2004. p352-360.
[75] Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotationinvariant texture classification with local binary patterns. Pattern Analysis andMachine Intelligence, IEEE Transactions on.2002,24(7),971-987.
[76] Schwarz G. Estimating the dimension of a model. The annals of statistics.1978,6(2),461-464.
[77] Jelinek F. Continuous speech recognition by statistical methods. Proceedings ofthe IEEE.1976,64(4),532-556.
[78] Bakis R. Continuous speech word recognition via centi-second acoustic states.Proc ASA Meeting.1976.
[79] Chen C, Liang J, Zhao H, et al. Frame difference energy image for gaitrecognition with incomplete silhouettes. Pattern Recognition Letters.2009,30(11),977-984.
[80] Chen C, Liang J, Zhao H, et al. Factorial HMM and parallel HMM for gaitrecognition. Systems, Man, and Cybernetics, Part C: Applications and Reviews,IEEE Transactions on.2009,39(1),114-123.
[81] Chen C, Liang J, Zhu X. Gait recognition based on improved dynamicBayesian networks. Pattern Recognition.2011,44(4),988-995.
[82] Krishnapuram R, Keller JM. A possibilistic approach to clustering. FuzzySystems, IEEE Transactions on.1993,1(2),98-110.
[83] Krishnapuram R, Keller JM. The possibilistic c-means algorithm: insights andrecommendations. Fuzzy Systems, IEEE Transactions on.1996,4(3),385-393.
[84] Pal NR, Pal K, Keller JM, et al. A possibilistic fuzzy c-means clusteringalgorithm. Fuzzy Systems, IEEE Transactions on.2005,13(4),517-530.
[85] Pal NR, Pal K, Bezdek JC. A mixed c-means clustering model.1997. IEEE. p11-21.
[86] P kalska E, Duin RP. Automatic pattern recognition by similarityrepresentations. Electronics Letters.2001,37(3),159-160.
[87] P kalska E, Duin RP. Dissimilarity representations allow for building goodclassifiers. Pattern Recognition Letters.2002,23(8),943-956.
[88] Pekalska E, Paclik P, Duin RP. A generalized kernel approach todissimilarity-based classification. The Journal of Machine Learning Research.2002,2,175-211.
[89] Bicego M. Odor classification using similarity-based representation. Sensorsand Actuators B: Chemical.2005,110(2),225-230.
[90] Milone DH, Di Persia LE. An EM algorithm to learn sequences in the waveletdomain. MICAI2007: Advances in Artificial Intelligence: Springer;2007. p518-528.
[91] Theodoridis S. L. koutroumbas. Pattern Recognition.1999.
[92] Kohavi R, Provost F. Glossary of terms. Machine Learning.1998,30(2-3),271-274.
[93] Sin B, Kim JH. Nonstationary hidden Markov model. Signal Processing.1995,46(1),31-46.
[94] Fasel GJ. Dayside poleward moving auroral forms: A statistical study. Journalof geophysical research.1995,100(A7),11891-11811,11905.
[95] Karlson K, ieroset M, Moen J, et al. A statistical study of flux transfer eventsignatures in the dayside aurora: The IMF By-related prenoon-postnoonsymmetry. Journal of geophysical research.1996,101(A1),59-68.
[96] Sandholt P, Deehr C, Egeland A, et al. Signatures in the dayside aurora ofplasma transfer from the magnetosheath. Journal of Geophysical Research:Space Physics (1978–2012).1986,91(A9),10063-10079.
[97] Sandholt P, Lockwood M, Oguti T, et al. Midday auroral breakup events andrelated energy and momentum transfer from the magnetosheath. Journal ofGeophysical Research: Space Physics (1978–2012).1990,95(A2),1039-1060.
[98] Sandholt P, Farrugia C. Role of poleward moving auroral forms in thedawn-dusk auroral precipitation asymmetries induced by IMF By. Journal ofgeophysical research.2007,112(A4), A04203.
[99] Sandholt P, Farrugia C. Poleward moving auroral forms (PMAFs) revisited:responses of aurorae, plasma convection and Birkeland currents in the pre-andpostnoon sectors under positive and negative IMF B yconditions.2007. p1629-1652.
[100] Vorobjev V, Gustafsson G, Starkov G, et al. Dynamics of day and night auroraduring substorms. Planetary and Space Science.1975,23(2),269-278.
[101] Zhang QH, Dunlop M, Lockwood M, et al. Simultaneous observations ofreconnection pulses at Cluster and their effects on the cusp aurora observed atthe Chinese Yellow River Station. Journal of Geophysical Research: SpacePhysics (1978–2012).2010,115(A10).
[102] Kozlovsky A, Kangas J. Motion and origin of noon high‐latitude polewardmoving auroral arcs on closed magnetic field lines. Journal of GeophysicalResearch: Space Physics (1978–2012).2002,107(A2), SMP1-1-SMP1-13.
[103] Milan S, Yeoman T, Lester M, et al. Post-noon two-minute period pulsatingaurora and their relationship to the dayside convection pattern.1999. Springer.p877-891.
[104] Yang Q, Wu X.10challenging problems in data mining research. InternationalJournal of Information Technology&Decision Making.2006,5(04),597-604.
[105] Barjatya A. Block matching algorithms for motion estimation. IEEETransactions Evolution Computation.2004,8(3),225-239.
[106] Kotsiantis S, Kanellopoulos D, Pintelas P. Handling imbalanced datasets: Areview. GESTS International Transactions on Computer Science andEngineering.2006,30(1),25-36.
[107] Vapnik V. Statistical learning theory.1998. Wiley, New York;1998.
[108] Akbani R, Kwek S, Japkowicz N. Applying support vector machines toimbalanced datasets. Machine Learning: ECML2004: Springer;2004. p39-50.
[109] Veropoulos K, Campbell C, Cristianini N. Controlling the sensitivity of supportvector machines.1999. Citeseer. p55-60.
[110] Florin R, Militaru D. A HMM/SVM hybrid method for speaker verification.2010. IEEE. p111-114.
[111] Huang B, Du C, Zhang Y, et al. A hybrid hmm-svm method for onlinehandwriting symbol recognition.2006. IEEE. p887-891.
[112] Lee H, Choi S. Pca+hmm+svm for eeg pattern classification.2003. IEEE. p541-544.
[113] Sloin A, Burshtein D. Support vector machine training for improved hiddenmarkov modeling. Signal Processing, IEEE Transactions on.2008,56(1),172-188.
[114] Wan W, Liu H, Wang L, et al. A hybrid HMM/SVM classifier for motionrecognition using μIMU data.2007. IEEE. p115-120.
[115] Callut J, Dupont P. Fβ support vector machines.2005. p1443-1448.
[116] Brefeld U, Scheffer T. AUC maximizing support vector learning.2005.Citeseer.
[117] Joachims T. A support vector method for multivariate performance measures.2005. ACM. p377-384.
[118] Lin Z, Hao Z, Yang X. Effects of several evaluation metrics on imbalanced datalearning. Journal of South China University of Technology (Natural ScienceEdition).2010,38,147-154.
[119] Yuan F. A fast accumulative motion orientation model based on integral imagefor video smoke detection. Pattern Recognition Letters.2008,29(7),925-932.
[120] Ferri C, Hernández-Orallo J, Modroiu R. An experimental comparison ofperformance measures for classification. Pattern Recognition Letters.2009,30(1),27-38.
[121] Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters.2006,27(8),861-874.
[122] Hand DJ, Till RJ. A simple generalisation of the area under the ROC curve formultiple class classification problems. Machine Learning.2001,45(2),171-186.
[123] Morik K, Brockhausen P, Joachims T. Combining statistical learning with aknowledge-based approach-a case study in intensive care monitoring.1999.MORGAN KAUFMANN PUBLISHERS, INC. p268-277.
[124] Chang C-C, Lin C-J. LIBSVM: a library for support vector machines. ACMTransactions on Intelligent Systems and Technology (TIST).2011,2(3),27.
[125] Wang BX, Japkowicz N. Boosting support vector machines for imbalanced datasets. Knowledge and Information Systems.2010,25(1),1-20.
[126] Davis RI, Lovell BC, Caelli T. Improved estimation of hidden markov modelparameters from multiple observation sequences.2002. IEEE. p168-171.
[127] Cogger L, Murphree J, Ismail S, et al. Characteristics of dayside5577and3914aurora. Geophysical research letters.1977,4(10),413-416.
[128] Sandholt P, Farrugia C. Does the aurora provide evidence for the occurrence ofantiparallel magnetopause reconnection? Journal of Geophysical Research:Space Physics (1978–2012).2003,108(A12).
[129]李柳元,曹晋滨,周国成.磁层相对论电子通量变化与磁暴/亚暴的关系.地球物理学报.2006,49(1),9-15.
[130] Li L, Cao J, Zhou G, et al. Statistical roles of storms and substorms in changingthe entire outer zone relativistic electron population. Journal of GeophysicalResearch: Space Physics (1978–2012).2009,114(A12).
[131] Sutcliffe P. Substorm onset identification using neural networks and Pi2pulsations.1997. Springer. p1257-1264.
[132] Murphy KR, Rae IJ, Mann IR, et al. Wavelet‐based ULF wave diagnosis ofsubstorm expansion phase onset. Journal of Geophysical Research: SpacePhysics (1978–2012).2009,114(A1).
[133] Tokunaga T, Yumoto K, Uozumi T. Identification of full-substorm onset fromground-magnetometer data by singular value transformation.2011.
[134] Kataoka R, Miyoshi Y, Morioka A. Hilbert‐Huang Transform of geomagneticpulsations at auroral expansion onset. Journal of Geophysical Research: SpacePhysics (1978–2012).2009,114(A9).
[135] Meng C-I, Liou K. Substorm timings and timescales: A new aspect. Spacescience reviews.2004,113(1-2),41-75.
[136] Liou K, Meng C-I, Lui T, et al. On relative timing in substorm onset signatures.Journal of geophysical research.1999,104(A10),22807-22822,22817.
[137] Liu JM, Zhang BC, Kamide Y, et al. Spontaneous and trigger‐associatedsubstorms compared: Electrodynamic parameters in the polar ionosphere.Journal of Geophysical Research: Space Physics (1978–2012).2011,116(A1).
[138] Frey H, Mende S. Substorm onsets as observed by IMAGE-FUV.2006. p71-76.
[139] Kullen A, Karlsson T. On the relation between solar wind, pseudobreakups, andsubstorms. Journal of Geophysical Research: Space Physics (1978–2012).2004,109(A12).
[140] McPherron RL. Substorm related changes in the geomagnetic tail: The growthphase. Planetary and Space Science.1972,20(9),1521-1539.
[141] Deng Y, Kenney C, Moore MS, et al. Peer group filtering and perceptual colorimage quantization.1999. IEEE. p21-24.
[142] Bezdek JC. Pattern recognition with fuzzy objective function algorithms:Kluwer Academic Publishers;1981.
[143] Bezdek JC, Ehrlich R, Full W. FCM: The fuzzy c-means clusteringalgorithm. Computers&Geosciences.1984,10(2),191-203.
[144] Chuang K-S, Tzeng H-L, Chen S, et al. Fuzzy c-means clustering with spatialinformation for image segmentation. computerized medical imaging andgraphics.2006,30(1),9-15.
[145] Sun H, Wang S, Jiang Q. FCM-based model selection algorithms fordetermining the number of clusters. Pattern Recognition.2004,37(10),2027-2037.
[146] Wang Y, Li C, Zuo Y. A selection model for optimal fuzzy clustering algorithmand number of clusters based on competitive comprehensive fuzzy evaluation.Fuzzy Systems, IEEE Transactions on.2009,17(3),568-577.
[147] Kisabeth JL, Rostoker G. The expansive phase of magnetospheric substorms:1.Development of the auroral electrojets and auroral arc configuration during asubstorm. Journal of geophysical research.1974,79(7),972-984.
[148] Liou K, Meng CI, Wu CC. On the interplanetary magnetic field By control ofsubstorm bulge expansion. Journal of Geophysical Research: Space Physics(1978–2012).2006,111(A9).
[149] Voronkov I, Donovan E, Samson J. Observations of the phases of the substorm.Journal of geophysical research.2003,108(A2),1073.
[150] Newell P, Gjerloev J. Evaluation of SuperMAG auroral electrojet indices asindicators of substorms and auroral power. Journal of Geophysical Research:Space Physics (1978–2012).2011,116(A12).
[151] Newell P, Gjerloev J. Substorm and magnetosphere characteristic scalesinferred from the SuperMAG auroral electrojet indices. Journal of GeophysicalResearch: Space Physics (1978–2012).2011,116(A12).
[152] Ohtani S, Anderson B, Sibeck D, et al. A multisatellite study of a pseudo‐substorm onset in the near‐Earth magnetotail. Journal of GeophysicalResearch: Space Physics (1978–2012).1993,98(A11),19355-19367.
[153] Rostoker G. On the place of the pseudo‐breakup in a magnetosphericsubstorm. Geophysical research letters.1998,25(2),217-220.
[154] Nakamura R, Baker DN, Yamamoto T, et al. Particle and field signatures duringpseudobreakup and major expansion onset. Journal of geophysical research.1994,99(A1),207-221.
[155] Aikio A, Sergeev V, Shukhtina M, et al. Characteristics of pseudobreakups andsubstorms observed in the ionosphere, at the geosynchronous orbit, and in themidtail. Journal of Geophysical Research: Space Physics (1978–2012).1999,104(A6),12263-12287.
[156] Kullen A, Karlsson T, Cumnock JA, et al. Occurrence and properties ofsubstorms associated with pseudobreakups. Journal of Geophysical Research:Space Physics (1978–2012).2010,115(A12).
[157] Boakes PD. Investigating the relationship between open magnetic flux and thesubstorm cycle;2010.
[158] Carbary JF. A Kp‐based model of auroral boundaries. Space Weather.2005,3(10).
[159] Zhang Y, Paxton L. An empirical Kp-dependent global auroral model based onTIMED/GUVI FUV data. Journal of Atmospheric and Solar-Terrestrial Physics.2008,70(8),1231-1242.
[160] Holzworth R, Meng CI. Mathematical representation of the auroral oval.Geophysical research letters.1975,2(9),377-380.
[161] Holzworth R, Meng C-I. Auroral boundary variations and the interplanetarymagnetic field. Planetary and Space Science.1984,32(1),25-29.
[162] Kauristie K. Statistical fits for auroral oval boundaries during the substormsequence. Journal of geophysical research.1995,100(A11),21885-21821,21895.
[163] Sigernes F, Dyrland M, Brekke P, et al. Real time aurora ovalforecasting-SvalTrackII. In other words.2011,3(3),3.
[164] Sigernes F, Dyrland M, Brekke P, et al. Two methods to forecast auroraldisplays. Journal of Space Weather and Space Climate.2011,1(1).
[165] Starkov G. Mathematical model of the auroral boundaries. GEOMAGNETISMAND AERONOMY C/C OF GEOMAGNETIZM I AERONOMIIA.1994,34,331-336.
[166] Starkov GV. Analiticheskoe predstavlenie ekvatorial’noy granitsy oval’noyzony polyarnikh siyaniy (Analytical representation of the equatorwardboundary of the polar auroral oval). Geomagn Aeron.1969,9,759.
[167] Hardy D, Burke W, Gussenhoven M, et al. DMSP/F2electron observations ofequatorward auroral boundaries and their relationship to the solar wind velocityand the north‐south component of the interplanetary magnetic field. Journalof Geophysical Research: Space Physics (1978–2012).1981,86(A12),9961-9974.
[168] Cowley S, Morelli J, Lockwood M. Dependence of convective flows andparticle precipitation in the high‐latitude dayside ionosphere on the x and ycomponents of the interplanetary magnetic field. Journal of GeophysicalResearch: Space Physics (1978–2012).1991,96(A4),5557-5564.
[169] Liou K, Newell P, Sibeck D, et al. Observation of IMF and seasonal effects inthe location of auroral substorm onset. Journal of Geophysical Research: SpacePhysics (1978–2012).2001,106(A4),5799-5810.
[170] Meng C-I. Polar cap variations and the interplanetary magnetic field. Dynamicsof the Magnetosphere: Springer;1979. p23-46.
[171] Yang Y, Lu J, Wang JS, et al. Influence of interplanetary magnetic field andsolar wind on auroral brightness in different regions. Journal of GeophysicalResearch: Space Physics.2013.
[172] Lee DY, Lyons L, Yumoto K. Sawtooth oscillations directly driven by solarwind dynamic pressure enhancements. Journal of Geophysical Research: SpacePhysics (1978–2012).2004,109(A4).
[173] Lyons L. Geomagnetic disturbances: Characteristics of, distinction betweentypes, and relations to interplanetary conditions. Journal of Atmospheric andSolar-Terrestrial Physics.2000,62(12),1087-1114.
[174] Shue J-H, Newell P, Liou K, et al. Solar wind density and velocity control ofauroral brightness under normal interplanetary magnetic field conditions.Journal of geophysical research.2002,107(A12),1428.
[175] Cho J-S, Lee D-Y, Kim K-C, et al. Response of the Poleward Boundary of theNightside Auroral Oval to Impacts of Solar Wind Dynamic PressureEnhancement. Journal of Astronomy and Space Sciences.2010,27,189-194.
[176] Brittnacher M, Fillingim M, Parks G, et al. Polar cap area and boundary motionduring substorms. Journal of Geophysical Research: Space Physics(1978–2012).1999,104(A6),12251-12262.
[177] Liou K, Newell P, Meng CI, et al. Characteristics of the solar wind controlledauroral emissions. Journal of Geophysical Research: Space Physics(1978–2012).1998,103(A8),17543-17557.
[178] Troshichev O, Andrezen V, Vennerstr m S, et al. Magnetic activity in the polarcap—A new index. Planetary and Space Science.1988,36(11),1095-1102.
[179] Kan J, Lee L. Energy coupling function and solar wind‐magnetospheredynamo. Geophysical research letters.1979,6(7),577-580.
[180] Hu ZJ, Yang HG, Han DS, et al. Dayside auroral emissions controlled by IMF:A survey for dayside auroral excitation at557.7and630.0nm in Ny‐lesund,Svalbard. Journal of Geophysical Research: Space Physics (1978–2012).2012,117(A2).
[181] Sykes A. An introduction to regression analysis: Law School, University ofChicago;1993.
[2]胡泽骏,杨惠根,艾勇,等.日侧极光卵的可见光多波段观测特征.极地研究.2005,17(2),107-114.
[3] http://sd-www.jhuapl.edu/Aurora/nightb/nightb.html.
[4] Liou K. Polar Ultraviolet Imager observation of auroral breakup. Journal ofGeophysical Research: Space Physics (1978–2012).2010,115(A12).
[5] Torr M, Torr D, Zukic M, et al. A far ultraviolet imager for the internationalsolar-terrestrial physics mission. Space science reviews.1995,71(1-4),329-383.
[6] Yang H, Sato N, Makita K, et al. Synoptic observations of auroras along thepostnoon oval: A survey with all-sky TV observations at Zhongshan, Antarctica.Journal of Atmospheric and Solar-Terrestrial Physics.2000,62(9),787-797.
[7] Sandholt PE, Carlson HC, Egeland A. Dayside and polar cap aurora: KluwerAcademic Pub;2002.
[8]胡红桥,刘瑞源,王敬芳,等.南极中山站极光形态的统计特征.极地研究.1999,11(1),8-18.
[9] Hu Z-J, Yang H, Huang D, et al. Synoptic distribution of dayside aurora:Multiple-wavelength all-sky observation at Yellow River Station inNy-lesund, Svalbard. Journal of Atmospheric and Solar-Terrestrial Physics.2009,71(8),794-804.
[10] Wang Q, Liang J, Hu Z-J, et al. Spatial texture based automatic classification ofdayside aurora in all-sky images. Journal of Atmospheric and Solar-TerrestrialPhysics.2010,72(5),498-508.
[11] Dungey JW. Interplanetary magnetic field and the auroral zones. PhysicalReview Letters.1961,6(2),47-48.
[12] Lemaire J, Roth M. Penetration of solar wind plasma elements into themagnetosphere. Journal of Atmospheric and Terrestrial Physics.1978,40(3),331-335.
[13] Wei C, Lee L, Belle‐Hamer L. A simulation study of the vortex structure inthe low‐latitude boundary layer. Journal of Geophysical Research: SpacePhysics (1978–2012).1990,95(A12),20793-20807.
[14] Xing Z, Yang H, Han D, et al. Poleward moving auroral forms (PMAFs)observed at the Yellow River Station: A statistical study of its dependence onthe solar wind conditions. Journal of Atmospheric and Solar-Terrestrial Physics.2012.
[15] Feldstein YI, Starkov G. Dynamics of auroral belt and polar geomagneticdisturbances. Planetary and Space Science.1967,15(2),209-229.
[16] Frank LA, Craven JD, Burch J, et al. Polar views of the Earth's aurora withDynamics Explorer. Geophysical research letters.1982,9(9),1001-1004.
[17] Akasofu S-I. The development of the auroral substorm. Planetary and SpaceScience.1964,12(4),273-282.
[18] Akasofu SI, Kan J. Dayside and nightside auroral arc systems. Geophysicalresearch letters.1980,7(10),753-756.
[19] Frey H, Mende S, Angelopoulos V, et al. Substorm onset observations byIMAGE‐FUV. Journal of Geophysical Research: Space Physics (1978–2012).2004,109(A10).
[20] Murphree J, Cogger L, Anger C, et al. Large scale6300,5577,3914dayside auroral morphology. Geophysical research letters.1980,7(4),239-242.
[21] Lui A, Venkatesan D, Murphree J. Auroral bright spots on the dayside oval.Journal of Geophysical Research: Space Physics (1978–2012).1989,94(A5),5515-5522.
[22] Liou K, Newell P, Meng CI, et al. Synoptic auroral distribution: A survey usingPolar ultraviolet imagery. Journal of Geophysical Research: Space Physics(1978–2012).1997,102(A12),27197-27205.
[23] Samadani R, Mihovilovic D, Clauer C, et al. Evaluation of an elastic curvetechnique for finding the auroral oval from satellite images automatically. IEEETransactions on Geoscience and Remote Sensing.1990,28,590-597.
[24] Daida J, Samadani R, Vesecky J. Adaptive Image Segmentation Applied ToExtracting The Auroral Oval From Satellite Images.1990. IEEE. p1649-1650.
[25] Samadani R, Mihovilovic D, Clauer CR, et al. A Computer Vision System forAutomatically Finding the Auroral Oval from Satellite Images. ImageProcessing Algorithms and Techniques, Proceedings of SPIE.1990,1244,68-75.
[26] Mihovilovic DA, Samadani R. Minimization-pruning algorithm for findingelliptical boundaries in images with nonconstant background and with missingdata.1992. International Society for Optics and Photonics. p546-557.
[27] Hung C, Germany G. K-means and iterative selection algorithms in imagesegmentation.2003.
[28] Li X, Ramachandran R, He M, et al. Comparing different thresholdingalgorithms for segmenting auroras.2004. IEEE. p594-601.
[29] Germany G, Parks G, Ranganath H, et al. Analysis of auroral morphology:Substorm precursor and onset on January10,1997. Geophysical researchletters.1998,25(15),3043-3046.
[30] Cao C, Newman T, Germany G. A Shape-Based Technique for Aurora OvalSegmentation From UVI Images.2005. p1285.
[31] Newman CCTS, Germany GA, Spann JF. A New Method for Auroral OvalDetection in FUV Imagery.2000.
[32] Cao C, Newman TS, Germany GA. New shape-based auroral ovalsegmentation driven by LLS-RHT. Pattern Recognition.2009,42(5),607-618.
[33] WANG Q, MENG Q, HU Z, et al. Extraction of auroral oval boundaries fromUVI images: A new FLICM clustering-based method and its evaluation.
[34] http://www.ava.fmi.fi/~syrjasuo/MV/.
[35] Syrj suo M, Kauristie K, Pulkkinen T. A search engine for auroral forms.Advances in Space Research.2001,28(11),1611-1616.
[36] Syrjasuo MT, Donovan EF, Cogger LL. Content-based retrieval of auroralimages-thousands of irregular shapes.2004. p224-228.
[37] Syrjasuo MT, Donovan EF. Using relevance feedback in retrieving auroralimages.2005.
[38] Kauristie K, Syrj suo M, Amm O, et al. A statistical study of evening sectorarcs and electrojets. Advances in Space Research.2001,28(11),1605-1610.
[39] Syrjasuo M, Donovan E, Qin X. Automatic classification of auroral images insubstorm studies.2007. p309-313.
[40] Syrjasuo M, Partamies N. Numeric image features for detection of aurora.Geoscience and Remote Sensing Letters, IEEE.2012,9(2),176-179.
[41] Baranoski GV, Rokne JG, Shirley P, et al. Simulating the aurora. The Journal ofVisualization and Computer Animation.2003,14(1),43-59.
[42] Baranoski GV, Wan J, Rokne JG, et al. Simulating the dynamics of auroralphenomena. ACM Transactions on Graphics (TOG).2005,24(1),37-59.
[43] Seviour R, Kosch M, Honary F. Identification of clouds and aurorae in opticaldata images. New Journal of Physics.2003,5(1),6.
[44] Kim SK, Ranganath HS. Content-Based Retrieval of Aurora Images Based onthe Hierarchical Representation.2010. Springer. p249-260.
[45] Biradar C, SB P. An Innovative Approach for Aurora Recognition. InternationalJournal of Engineering.2012,1(7).
[46] Whiter D, Gustavsson B, Partamies N, et al. A new automatic method forestimating the peak auroral emission height from all-sky camera images.Geoscientific Instrumentation, Methods and Data Systems Discussions.2012,2,893-924.
[47] Fu R, Li J, Gao X, et al. Automatic aurora images classification algorithmbased on separated texture.2009. IEEE. p1331-1335.
[48] Han S, Wu Z, Wu G, et al. Automatic Classification of Ultraviolet AuroraImages Based on Texture and Shape Features.2011. IEEE. p527-532.
[49] Gao X, Fu R, Li X, et al. Aurora image segmentation by combining patch andtexture thresholding. Computer Vision and Image Understanding.2011,115(3),390-402.
[50] Blixt EM, Semeter J, Ivchenko N. Optical flow analysis of the aurora borealis.Geoscience and Remote Sensing Letters, IEEE.2006,3(1),159-163.
[51] Syrj suo MT. Auroral monitoring network: From all-sky camera system toautomated image analysis: Finnish Meteorological Institute;2001.
[52] Feldstein Y, Elphinstone R. Aurorae and the large-scale structure of themagnetosphere. Journal of geomagnetism and geoelectricity.1992,44(12),1159-1174.
[53] Kullen A, Brittnacher M, Cumnock J, et al. Solar wind dependence of theoccurrence and motion of polar auroral arcs: A statistical study. Journal ofgeophysical research.2002,107(A11),1362.
[54] Knudsen D, Donovan E, Cogger L, et al. Width and structure of mesoscaleoptical auroral arcs. Geophysical research letters.2001,28(4),705-708.
[55] Thorolfsson A, Cerisier J-C, Lockwood M, et al. Simultaneous optical andradar signatures of poleward-moving auroral forms.2000. Springer. p1054-1066.
[56] Denig W, Burke W, Maynard N, et al. Ionospheric signatures of daysidemagnetopause transients: A case study using satellite and ground measurements.Journal of geophysical research.1993,98(A4),5969-5980.
[57] Milan S, Lester M, Cowley S, et al. Meridian-scanning photometer, coherentHF radar, and magnetometer observations of the cusp: a case study.1999.Springer. p159-172.
[58] Farrugia C, Sandholt P, Torbert R, et al. Temporal and spatial aspects of thecusp inferred from local and global ground‐and space‐based observations ina case study. Journal of Geophysical Research: Space Physics (1978–2012).2004,109(A4).
[59] Sandholt PE, Farrugia CJ, Moen J, et al. A classification of dayside auroralforms and activities as a function of interplanetary magnetic field orientation.Journal of Geophysical Research: Space Physics (1978–2012).1998,103(A10),23325-23345.
[60] Safargaleev V, Tagirov V, Osipenko S. Multiple poleward propagating auroralevents in the15-16MLT sector. Sodankyl Geophys Observatory Publ.2003,92,15-18.
[61] Rabiner LR. A tutorial on hidden Markov models and selected applications inspeech recognition. Proceedings of the IEEE.1989,77(2),257-286.
[62] Benítez M, Ramírez J, Segura JC, et al. Continuous HMM-based seismic-eventclassification at Deception Island, Antarctica. Geoscience and Remote Sensing,IEEE Transactions on.2007,45(1),138-146.
[63] Chen C, Liang J, Zhao H, et al. Gait recognition using Hidden Markov model.Advances in Natural Computation: Springer;2006. p399-407.
[64] Shafiei MM, Rabiee HR. A new online signature verification algorithm usingvariable length segmentation and hidden Markov models.2003. IEEE. p443-446.
[65] Van BL, Garcia-Salicetti S, Dorizzi B. On using the Viterbi path along withHMM likelihood information for online signature verification. Systems, Man,and Cybernetics, Part B: Cybernetics, IEEE Transactions on.2007,37(5),1237-1247.
[66] Hughey R, Krogh A. Hidden Markov models for sequence analysis: extensionand analysis of the basic method. Computer applications in the biosciences:CABIOS.1996,12(2),95-107.
[67] Mitra S, Acharya T. Gesture recognition: A survey. Systems, Man, andCybernetics, Part C: Applications and Reviews, IEEE Transactions on.2007,37(3),311-324.
[68] Yang M-H, Kriegman DJ, Ahuja N. Detecting faces in images: A survey.Pattern Analysis and Machine Intelligence, IEEE Transactions on.2002,24(1),34-58.
[69] Missaoui O, Frigui H, Gader P. Land-mine detection with ground-penetratingradar using multistream discrete hidden Markov models. Geoscience andRemote Sensing, IEEE Transactions on.2011,49(6),2080-2099.
[70] García-García D, Hernández EP, Diaz de Maria F. A new distance measure formodel-based sequence clustering. Pattern Analysis and Machine Intelligence,IEEE Transactions on.2009,31(7),1325-1331.
[71] Duda RO, Hart PE, Stork DG. Pattern classifcation. New York: John Wiley,Section.2001,10, l.
[72] Ortega-Garcia J, Fiérrez-Aguilar J, Martin-Rello J, et al. Complete signalmodeling and score normalization for function-based dynamic signatureverification.2003. Springer. p658-667.
[73] Bicego M, Murino V, Figueiredo MA. Similarity-based classification ofsequences using hidden Markov models. Pattern Recognition.2004,37(12),2281-2291.
[74] Porikli F. Clustering variable length sequences by eigenvector decompositionusing HMM. Structural, Syntactic, and Statistical Pattern Recognition: Springer;2004. p352-360.
[75] Ojala T, Pietikainen M, Maenpaa T. Multiresolution gray-scale and rotationinvariant texture classification with local binary patterns. Pattern Analysis andMachine Intelligence, IEEE Transactions on.2002,24(7),971-987.
[76] Schwarz G. Estimating the dimension of a model. The annals of statistics.1978,6(2),461-464.
[77] Jelinek F. Continuous speech recognition by statistical methods. Proceedings ofthe IEEE.1976,64(4),532-556.
[78] Bakis R. Continuous speech word recognition via centi-second acoustic states.Proc ASA Meeting.1976.
[79] Chen C, Liang J, Zhao H, et al. Frame difference energy image for gaitrecognition with incomplete silhouettes. Pattern Recognition Letters.2009,30(11),977-984.
[80] Chen C, Liang J, Zhao H, et al. Factorial HMM and parallel HMM for gaitrecognition. Systems, Man, and Cybernetics, Part C: Applications and Reviews,IEEE Transactions on.2009,39(1),114-123.
[81] Chen C, Liang J, Zhu X. Gait recognition based on improved dynamicBayesian networks. Pattern Recognition.2011,44(4),988-995.
[82] Krishnapuram R, Keller JM. A possibilistic approach to clustering. FuzzySystems, IEEE Transactions on.1993,1(2),98-110.
[83] Krishnapuram R, Keller JM. The possibilistic c-means algorithm: insights andrecommendations. Fuzzy Systems, IEEE Transactions on.1996,4(3),385-393.
[84] Pal NR, Pal K, Keller JM, et al. A possibilistic fuzzy c-means clusteringalgorithm. Fuzzy Systems, IEEE Transactions on.2005,13(4),517-530.
[85] Pal NR, Pal K, Bezdek JC. A mixed c-means clustering model.1997. IEEE. p11-21.
[86] P kalska E, Duin RP. Automatic pattern recognition by similarityrepresentations. Electronics Letters.2001,37(3),159-160.
[87] P kalska E, Duin RP. Dissimilarity representations allow for building goodclassifiers. Pattern Recognition Letters.2002,23(8),943-956.
[88] Pekalska E, Paclik P, Duin RP. A generalized kernel approach todissimilarity-based classification. The Journal of Machine Learning Research.2002,2,175-211.
[89] Bicego M. Odor classification using similarity-based representation. Sensorsand Actuators B: Chemical.2005,110(2),225-230.
[90] Milone DH, Di Persia LE. An EM algorithm to learn sequences in the waveletdomain. MICAI2007: Advances in Artificial Intelligence: Springer;2007. p518-528.
[91] Theodoridis S. L. koutroumbas. Pattern Recognition.1999.
[92] Kohavi R, Provost F. Glossary of terms. Machine Learning.1998,30(2-3),271-274.
[93] Sin B, Kim JH. Nonstationary hidden Markov model. Signal Processing.1995,46(1),31-46.
[94] Fasel GJ. Dayside poleward moving auroral forms: A statistical study. Journalof geophysical research.1995,100(A7),11891-11811,11905.
[95] Karlson K, ieroset M, Moen J, et al. A statistical study of flux transfer eventsignatures in the dayside aurora: The IMF By-related prenoon-postnoonsymmetry. Journal of geophysical research.1996,101(A1),59-68.
[96] Sandholt P, Deehr C, Egeland A, et al. Signatures in the dayside aurora ofplasma transfer from the magnetosheath. Journal of Geophysical Research:Space Physics (1978–2012).1986,91(A9),10063-10079.
[97] Sandholt P, Lockwood M, Oguti T, et al. Midday auroral breakup events andrelated energy and momentum transfer from the magnetosheath. Journal ofGeophysical Research: Space Physics (1978–2012).1990,95(A2),1039-1060.
[98] Sandholt P, Farrugia C. Role of poleward moving auroral forms in thedawn-dusk auroral precipitation asymmetries induced by IMF By. Journal ofgeophysical research.2007,112(A4), A04203.
[99] Sandholt P, Farrugia C. Poleward moving auroral forms (PMAFs) revisited:responses of aurorae, plasma convection and Birkeland currents in the pre-andpostnoon sectors under positive and negative IMF B yconditions.2007. p1629-1652.
[100] Vorobjev V, Gustafsson G, Starkov G, et al. Dynamics of day and night auroraduring substorms. Planetary and Space Science.1975,23(2),269-278.
[101] Zhang QH, Dunlop M, Lockwood M, et al. Simultaneous observations ofreconnection pulses at Cluster and their effects on the cusp aurora observed atthe Chinese Yellow River Station. Journal of Geophysical Research: SpacePhysics (1978–2012).2010,115(A10).
[102] Kozlovsky A, Kangas J. Motion and origin of noon high‐latitude polewardmoving auroral arcs on closed magnetic field lines. Journal of GeophysicalResearch: Space Physics (1978–2012).2002,107(A2), SMP1-1-SMP1-13.
[103] Milan S, Yeoman T, Lester M, et al. Post-noon two-minute period pulsatingaurora and their relationship to the dayside convection pattern.1999. Springer.p877-891.
[104] Yang Q, Wu X.10challenging problems in data mining research. InternationalJournal of Information Technology&Decision Making.2006,5(04),597-604.
[105] Barjatya A. Block matching algorithms for motion estimation. IEEETransactions Evolution Computation.2004,8(3),225-239.
[106] Kotsiantis S, Kanellopoulos D, Pintelas P. Handling imbalanced datasets: Areview. GESTS International Transactions on Computer Science andEngineering.2006,30(1),25-36.
[107] Vapnik V. Statistical learning theory.1998. Wiley, New York;1998.
[108] Akbani R, Kwek S, Japkowicz N. Applying support vector machines toimbalanced datasets. Machine Learning: ECML2004: Springer;2004. p39-50.
[109] Veropoulos K, Campbell C, Cristianini N. Controlling the sensitivity of supportvector machines.1999. Citeseer. p55-60.
[110] Florin R, Militaru D. A HMM/SVM hybrid method for speaker verification.2010. IEEE. p111-114.
[111] Huang B, Du C, Zhang Y, et al. A hybrid hmm-svm method for onlinehandwriting symbol recognition.2006. IEEE. p887-891.
[112] Lee H, Choi S. Pca+hmm+svm for eeg pattern classification.2003. IEEE. p541-544.
[113] Sloin A, Burshtein D. Support vector machine training for improved hiddenmarkov modeling. Signal Processing, IEEE Transactions on.2008,56(1),172-188.
[114] Wan W, Liu H, Wang L, et al. A hybrid HMM/SVM classifier for motionrecognition using μIMU data.2007. IEEE. p115-120.
[115] Callut J, Dupont P. Fβ support vector machines.2005. p1443-1448.
[116] Brefeld U, Scheffer T. AUC maximizing support vector learning.2005.Citeseer.
[117] Joachims T. A support vector method for multivariate performance measures.2005. ACM. p377-384.
[118] Lin Z, Hao Z, Yang X. Effects of several evaluation metrics on imbalanced datalearning. Journal of South China University of Technology (Natural ScienceEdition).2010,38,147-154.
[119] Yuan F. A fast accumulative motion orientation model based on integral imagefor video smoke detection. Pattern Recognition Letters.2008,29(7),925-932.
[120] Ferri C, Hernández-Orallo J, Modroiu R. An experimental comparison ofperformance measures for classification. Pattern Recognition Letters.2009,30(1),27-38.
[121] Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters.2006,27(8),861-874.
[122] Hand DJ, Till RJ. A simple generalisation of the area under the ROC curve formultiple class classification problems. Machine Learning.2001,45(2),171-186.
[123] Morik K, Brockhausen P, Joachims T. Combining statistical learning with aknowledge-based approach-a case study in intensive care monitoring.1999.MORGAN KAUFMANN PUBLISHERS, INC. p268-277.
[124] Chang C-C, Lin C-J. LIBSVM: a library for support vector machines. ACMTransactions on Intelligent Systems and Technology (TIST).2011,2(3),27.
[125] Wang BX, Japkowicz N. Boosting support vector machines for imbalanced datasets. Knowledge and Information Systems.2010,25(1),1-20.
[126] Davis RI, Lovell BC, Caelli T. Improved estimation of hidden markov modelparameters from multiple observation sequences.2002. IEEE. p168-171.
[127] Cogger L, Murphree J, Ismail S, et al. Characteristics of dayside5577and3914aurora. Geophysical research letters.1977,4(10),413-416.
[128] Sandholt P, Farrugia C. Does the aurora provide evidence for the occurrence ofantiparallel magnetopause reconnection? Journal of Geophysical Research:Space Physics (1978–2012).2003,108(A12).
[129]李柳元,曹晋滨,周国成.磁层相对论电子通量变化与磁暴/亚暴的关系.地球物理学报.2006,49(1),9-15.
[130] Li L, Cao J, Zhou G, et al. Statistical roles of storms and substorms in changingthe entire outer zone relativistic electron population. Journal of GeophysicalResearch: Space Physics (1978–2012).2009,114(A12).
[131] Sutcliffe P. Substorm onset identification using neural networks and Pi2pulsations.1997. Springer. p1257-1264.
[132] Murphy KR, Rae IJ, Mann IR, et al. Wavelet‐based ULF wave diagnosis ofsubstorm expansion phase onset. Journal of Geophysical Research: SpacePhysics (1978–2012).2009,114(A1).
[133] Tokunaga T, Yumoto K, Uozumi T. Identification of full-substorm onset fromground-magnetometer data by singular value transformation.2011.
[134] Kataoka R, Miyoshi Y, Morioka A. Hilbert‐Huang Transform of geomagneticpulsations at auroral expansion onset. Journal of Geophysical Research: SpacePhysics (1978–2012).2009,114(A9).
[135] Meng C-I, Liou K. Substorm timings and timescales: A new aspect. Spacescience reviews.2004,113(1-2),41-75.
[136] Liou K, Meng C-I, Lui T, et al. On relative timing in substorm onset signatures.Journal of geophysical research.1999,104(A10),22807-22822,22817.
[137] Liu JM, Zhang BC, Kamide Y, et al. Spontaneous and trigger‐associatedsubstorms compared: Electrodynamic parameters in the polar ionosphere.Journal of Geophysical Research: Space Physics (1978–2012).2011,116(A1).
[138] Frey H, Mende S. Substorm onsets as observed by IMAGE-FUV.2006. p71-76.
[139] Kullen A, Karlsson T. On the relation between solar wind, pseudobreakups, andsubstorms. Journal of Geophysical Research: Space Physics (1978–2012).2004,109(A12).
[140] McPherron RL. Substorm related changes in the geomagnetic tail: The growthphase. Planetary and Space Science.1972,20(9),1521-1539.
[141] Deng Y, Kenney C, Moore MS, et al. Peer group filtering and perceptual colorimage quantization.1999. IEEE. p21-24.
[142] Bezdek JC. Pattern recognition with fuzzy objective function algorithms:Kluwer Academic Publishers;1981.
[143] Bezdek JC, Ehrlich R, Full W. FCM: The fuzzy c-means clusteringalgorithm. Computers&Geosciences.1984,10(2),191-203.
[144] Chuang K-S, Tzeng H-L, Chen S, et al. Fuzzy c-means clustering with spatialinformation for image segmentation. computerized medical imaging andgraphics.2006,30(1),9-15.
[145] Sun H, Wang S, Jiang Q. FCM-based model selection algorithms fordetermining the number of clusters. Pattern Recognition.2004,37(10),2027-2037.
[146] Wang Y, Li C, Zuo Y. A selection model for optimal fuzzy clustering algorithmand number of clusters based on competitive comprehensive fuzzy evaluation.Fuzzy Systems, IEEE Transactions on.2009,17(3),568-577.
[147] Kisabeth JL, Rostoker G. The expansive phase of magnetospheric substorms:1.Development of the auroral electrojets and auroral arc configuration during asubstorm. Journal of geophysical research.1974,79(7),972-984.
[148] Liou K, Meng CI, Wu CC. On the interplanetary magnetic field By control ofsubstorm bulge expansion. Journal of Geophysical Research: Space Physics(1978–2012).2006,111(A9).
[149] Voronkov I, Donovan E, Samson J. Observations of the phases of the substorm.Journal of geophysical research.2003,108(A2),1073.
[150] Newell P, Gjerloev J. Evaluation of SuperMAG auroral electrojet indices asindicators of substorms and auroral power. Journal of Geophysical Research:Space Physics (1978–2012).2011,116(A12).
[151] Newell P, Gjerloev J. Substorm and magnetosphere characteristic scalesinferred from the SuperMAG auroral electrojet indices. Journal of GeophysicalResearch: Space Physics (1978–2012).2011,116(A12).
[152] Ohtani S, Anderson B, Sibeck D, et al. A multisatellite study of a pseudo‐substorm onset in the near‐Earth magnetotail. Journal of GeophysicalResearch: Space Physics (1978–2012).1993,98(A11),19355-19367.
[153] Rostoker G. On the place of the pseudo‐breakup in a magnetosphericsubstorm. Geophysical research letters.1998,25(2),217-220.
[154] Nakamura R, Baker DN, Yamamoto T, et al. Particle and field signatures duringpseudobreakup and major expansion onset. Journal of geophysical research.1994,99(A1),207-221.
[155] Aikio A, Sergeev V, Shukhtina M, et al. Characteristics of pseudobreakups andsubstorms observed in the ionosphere, at the geosynchronous orbit, and in themidtail. Journal of Geophysical Research: Space Physics (1978–2012).1999,104(A6),12263-12287.
[156] Kullen A, Karlsson T, Cumnock JA, et al. Occurrence and properties ofsubstorms associated with pseudobreakups. Journal of Geophysical Research:Space Physics (1978–2012).2010,115(A12).
[157] Boakes PD. Investigating the relationship between open magnetic flux and thesubstorm cycle;2010.
[158] Carbary JF. A Kp‐based model of auroral boundaries. Space Weather.2005,3(10).
[159] Zhang Y, Paxton L. An empirical Kp-dependent global auroral model based onTIMED/GUVI FUV data. Journal of Atmospheric and Solar-Terrestrial Physics.2008,70(8),1231-1242.
[160] Holzworth R, Meng CI. Mathematical representation of the auroral oval.Geophysical research letters.1975,2(9),377-380.
[161] Holzworth R, Meng C-I. Auroral boundary variations and the interplanetarymagnetic field. Planetary and Space Science.1984,32(1),25-29.
[162] Kauristie K. Statistical fits for auroral oval boundaries during the substormsequence. Journal of geophysical research.1995,100(A11),21885-21821,21895.
[163] Sigernes F, Dyrland M, Brekke P, et al. Real time aurora ovalforecasting-SvalTrackII. In other words.2011,3(3),3.
[164] Sigernes F, Dyrland M, Brekke P, et al. Two methods to forecast auroraldisplays. Journal of Space Weather and Space Climate.2011,1(1).
[165] Starkov G. Mathematical model of the auroral boundaries. GEOMAGNETISMAND AERONOMY C/C OF GEOMAGNETIZM I AERONOMIIA.1994,34,331-336.
[166] Starkov GV. Analiticheskoe predstavlenie ekvatorial’noy granitsy oval’noyzony polyarnikh siyaniy (Analytical representation of the equatorwardboundary of the polar auroral oval). Geomagn Aeron.1969,9,759.
[167] Hardy D, Burke W, Gussenhoven M, et al. DMSP/F2electron observations ofequatorward auroral boundaries and their relationship to the solar wind velocityand the north‐south component of the interplanetary magnetic field. Journalof Geophysical Research: Space Physics (1978–2012).1981,86(A12),9961-9974.
[168] Cowley S, Morelli J, Lockwood M. Dependence of convective flows andparticle precipitation in the high‐latitude dayside ionosphere on the x and ycomponents of the interplanetary magnetic field. Journal of GeophysicalResearch: Space Physics (1978–2012).1991,96(A4),5557-5564.
[169] Liou K, Newell P, Sibeck D, et al. Observation of IMF and seasonal effects inthe location of auroral substorm onset. Journal of Geophysical Research: SpacePhysics (1978–2012).2001,106(A4),5799-5810.
[170] Meng C-I. Polar cap variations and the interplanetary magnetic field. Dynamicsof the Magnetosphere: Springer;1979. p23-46.
[171] Yang Y, Lu J, Wang JS, et al. Influence of interplanetary magnetic field andsolar wind on auroral brightness in different regions. Journal of GeophysicalResearch: Space Physics.2013.
[172] Lee DY, Lyons L, Yumoto K. Sawtooth oscillations directly driven by solarwind dynamic pressure enhancements. Journal of Geophysical Research: SpacePhysics (1978–2012).2004,109(A4).
[173] Lyons L. Geomagnetic disturbances: Characteristics of, distinction betweentypes, and relations to interplanetary conditions. Journal of Atmospheric andSolar-Terrestrial Physics.2000,62(12),1087-1114.
[174] Shue J-H, Newell P, Liou K, et al. Solar wind density and velocity control ofauroral brightness under normal interplanetary magnetic field conditions.Journal of geophysical research.2002,107(A12),1428.
[175] Cho J-S, Lee D-Y, Kim K-C, et al. Response of the Poleward Boundary of theNightside Auroral Oval to Impacts of Solar Wind Dynamic PressureEnhancement. Journal of Astronomy and Space Sciences.2010,27,189-194.
[176] Brittnacher M, Fillingim M, Parks G, et al. Polar cap area and boundary motionduring substorms. Journal of Geophysical Research: Space Physics(1978–2012).1999,104(A6),12251-12262.
[177] Liou K, Newell P, Meng CI, et al. Characteristics of the solar wind controlledauroral emissions. Journal of Geophysical Research: Space Physics(1978–2012).1998,103(A8),17543-17557.
[178] Troshichev O, Andrezen V, Vennerstr m S, et al. Magnetic activity in the polarcap—A new index. Planetary and Space Science.1988,36(11),1095-1102.
[179] Kan J, Lee L. Energy coupling function and solar wind‐magnetospheredynamo. Geophysical research letters.1979,6(7),577-580.
[180] Hu ZJ, Yang HG, Han DS, et al. Dayside auroral emissions controlled by IMF:A survey for dayside auroral excitation at557.7and630.0nm in Ny‐lesund,Svalbard. Journal of Geophysical Research: Space Physics (1978–2012).2012,117(A2).
[181] Sykes A. An introduction to regression analysis: Law School, University ofChicago;1993.