日侧全天空极光图像分类及动态过程分析方法研究
|
摘要
极光是沿磁力线运动的高能带电粒子（主要是电子）沉降到极区电离层高度激发大气粒子后产生的光激发现象。这些高能粒子既有通过太阳风—磁层相互作用的磁动力学过程产生的，也有部分是直接由太阳风携带的。极光是人们唯一能够直接用肉眼观测到的具有极区特征的地球物理现象，是极区日地物理过程（特别是磁层-电离层相互作用）的最集中的表现形式，是研究太阳风暴的最佳窗口。对其形态和演变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。
多变的极光形态是磁层和电离层某种确定的动力学过程对大气层的反应，是这个过程投影到地球大气层的映射。研究已经证实，不同的极光形态类型与一些确定的磁层边界层动力学过程有着密切的关系，而且太阳风的参数变化直接影响极光的形态。高分辨率的极光成像设备（例如全天空观测仪）能够获取极光的二维形态信息，从而使得研究人员可以连续观测极光的演变。因此，从图像的角度研究极光现象越来越受到人们的关注。然而，现在的研究大多还停留在人工的案例分析，以至每年都在不断增长的庞大的极光图像成为闲置资源。面对如此庞大的数据量，如何有效利用和自动分析这些全天空极光图像已成为各国极光研究人员亟待解决的问题。
本文基于图像处理和模式识别技术，提出了多种适合极光数据特征的静态和动态表征方法，初步实现了对海量极光数据的自动化分析，研究成果为很多现有的理论提供了形态学证明，验证了模式识别技术应用于空间物理研究的可行性和有效性。
本文的主要研究成果有：
(1)对极光图像特征进行了系统分析和多种尝试，提出使用基于空间纹理的全天空极光图像表征方法。该方法能够表征极光图像的全局形状和局部纹理，不同的极光类型都得到了有效的表征，并且对一定范围内的尺度和亮度变化以及形变具有很好的鲁棒性。基于这一表征方法，实现了高正确率的极光图像自动分类。
(2)基于表象特征的极光自动分类技术用于对日侧极光类型的统计分析。无论极光类型是定义在极光的形态学信息还是其他物理属性上的，这些极光类型在极光卵上的发生总是呈现出周期性和规律性，这已经在许多空间科学研究中得到证实。本文得到了四类日侧极光的发生分布，分布的特征说明发生在日侧极光卵不同区域的极光确实具有各自明显的形态学特征。四种特殊的极光类型占据不同的极光卵区域，它们发生分布的峰值与基于三波段(427.8,557.7和630.0nm)沿磁子午面激发谱线能量的综观分布图是完全一致的。这一结论支持了“极光形态可分”这一观点，为极光分类机制提供了形态学证据，同时给出了极光类型的形态学解释。
(3)对于极光数据的形态特征而言，尚无被广泛接受的分类机制。本文利用了无监督的聚类算法对极光数据进行了聚类，从而避免了有监督分类中被诟病的分类机制问题。在本文中，不基于任何先验知识，通过无监督的方法，日侧极光的主要形态特征得到总结，同时对现有的基于极光形态特征的分类机制进行了验证和补充。聚类标记的发生分布规律性证明了依赖于极光表象特征研究极光现象的可行性。这些成果进而验证了机器视觉方法用于研究极光现象的有效性。
(4)因为极光运动具有非刚体性质，本文首次把针对流体运动的流体运动场估计算法应用到极光图像中，并根据极光图像在不同尺度上具有不同运动方式的特性，提出了非一致多尺度流体运动场估计算法。这一方法不再依赖于亮度不变假设，而且为不同尺度上的极光运动选择匹配的正则化因子。改进后的基于流体的运动场估计算法能够有效、准确、鲁棒地提取各种类型极光的运动。
(5)基于提取出的极光瞬时运动场表征极光序列，实现极光事件的自动检测。由于极光活动复杂多变，随机性强，本文利用时空累计方法对极光视频序列进行运动特征描述。基于运动向量的时空统计特征能够表征极光的运动方向和运动速度，可以实现不同长度极光序列的比较。在此基础上，实现了对多个特殊极光事件的准确检索，为实现基于大量连续观测的极光视频对极光动态过程进行统计分析奠定了基础。
(6)提出了一种新的基于局部向量差的时空统计特征，它能够融合运动的时间性和空间性，可以实现不同长度序列的比较。该方法可以有效、可靠地描述极光视频序列的运动特征，成功地实现了极光序列的自动分割和事件聚类。利用本文提出的极光事件自动分析方法，在无监督的条件下，即可实现对大量的全天空极光活动进行高速地浏览、检查和排除。本文受国家自然科学基金（60872154,41031064）、、海洋公益性行业科研专项（201005017）等项目的经费资助。
Aurora is created by atmospheric atoms and molecules colliding with electronsand protons from outer space when they precipitate into the atmosphere. Theseenergetic particles are composed of those particles discharged in magneto dynamicsprocess of the solar wind-magnetosphere interaction, and some particles from the solarwind. Among geophysical phenomena processing features of polar region, aurora is theonly one can be seen by naked eye. The aurora phenomena provide a convenientprojection of effects from complex and energetic plasma processes of the outermagnetosphere. Much has been learned about the ionosphere and magnetosphere fromauroral events. The spatial structure and temporal evolution of auroral luminosity areascribed to cumulative effects of the solar wind–magnetosphere interaction and thephysics of the magnetosphere–ionosphere interaction. Therefore, the study on auroralappearance and evolvement is helpful to study influence of the sun to earth, and issignificant to acquire information about space weather. Thus this study has greatsignificance for science research and practical application.
Certain physical processes in the magnetosphere and ionosphere are responsiblefor the auroral appearance. In earlier studies, there have been several types of auroraidentified, which have turned out to be correlated with specific magnetosphericregimes and dynamic activities. Variations in the solar wind parameters seem to havea strong influence on auroral appearance. The valuable two-dimensionalmorphological information can be acquired by high spatial and temporal resolutions,which enables scientists to successively observe auroral behavior. The auroral studybased on image is attracting more attentions. However, most of current studies are―case study‖which is performed manually, so that annually increasing huge numberof auroral images haven’t been utilized. Provided with so abundant valuable data,tools which automatically analyze aurora are urgently needed, which has beensubjects of many research efforts.
The study of applying image processing and pattern recognition techniques inautomatic auroral image analysis is far from complete. Several problems need furtherinvestigation. In this thesis, several good techniques in image processing and patternrecognition are introduced and improved in studying aurora phenomena. Manyrepresentation methods properly characterizing auroral statistic and dynamicinformation are proposed. Automatic analysis on huge number of auroral images is achieved. The experimental results provide morphological proofs for classificationschemes available, and also offer morphological interpretation of auroral types. Thus,the feasibility and effectiveness of techniques of image processing and patternrecognition in studying space science are verified.
In sum, the author’s major contributions are outlined as follows:
1. By systematically analyzing dayside aurora characters in all-sky imager (ASI)observations and having attempted various methods, a spatial texture basedrepresentation method including features of intensity, shape and texture, was utilizedto characterize ASI images. The combination of the local binary pattern (LBP)operator and a delicately designed block partition scheme achieved capabilities ofrepresenting both global shapes and local textures. The representation method wasused in automatic recognition of four primary categories of discrete dayside aurorausing observations between years2003–2009at the Yellow River Station,Ny-lesund, Svalbard. The supervised classification results on labeled data in2003were in accordance with the results labeled by scientists considering both spectral andmorphological information.
2. The occurrence distributions of the four categories were obtained byautomatic classification technologies. Regardless of being discriminated by somephysical properties or morphology, certain auroral types may occur periodically andrepeatedly in the auroral oval, which has been verified in numerous space sciencestudies. The obtained occurrence distributions of four dayside auroral categoriesbased on morphology confirm that there are dominant morphological characteristicsin different regions of the dayside oval. The peak positions of occurrence of the fourauroral types are in accordance with synoptic distribution based on the intensities ofphoto-emissions on three wavelengths (427.8,557.7and630.0nm) along themagnetic meridian during daytime (0300–1500UT/0600–1800MLT). Theexperimental results support that auroral morphology identifiable. The results providemorphological evidence for available classification schemes and offers morphologicalinterpretation of auroral types.
3. By now, there is no generally accepted set of auroral types for auroralclassification studies. It's uncertain that the auroral classification schemes availablereflect the real natural structure of the auroral data. The unsupervised clusteringmethod is utilized in classifying ASI images, which don’t depend on any predefinedclassification. In unsupervised way, dominant morphological characteristicsthroughout dayside oval are found, which thereby test and complement the available classification schemes. Being independent of any prior knowledge or temporalinformation, auroral data exhibit clustering tendency. The regularity in occurrencedistributions of clustering labels verifies the feasible of studying auroral dynamicprocesses depending on morphology. These achievements properly verifyeffectiveness of pattern recognition method in studying aurora phenomena.
4. Considering the fluid nature of auroral motions and non-uniformity of motionson different scales, we introduce a fluid flow algorithm and modified theregularization methods to estimate motion fields under dynamic auroral situations.The modified motion estimation method doesn’t depend on brightness constancyassumption and use the matching regularizer for auroral motions on different scale.The modified method can effectively, accurately and robustly capture the auroralappearance structures and estimate auroral motion.
5. Based on extracted motion fields, the auroral event can be characterized andautomatic event detection is achieved. Because auroral activities are extremelycomplex, changeful and predictable, we use spatiotemporally statistics of motionvectors to represent auroral motions. By using the representation, two sequences withdifferent lengths can be compared. Automatic multi-event detection is achieved,which support statistical analyzing auroral dynamic process based on numeroussuccessive observation.
6. A novel method of modeling dynamic texture (DT) is proposed to characterizeauroral motions. The method synergizes spatial and temporal aspects of DTs, anddoesn’t limit the length of sequence. Its validity is firstly verified in a common DTdataset. The new method can effectively and robustly represent auroral motions,which support segmentation and clustering of auroral events. Thus, by using theproposed method of automatically analyzing auroral event, great amount of auroralactivities can be quickly browsed, inspected and eliminated.
多变的极光形态是磁层和电离层某种确定的动力学过程对大气层的反应，是这个过程投影到地球大气层的映射。研究已经证实，不同的极光形态类型与一些确定的磁层边界层动力学过程有着密切的关系，而且太阳风的参数变化直接影响极光的形态。高分辨率的极光成像设备（例如全天空观测仪）能够获取极光的二维形态信息，从而使得研究人员可以连续观测极光的演变。因此，从图像的角度研究极光现象越来越受到人们的关注。然而，现在的研究大多还停留在人工的案例分析，以至每年都在不断增长的庞大的极光图像成为闲置资源。面对如此庞大的数据量，如何有效利用和自动分析这些全天空极光图像已成为各国极光研究人员亟待解决的问题。
本文基于图像处理和模式识别技术，提出了多种适合极光数据特征的静态和动态表征方法，初步实现了对海量极光数据的自动化分析，研究成果为很多现有的理论提供了形态学证明，验证了模式识别技术应用于空间物理研究的可行性和有效性。
本文的主要研究成果有：
(1)对极光图像特征进行了系统分析和多种尝试，提出使用基于空间纹理的全天空极光图像表征方法。该方法能够表征极光图像的全局形状和局部纹理，不同的极光类型都得到了有效的表征，并且对一定范围内的尺度和亮度变化以及形变具有很好的鲁棒性。基于这一表征方法，实现了高正确率的极光图像自动分类。
(2)基于表象特征的极光自动分类技术用于对日侧极光类型的统计分析。无论极光类型是定义在极光的形态学信息还是其他物理属性上的，这些极光类型在极光卵上的发生总是呈现出周期性和规律性，这已经在许多空间科学研究中得到证实。本文得到了四类日侧极光的发生分布，分布的特征说明发生在日侧极光卵不同区域的极光确实具有各自明显的形态学特征。四种特殊的极光类型占据不同的极光卵区域，它们发生分布的峰值与基于三波段(427.8,557.7和630.0nm)沿磁子午面激发谱线能量的综观分布图是完全一致的。这一结论支持了“极光形态可分”这一观点，为极光分类机制提供了形态学证据，同时给出了极光类型的形态学解释。
(3)对于极光数据的形态特征而言，尚无被广泛接受的分类机制。本文利用了无监督的聚类算法对极光数据进行了聚类，从而避免了有监督分类中被诟病的分类机制问题。在本文中，不基于任何先验知识，通过无监督的方法，日侧极光的主要形态特征得到总结，同时对现有的基于极光形态特征的分类机制进行了验证和补充。聚类标记的发生分布规律性证明了依赖于极光表象特征研究极光现象的可行性。这些成果进而验证了机器视觉方法用于研究极光现象的有效性。
(4)因为极光运动具有非刚体性质，本文首次把针对流体运动的流体运动场估计算法应用到极光图像中，并根据极光图像在不同尺度上具有不同运动方式的特性，提出了非一致多尺度流体运动场估计算法。这一方法不再依赖于亮度不变假设，而且为不同尺度上的极光运动选择匹配的正则化因子。改进后的基于流体的运动场估计算法能够有效、准确、鲁棒地提取各种类型极光的运动。
(5)基于提取出的极光瞬时运动场表征极光序列，实现极光事件的自动检测。由于极光活动复杂多变，随机性强，本文利用时空累计方法对极光视频序列进行运动特征描述。基于运动向量的时空统计特征能够表征极光的运动方向和运动速度，可以实现不同长度极光序列的比较。在此基础上，实现了对多个特殊极光事件的准确检索，为实现基于大量连续观测的极光视频对极光动态过程进行统计分析奠定了基础。
(6)提出了一种新的基于局部向量差的时空统计特征，它能够融合运动的时间性和空间性，可以实现不同长度序列的比较。该方法可以有效、可靠地描述极光视频序列的运动特征，成功地实现了极光序列的自动分割和事件聚类。利用本文提出的极光事件自动分析方法，在无监督的条件下，即可实现对大量的全天空极光活动进行高速地浏览、检查和排除。本文受国家自然科学基金（60872154,41031064）、、海洋公益性行业科研专项（201005017）等项目的经费资助。
Aurora is created by atmospheric atoms and molecules colliding with electronsand protons from outer space when they precipitate into the atmosphere. Theseenergetic particles are composed of those particles discharged in magneto dynamicsprocess of the solar wind-magnetosphere interaction, and some particles from the solarwind. Among geophysical phenomena processing features of polar region, aurora is theonly one can be seen by naked eye. The aurora phenomena provide a convenientprojection of effects from complex and energetic plasma processes of the outermagnetosphere. Much has been learned about the ionosphere and magnetosphere fromauroral events. The spatial structure and temporal evolution of auroral luminosity areascribed to cumulative effects of the solar wind–magnetosphere interaction and thephysics of the magnetosphere–ionosphere interaction. Therefore, the study on auroralappearance and evolvement is helpful to study influence of the sun to earth, and issignificant to acquire information about space weather. Thus this study has greatsignificance for science research and practical application.
Certain physical processes in the magnetosphere and ionosphere are responsiblefor the auroral appearance. In earlier studies, there have been several types of auroraidentified, which have turned out to be correlated with specific magnetosphericregimes and dynamic activities. Variations in the solar wind parameters seem to havea strong influence on auroral appearance. The valuable two-dimensionalmorphological information can be acquired by high spatial and temporal resolutions,which enables scientists to successively observe auroral behavior. The auroral studybased on image is attracting more attentions. However, most of current studies are―case study‖which is performed manually, so that annually increasing huge numberof auroral images haven’t been utilized. Provided with so abundant valuable data,tools which automatically analyze aurora are urgently needed, which has beensubjects of many research efforts.
The study of applying image processing and pattern recognition techniques inautomatic auroral image analysis is far from complete. Several problems need furtherinvestigation. In this thesis, several good techniques in image processing and patternrecognition are introduced and improved in studying aurora phenomena. Manyrepresentation methods properly characterizing auroral statistic and dynamicinformation are proposed. Automatic analysis on huge number of auroral images is achieved. The experimental results provide morphological proofs for classificationschemes available, and also offer morphological interpretation of auroral types. Thus,the feasibility and effectiveness of techniques of image processing and patternrecognition in studying space science are verified.
In sum, the author’s major contributions are outlined as follows:
1. By systematically analyzing dayside aurora characters in all-sky imager (ASI)observations and having attempted various methods, a spatial texture basedrepresentation method including features of intensity, shape and texture, was utilizedto characterize ASI images. The combination of the local binary pattern (LBP)operator and a delicately designed block partition scheme achieved capabilities ofrepresenting both global shapes and local textures. The representation method wasused in automatic recognition of four primary categories of discrete dayside aurorausing observations between years2003–2009at the Yellow River Station,Ny-lesund, Svalbard. The supervised classification results on labeled data in2003were in accordance with the results labeled by scientists considering both spectral andmorphological information.
2. The occurrence distributions of the four categories were obtained byautomatic classification technologies. Regardless of being discriminated by somephysical properties or morphology, certain auroral types may occur periodically andrepeatedly in the auroral oval, which has been verified in numerous space sciencestudies. The obtained occurrence distributions of four dayside auroral categoriesbased on morphology confirm that there are dominant morphological characteristicsin different regions of the dayside oval. The peak positions of occurrence of the fourauroral types are in accordance with synoptic distribution based on the intensities ofphoto-emissions on three wavelengths (427.8,557.7and630.0nm) along themagnetic meridian during daytime (0300–1500UT/0600–1800MLT). Theexperimental results support that auroral morphology identifiable. The results providemorphological evidence for available classification schemes and offers morphologicalinterpretation of auroral types.
3. By now, there is no generally accepted set of auroral types for auroralclassification studies. It's uncertain that the auroral classification schemes availablereflect the real natural structure of the auroral data. The unsupervised clusteringmethod is utilized in classifying ASI images, which don’t depend on any predefinedclassification. In unsupervised way, dominant morphological characteristicsthroughout dayside oval are found, which thereby test and complement the available classification schemes. Being independent of any prior knowledge or temporalinformation, auroral data exhibit clustering tendency. The regularity in occurrencedistributions of clustering labels verifies the feasible of studying auroral dynamicprocesses depending on morphology. These achievements properly verifyeffectiveness of pattern recognition method in studying aurora phenomena.
4. Considering the fluid nature of auroral motions and non-uniformity of motionson different scales, we introduce a fluid flow algorithm and modified theregularization methods to estimate motion fields under dynamic auroral situations.The modified motion estimation method doesn’t depend on brightness constancyassumption and use the matching regularizer for auroral motions on different scale.The modified method can effectively, accurately and robustly capture the auroralappearance structures and estimate auroral motion.
5. Based on extracted motion fields, the auroral event can be characterized andautomatic event detection is achieved. Because auroral activities are extremelycomplex, changeful and predictable, we use spatiotemporally statistics of motionvectors to represent auroral motions. By using the representation, two sequences withdifferent lengths can be compared. Automatic multi-event detection is achieved,which support statistical analyzing auroral dynamic process based on numeroussuccessive observation.
6. A novel method of modeling dynamic texture (DT) is proposed to characterizeauroral motions. The method synergizes spatial and temporal aspects of DTs, anddoesn’t limit the length of sequence. Its validity is firstly verified in a common DTdataset. The new method can effectively and robustly represent auroral motions,which support segmentation and clustering of auroral events. Thus, by using theproposed method of automatically analyzing auroral event, great amount of auroralactivities can be quickly browsed, inspected and eliminated.
引文
[1] Real Time Geophysical Data Links. http://odin.gi.alaska.edu/lumm/RealTime/.
[2] Hu, Z.-J., Yang, H., Huang, D., Araki, T., Sato, N., et al. Synoptic distribution ofdayside aurora: multiple-wavelength all-sky observation at Yellow River Stationin Ny-lesund, Svalbard. Journal of Atmospheric and Solar-Terrestrial Physics,2009,71:794–804.
[3] Feldstein, Y., Elphinstone, R. Aurorae and the large-scale structure of themagnetosphere. Journal of Geomagnetism and Geoelectricity,1992,44:1159–1174.
[4] Kullen, A., Brittnacher, M., Cumnock, J.A., Blomberg, L.G. Solar winddependence of the occurrence and motion of polar auroral arcs: a statistical study.Journal of Geophysical Research,2002,107(A11):1362.
[5] THEMIS project. http://aurora.phys.ucalgary.ca/themis/themis_main.html.
[6] St rmer, C. The Polar Aurora. Clarendon Press,1955, Oxford.
[7] Vallance, J.A. Aurora.1974, D. Riedel, Dordrecht, Holland.
[8] Steen,, Br ndstr m, U., Gustavsson, B. ALIS—a multi station imaging systemat high latitude with multi disciplinary scientific objectives. In: Proceedings ofthe13th ESA Symposium on European Rocket and Balloon Programmes andRelated Research,1997:261-266.
[9] Marklund, G.T. Auroral arc classification scheme based on the observedarc-associated electric field pattern. Planetary and Space Science,1984,32(2):193-211.
[10] Ayukawa, M., Makita, K., Yamagishi, H., Ejiri, M., Sakanoi, T. Characteristics ofpolar cap aurora. Journal of Atmospheric and Terrestrial Physics,1996,58:1885-1894.
[11] Yang, H., Sato, N., Makita, K., Kikuchi, M., Kadokura, A., Ayukawa, M., Hu,H.Q., Liu, R.Y., H ggstr m, I. Synoptic observations of auroras along thepost-noon oval: a survey with all-sky TV observations at Zhongshan, Antarctica.Journal of Atmospheric and Solar-Terrestrial Physics,2000,62:787-797.
[12] Syrj suo, M.T., Donovan, E.F., Qin, X., Yang, Y.-H. Automatic classification ofauroral images in substorm studies. In: Proceedings of the Eight InternationalConference on Substorms (ICS8), Canada,2006:309-313.
[13] Donovan, E. F., Trondsen, T. S., Cogger, L. L. and Jackel, B. J., Auroral imagingin Canadian CANOPUS and NORSTAR programs, Proc. of Atmospheric Studiesby Optical Methods,2003:109–112.
[14] Syrj suo, M.T.,2001. Auroral monitoring network: from all-sky camera systemto automated image analysis. Ph.D Thesis, Finnish Meteorological Institute,Helsinki, Finland, ISBN951-697-551-8.
[15] Syrj suo, M.T., Donovan, E.F. Content-based retrieval of auroralimages—thousands of irregular shapes. In: Proceedings of the IASTEDInternational Conference on Visualization, Imaging, and Image Processing,2004:224-228.
[16] Syrj suo, M.T., Donovan, E.F. Diurnal auroral occurrence statistics obtained viamachine vision. Annales Geophysicae,2004,22(4):1103–1113.
[17] Syrj suo, M.T., Donovan, E.F. Using relevance feedback in retrieving auroralimages. In: Proceedings of the IASTED Computational Intelligence,2005:420-425.
[18] Syrj suo, M.T., Kauristie, K., Pulkkinen, T.I. A search engine for auroral forms.Advances in Space Research,2001:28(11),1611-1616.
[19] G. V. G. Baranoski, J. G. Rokne, P. Shirley, T. S. Trondsen, and R. Bastos.Simulating the aurora. The Journal of Visualization and Computer Animation,2003,14:43-59.
[20] G. V. G. Baranoski, J. Wan, J. G. Rokne, and I. Bell. Simulating the dynamics ofauroral phenomena. ACM Trans. Graph,2005,24:37-59.
[21] Kauristie, K., Syrj suo, M.T., Amm, O., Viljanen, A., Pulkkinen, T.I.,Opgenoorth, H.J. A Statistical study of evening sector arcs and electrojets.Advances in Space Research,2001,28(11):1605–1610.
[22] Qin, X., Yang, Y.-H. Basic Gray Level Aura Matrices: Theory and its Applicationto Texture Synthesis. IEEE ICCV, Beijing, China,2005:128–135.
[23] Blixt, E., Semeter, J., Ivchenko, N. Optical flow analysis of the aurora borealis.Geoscience and Remote Sensing Letters, IEEE,2006,3:159-163.
[24] Horn, B., Schunck, B. Determining optical flow. Artificial intelligence,1981,17:185-203.
[25]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——中国北极黄河站首次极光观测初步分析.极地研究,2005,17:107-114.
[26] Ojala, T., Pietik inen, M., Harwood, D. A comparative study of texture measureswith classification based on feature distributions. Pattern Recognition,1996,29:51–59.
[27] Ojala, T., Pietik inen, M., M enp, T. Multiresolution grayscale and rotationinvariant texture classification with local binary patterns. IEEE Transactions onPattern Analysis and Machine Intelligence,2002,24(7):971–987.
[28] Ahonen, T., Hadid, A., Pietika¨inen, M. Face description with local binarypatterns: application to face recognition. IEEE Transactions on Pattern Analysisand Machine Intelligence,2006,28(12):2037–2041.
[29] Theodoridis, S., Koutroumbas, K. Pattern Recognition, third ed. Elsevier,2006.
[30] Duda, R.O., Hart, P.E., Stork, D.G. Pattern Classification Second Edition JohnWiley&Sons,2001.
[31] NSTC Subcommittee on Biometrics. Biometrics testing and statistics. Availableonline: http://www.biometrics.gov/Documents/BioTestingAndStats.pdf.
[32] Akasofu, S.I., Kan, J.R. Dayside and nightside auroral arc systems. GeophysicalResearch Letters,1980,7(10):753–756.
[33] Liou, K., Newell, P.T., Meng, C.-I., Brittnacher, M. Parks, G. Synoptic auroraldistribution: a survey using polar ultraviolet imagery. Journal of GeophysicalResearch,1997,102:27197–27205.
[34] Newell, P.T., Lyons, K.M., Meng, C.-I. A large survey of electron accelerationevents. Journal of Geophysical Research,1996,101:2599–2614.
[35] Wang Q, Liang J, Hu Z J, Hu H H, Zhao H, Hu H Q, Gao X B, Yang H G. SpatialTexture Based Automatic Classification of Dayside Aurora in All-Sky Images.Journal of Atmospheric and Solar-Terrestrial Physics,2010,72(5-6):498-508.
[36] Xu, R., Wunsch, D. Survey of clustering algorithms. IEEE Transactions onneural network,2005,16:645-678.
[37] Shi, J., Malik, J. Normalized cuts and image segmentation. IEEE Transactions onPattern Analysis and Machine Intelligence,2002,22(8):888-905.
[38] Jolliffe, I. Principal component analysis,2002, Springer Verlag.
[39] Jain, A., Dubes, R. Algorithms for clustering data. Prentice-Hall AdvancedReference Series,1988,320.
[40] Jain, A., Murty, M., Flynn, P. Data Clustering: A Review. ACM computingSurveys,1999,31:264-323.
[41] Chris, H., Ding, Q., Li, T. Adaptive dimension reduction using discriminantanalysis and k-means clustering. In International Conference on MachineLearning,2007:521-528.
[42] Halkidi, M., Batistakis, Y., Vazirgiannis, M. On Clustering Validation Techniques.Intelligent Information Systems,2001,17(2-3):107-145.
[43] Halkidi, M., Batistakis, Y., Vazirgiannis, M. Cluster validity methods: part I.ACM SIGMOD Record,2002,31, Issue2:40-45.
[44] Halkidi, M., Batistakis, Y., Vazirgiannis, M. Cluster validity methods: part II.ACM SIGMOD Record,2002,31, Issue3:19-27.
[45] Yang, Y., Xu, D., Nie, F., Yan, S., Zhuang, Y. Image clustering using localdiscriminant models and global integration. IEEE Transactions on ImageProcessing,2010,19:2761-2773.
[46] Chen, W., Song, Y., Bai, H., Lin, C., Chang, E. Parallel spectral clustering indistributed systems. IEEE Transactions on Pattern Analysis and MachineIntelligence,2011,33(3):568-586.
[47] Von Luxburg, U. A tutorial on spectral clustering. Statistics and Computing,2007,17(4):395-416.
[48] Abascal, F. and Valencia, A. Clustering of proximal sequence space for theidentification of protein families, Bioinformatics,2002,18(7):908.
[49] Chen, C. Handbook of pattern recognition and computer vision. World Scientific,2010.
[50] Legány, C. and Juhász, S. and Babos, A. Cluster validity measurementtechniques. Proceedings of the5th WSEAS International Conference onArtificial Intelligence, Knowledge Engineering and Data Bases,2006:388-393.
[51] Everitt, B. and Landau, S. and M. Leese. Cluster Analysis. New York: OxfordUniv. Press,2001.
[52] Davies, DL and Bouldin, DW. A cluster separation measure. IEEE Transactionson Pattern Analysis and Machine Intelligence,1979,1:224-227.
[53] Dunn, J.C. Well separated clusters and optimal fuzzy partitions. Journal ofCybernetica,1974,4:95-104.
[54] Halkidi, M. and Vazirgiannis, M. Clustering validity assessment: Finding theoptimal partitioning of a data set. Proceedings IEEE International Conference onData Mining (ICDM),2002:187-194.
[55] Halkidi, M. and Vazirgiannis, M. and Batistakis, Y. Quality scheme assessment inthe clustering process. Principles of Data Mining and Knowledge Discovery,2000:265-276.
[56] Caliński, T. and Harabasz, J. A dendrite method for cluster analysis.Communications in Statistics-Theory and Methods,1974,3(1):1-27.
[57] Kaufman, L. and Rousseeuw, P.J. Finding groups in data—an introduction tocluster analysis. New York: Wiley,1990.
[58] Maulik, U. and Bandyopadhyay, S. Performance evaluation of some clusteringalgorithms and validity indices. IEEE Transactions on Pattern Analysis andMachine Intelligence,2002,24(12):1650-1654.
[59] Xie, X.L. and Beni, G. A validity measure for fuzzy clustering. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,13(8):841-847.
[60] Kim, D.J. and Park, Y.W. and Park, D.J. A novel validity index for determinationof the optimal number of clusters. IEICE transactions on information andsystems,2001,84(2):281-285.
[61] Kwon, S.H. Cluster validity index for fuzzy clusterin. Electronics Letters,2002,34(22):2176-2177.
[62] Dungey, J. W. Interplanetary Magnetic Field and the Auroral Zones. PhysicalReview Letters,1961,6(2):47-48.
[63] Lemaire, J. and Roth, M. Penetration of solar wind plasma elements into themagnetosphere. Journal of Atmospheric and Terrestrial Physics,1978,40(3):331-335.
[64] Wei, C. and Lee, L. and La Belle-Hamer, A. A simulation study of the vortexstructure in the low-latitude boundary layer. Journal of Geophysical Research,1990,95(A12):20793.
[65] Haerendel, G. and Buchert, S. and La Hoz, C. and Raaf, B. and Rieger, E. On theproper motion of auroral arcs. Journal of Geophysical Research,1993,98(A4):6087-6099.
[66] Kimball, J. and Hallinan, T. Observations of black auroral patches and of theirrelationship to other types of aurora. Journal of Geophysical Research,1998,103(A7):14671.
[67] Blixt, E.M. and Semeter, J. Relative drift between black aurora and theionospheric plasma. Ann. Geophys,2005,23(5):1611-1621.
[68] Simpson, J.J. and Gobat, J.I. Robust velocity estimates, stream functions, andsimulated Lagrangian drifters from sequential spacecraft data., IEEETransactions on Geoscience and Remote Sensing,2002,32(3):479-493.
[69] Shukla, J. and Saha, K. Computation of non-divergent streamfunction andirrotational velocity potential from the observed winds. Monthly weather review,1974,102(6):419-425.
[70] Jain, R. and Nagel, H.H. On the analysis of accumulative difference picturesfrom image sequences of real world scenes. IEEE Transactions on PatternAnalysis and Machine Intelligence,1979, PAMI-1(2):206-214.
[71] Koenderink, J.J. and Doorn, A.J. Visual perception of rigidity of solid shape.Journal of Mathematical Biology,1976,3(1):79-85.
[72] Corpetti, T., Mémin, é., Pérez, P. Dense estimation of fluid flows. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(3):365-380.
[73] Wildes, R.P. and Amabile, M.J. and Lanzillotto, A.M. and Leu, T.S. Recoveringestimates of fluid flow from image sequence data. Computer vision and Imageunderstanding,2000,80(2):246-266.
[74] Zhou, L. and Kambhamettu, C. and Goldgof, D.B. Fluid Structure and MotionAnalysis from Multi-spectrum2D Cloud Image Sequences. In Proceedings IEEEConference on Computer Vision and Pattern Recognition,2000,2:2744.
[75] Fitzpatrick, J.M. The existence of geometrical density-image transformationscorresponding to object motion. Computer Vision, Graphics, and ImageProcessing,1988,44(2):155-174.
[76] Verri, A. and Poggio, T. Against quantitative optical flow. Proceedings of the1stInternational Conference on Computer Vision London England,1987:171-180.
[77] Cao, C. and Newman, T.S. and Germany. G.Shape-based mechanisms forcontent-based retrieval of aurora images. In Proceedings of SPIE, theInternational Society for Optical Engineering,2006.
[78] Davis, N. The aurora watcher's handbook. University of Alaska Press Fairbanks,1992.
[79] Wallace, J.M. and Foss, J.F. The measurement of vorticity in turbulent flows.Annual Review of Fluid Mechanics,1995,27(1):469-514.
[80] Black, M.J. Robust incremental optical flow. Yale University, New Haven, CT,1992:280.
[81] Black, M.J. and Anandan, P. The robust estimation of multiple motions:parametric and piecewise-smooth flow fields. Computer Vision ImageUnderstanting,1996,63(1):75-104.
[82] Lucas, B.D. and Kanade, T. An iterative image registration technique with anapplication to stereo vision. International joint conference on artificialintelligence,1981,3:674-679.
[83] Horwitz, J.L. and Akasofu, S.I. The response of the dayside aurora to sharpnorthward and southward transitions of the interplanetary magnetic field and tomagnetospheric substorms. Journal of Geophysical Research,1977,82(19):2723-2734.
[84] Sandholt, PE and Farrugia, CJ and Cowley, SWH and Lester, M. Dayside auroralbifurcation sequence during By-dominated interplanetary magnetic field:Relationship with merging and lobe convection cells. Journal of GeophysicalResearch,2001,106(A8):15429.
[85] Drury, E.E. and Mende, S.B. and Frey, H.U. and Doolittle, J.H. SouthernHemisphere poleward moving auroral forms. Journal of Geophysical Research.,2004,108(A3):1114.
[86] Nelson, R. C. and Polana, R. Qualitative recognition of motion using temporaltexture. CVGIP: Image Understand,1992,56(1):78-89.
[87] Chetverikov, D., Péteri, R. A Brief Survey of Dynamic Texture Description andRecognition. Computer Recognition Systems,2005:17-26.
[88] Peh, C.H. and Cheong, L.F. Synergizing spatial and temporal texture. IEEE Transon Image Processing,2002,11(10):1179-1191.
[89] Fablet, R. and Bouthemy, P. Motion recognition using nonparametric imagemotion models estimated from temporal and multiscale co-occurrence statistics.IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(12):1619-1624.
[90] Saisan P. and Doretto G. and Wu Y. and Soatto S. Dynamic texture recognition.In Proceedings of the2001IEEE Computer Society Conference on ComputerVision and Pattern Recognition,2001, pages II:58-63.
[91] Fazekas, S., Amiaz, T., Chetverikov, D., Kiryat, N. Dynamic texture detectionbased on motion analysis. International journal of computer vision,2009,82(1):48-63.
[92] Ghanem, B. and Ahuja, N. Extracting a fluid dynamic texture and thebackground from video. IEEE Conference on Computer Vision and PatternRecognition,2008:1-8.
[93]卢宗庆.运动图像分析中的光流计算方法研究.博士学位论文,西安电子科技大学，2007.
[94]高琴,刘立波,赵必强,万卫星.基于修正Kp指数的中纬区单站电离层暴预报方法研究.地球物理学进展,2009,6:1943-1950.
[95] Ho, T.K. and Basu, M. Complexity measures of supervised classificationproblems. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002:289-300.
[96]李洁.基于自然计算的模糊聚类新算法研究.博士学位论文,西安电子科技大学，2004.
[97]赵恒.数据挖掘中聚类若干问题研究.博士学位论文,西安电子科技大学，2005.
[98] Anderberg, M.R. Cluster analysis for applications. Academic Press,1973.
[99] Chen, M.S. and Han, J. and Yu, P.S. Data mining: An overview from a databaseperspective. IEEE Transactions on Knowledge and data Engineering,1996,8(6):866-883.
[100] Rastogi, R. and Shim, K. Mining optimized association rules with categoricaland numeric attributes. IEEE Transactions on Knowledge and Data Engineering,2002:29-50.
[101] Mahdavi, M. and Chehreghani, M.H. and Abolhassani, H. and Forsati, R. Novelmeta-heuristic algorithms for clustering web documents. Applied Mathematicsand Computation,2008,201(1-2),441-451.
[102] Zhou, D. and Bousquet, O. and Lai, T.N. and Weston, J. and Scholkopf, B.Learning with Local and Global Consistency. Advances in neural informationprocessing systems,2004,16:321.
[2] Hu, Z.-J., Yang, H., Huang, D., Araki, T., Sato, N., et al. Synoptic distribution ofdayside aurora: multiple-wavelength all-sky observation at Yellow River Stationin Ny-lesund, Svalbard. Journal of Atmospheric and Solar-Terrestrial Physics,2009,71:794–804.
[3] Feldstein, Y., Elphinstone, R. Aurorae and the large-scale structure of themagnetosphere. Journal of Geomagnetism and Geoelectricity,1992,44:1159–1174.
[4] Kullen, A., Brittnacher, M., Cumnock, J.A., Blomberg, L.G. Solar winddependence of the occurrence and motion of polar auroral arcs: a statistical study.Journal of Geophysical Research,2002,107(A11):1362.
[5] THEMIS project. http://aurora.phys.ucalgary.ca/themis/themis_main.html.
[6] St rmer, C. The Polar Aurora. Clarendon Press,1955, Oxford.
[7] Vallance, J.A. Aurora.1974, D. Riedel, Dordrecht, Holland.
[8] Steen,, Br ndstr m, U., Gustavsson, B. ALIS—a multi station imaging systemat high latitude with multi disciplinary scientific objectives. In: Proceedings ofthe13th ESA Symposium on European Rocket and Balloon Programmes andRelated Research,1997:261-266.
[9] Marklund, G.T. Auroral arc classification scheme based on the observedarc-associated electric field pattern. Planetary and Space Science,1984,32(2):193-211.
[10] Ayukawa, M., Makita, K., Yamagishi, H., Ejiri, M., Sakanoi, T. Characteristics ofpolar cap aurora. Journal of Atmospheric and Terrestrial Physics,1996,58:1885-1894.
[11] Yang, H., Sato, N., Makita, K., Kikuchi, M., Kadokura, A., Ayukawa, M., Hu,H.Q., Liu, R.Y., H ggstr m, I. Synoptic observations of auroras along thepost-noon oval: a survey with all-sky TV observations at Zhongshan, Antarctica.Journal of Atmospheric and Solar-Terrestrial Physics,2000,62:787-797.
[12] Syrj suo, M.T., Donovan, E.F., Qin, X., Yang, Y.-H. Automatic classification ofauroral images in substorm studies. In: Proceedings of the Eight InternationalConference on Substorms (ICS8), Canada,2006:309-313.
[13] Donovan, E. F., Trondsen, T. S., Cogger, L. L. and Jackel, B. J., Auroral imagingin Canadian CANOPUS and NORSTAR programs, Proc. of Atmospheric Studiesby Optical Methods,2003:109–112.
[14] Syrj suo, M.T.,2001. Auroral monitoring network: from all-sky camera systemto automated image analysis. Ph.D Thesis, Finnish Meteorological Institute,Helsinki, Finland, ISBN951-697-551-8.
[15] Syrj suo, M.T., Donovan, E.F. Content-based retrieval of auroralimages—thousands of irregular shapes. In: Proceedings of the IASTEDInternational Conference on Visualization, Imaging, and Image Processing,2004:224-228.
[16] Syrj suo, M.T., Donovan, E.F. Diurnal auroral occurrence statistics obtained viamachine vision. Annales Geophysicae,2004,22(4):1103–1113.
[17] Syrj suo, M.T., Donovan, E.F. Using relevance feedback in retrieving auroralimages. In: Proceedings of the IASTED Computational Intelligence,2005:420-425.
[18] Syrj suo, M.T., Kauristie, K., Pulkkinen, T.I. A search engine for auroral forms.Advances in Space Research,2001:28(11),1611-1616.
[19] G. V. G. Baranoski, J. G. Rokne, P. Shirley, T. S. Trondsen, and R. Bastos.Simulating the aurora. The Journal of Visualization and Computer Animation,2003,14:43-59.
[20] G. V. G. Baranoski, J. Wan, J. G. Rokne, and I. Bell. Simulating the dynamics ofauroral phenomena. ACM Trans. Graph,2005,24:37-59.
[21] Kauristie, K., Syrj suo, M.T., Amm, O., Viljanen, A., Pulkkinen, T.I.,Opgenoorth, H.J. A Statistical study of evening sector arcs and electrojets.Advances in Space Research,2001,28(11):1605–1610.
[22] Qin, X., Yang, Y.-H. Basic Gray Level Aura Matrices: Theory and its Applicationto Texture Synthesis. IEEE ICCV, Beijing, China,2005:128–135.
[23] Blixt, E., Semeter, J., Ivchenko, N. Optical flow analysis of the aurora borealis.Geoscience and Remote Sensing Letters, IEEE,2006,3:159-163.
[24] Horn, B., Schunck, B. Determining optical flow. Artificial intelligence,1981,17:185-203.
[25]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——中国北极黄河站首次极光观测初步分析.极地研究,2005,17:107-114.
[26] Ojala, T., Pietik inen, M., Harwood, D. A comparative study of texture measureswith classification based on feature distributions. Pattern Recognition,1996,29:51–59.
[27] Ojala, T., Pietik inen, M., M enp, T. Multiresolution grayscale and rotationinvariant texture classification with local binary patterns. IEEE Transactions onPattern Analysis and Machine Intelligence,2002,24(7):971–987.
[28] Ahonen, T., Hadid, A., Pietika¨inen, M. Face description with local binarypatterns: application to face recognition. IEEE Transactions on Pattern Analysisand Machine Intelligence,2006,28(12):2037–2041.
[29] Theodoridis, S., Koutroumbas, K. Pattern Recognition, third ed. Elsevier,2006.
[30] Duda, R.O., Hart, P.E., Stork, D.G. Pattern Classification Second Edition JohnWiley&Sons,2001.
[31] NSTC Subcommittee on Biometrics. Biometrics testing and statistics. Availableonline: http://www.biometrics.gov/Documents/BioTestingAndStats.pdf.
[32] Akasofu, S.I., Kan, J.R. Dayside and nightside auroral arc systems. GeophysicalResearch Letters,1980,7(10):753–756.
[33] Liou, K., Newell, P.T., Meng, C.-I., Brittnacher, M. Parks, G. Synoptic auroraldistribution: a survey using polar ultraviolet imagery. Journal of GeophysicalResearch,1997,102:27197–27205.
[34] Newell, P.T., Lyons, K.M., Meng, C.-I. A large survey of electron accelerationevents. Journal of Geophysical Research,1996,101:2599–2614.
[35] Wang Q, Liang J, Hu Z J, Hu H H, Zhao H, Hu H Q, Gao X B, Yang H G. SpatialTexture Based Automatic Classification of Dayside Aurora in All-Sky Images.Journal of Atmospheric and Solar-Terrestrial Physics,2010,72(5-6):498-508.
[36] Xu, R., Wunsch, D. Survey of clustering algorithms. IEEE Transactions onneural network,2005,16:645-678.
[37] Shi, J., Malik, J. Normalized cuts and image segmentation. IEEE Transactions onPattern Analysis and Machine Intelligence,2002,22(8):888-905.
[38] Jolliffe, I. Principal component analysis,2002, Springer Verlag.
[39] Jain, A., Dubes, R. Algorithms for clustering data. Prentice-Hall AdvancedReference Series,1988,320.
[40] Jain, A., Murty, M., Flynn, P. Data Clustering: A Review. ACM computingSurveys,1999,31:264-323.
[41] Chris, H., Ding, Q., Li, T. Adaptive dimension reduction using discriminantanalysis and k-means clustering. In International Conference on MachineLearning,2007:521-528.
[42] Halkidi, M., Batistakis, Y., Vazirgiannis, M. On Clustering Validation Techniques.Intelligent Information Systems,2001,17(2-3):107-145.
[43] Halkidi, M., Batistakis, Y., Vazirgiannis, M. Cluster validity methods: part I.ACM SIGMOD Record,2002,31, Issue2:40-45.
[44] Halkidi, M., Batistakis, Y., Vazirgiannis, M. Cluster validity methods: part II.ACM SIGMOD Record,2002,31, Issue3:19-27.
[45] Yang, Y., Xu, D., Nie, F., Yan, S., Zhuang, Y. Image clustering using localdiscriminant models and global integration. IEEE Transactions on ImageProcessing,2010,19:2761-2773.
[46] Chen, W., Song, Y., Bai, H., Lin, C., Chang, E. Parallel spectral clustering indistributed systems. IEEE Transactions on Pattern Analysis and MachineIntelligence,2011,33(3):568-586.
[47] Von Luxburg, U. A tutorial on spectral clustering. Statistics and Computing,2007,17(4):395-416.
[48] Abascal, F. and Valencia, A. Clustering of proximal sequence space for theidentification of protein families, Bioinformatics,2002,18(7):908.
[49] Chen, C. Handbook of pattern recognition and computer vision. World Scientific,2010.
[50] Legány, C. and Juhász, S. and Babos, A. Cluster validity measurementtechniques. Proceedings of the5th WSEAS International Conference onArtificial Intelligence, Knowledge Engineering and Data Bases,2006:388-393.
[51] Everitt, B. and Landau, S. and M. Leese. Cluster Analysis. New York: OxfordUniv. Press,2001.
[52] Davies, DL and Bouldin, DW. A cluster separation measure. IEEE Transactionson Pattern Analysis and Machine Intelligence,1979,1:224-227.
[53] Dunn, J.C. Well separated clusters and optimal fuzzy partitions. Journal ofCybernetica,1974,4:95-104.
[54] Halkidi, M. and Vazirgiannis, M. Clustering validity assessment: Finding theoptimal partitioning of a data set. Proceedings IEEE International Conference onData Mining (ICDM),2002:187-194.
[55] Halkidi, M. and Vazirgiannis, M. and Batistakis, Y. Quality scheme assessment inthe clustering process. Principles of Data Mining and Knowledge Discovery,2000:265-276.
[56] Caliński, T. and Harabasz, J. A dendrite method for cluster analysis.Communications in Statistics-Theory and Methods,1974,3(1):1-27.
[57] Kaufman, L. and Rousseeuw, P.J. Finding groups in data—an introduction tocluster analysis. New York: Wiley,1990.
[58] Maulik, U. and Bandyopadhyay, S. Performance evaluation of some clusteringalgorithms and validity indices. IEEE Transactions on Pattern Analysis andMachine Intelligence,2002,24(12):1650-1654.
[59] Xie, X.L. and Beni, G. A validity measure for fuzzy clustering. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,13(8):841-847.
[60] Kim, D.J. and Park, Y.W. and Park, D.J. A novel validity index for determinationof the optimal number of clusters. IEICE transactions on information andsystems,2001,84(2):281-285.
[61] Kwon, S.H. Cluster validity index for fuzzy clusterin. Electronics Letters,2002,34(22):2176-2177.
[62] Dungey, J. W. Interplanetary Magnetic Field and the Auroral Zones. PhysicalReview Letters,1961,6(2):47-48.
[63] Lemaire, J. and Roth, M. Penetration of solar wind plasma elements into themagnetosphere. Journal of Atmospheric and Terrestrial Physics,1978,40(3):331-335.
[64] Wei, C. and Lee, L. and La Belle-Hamer, A. A simulation study of the vortexstructure in the low-latitude boundary layer. Journal of Geophysical Research,1990,95(A12):20793.
[65] Haerendel, G. and Buchert, S. and La Hoz, C. and Raaf, B. and Rieger, E. On theproper motion of auroral arcs. Journal of Geophysical Research,1993,98(A4):6087-6099.
[66] Kimball, J. and Hallinan, T. Observations of black auroral patches and of theirrelationship to other types of aurora. Journal of Geophysical Research,1998,103(A7):14671.
[67] Blixt, E.M. and Semeter, J. Relative drift between black aurora and theionospheric plasma. Ann. Geophys,2005,23(5):1611-1621.
[68] Simpson, J.J. and Gobat, J.I. Robust velocity estimates, stream functions, andsimulated Lagrangian drifters from sequential spacecraft data., IEEETransactions on Geoscience and Remote Sensing,2002,32(3):479-493.
[69] Shukla, J. and Saha, K. Computation of non-divergent streamfunction andirrotational velocity potential from the observed winds. Monthly weather review,1974,102(6):419-425.
[70] Jain, R. and Nagel, H.H. On the analysis of accumulative difference picturesfrom image sequences of real world scenes. IEEE Transactions on PatternAnalysis and Machine Intelligence,1979, PAMI-1(2):206-214.
[71] Koenderink, J.J. and Doorn, A.J. Visual perception of rigidity of solid shape.Journal of Mathematical Biology,1976,3(1):79-85.
[72] Corpetti, T., Mémin, é., Pérez, P. Dense estimation of fluid flows. IEEETransactions on Pattern Analysis and Machine Intelligence,2002,24(3):365-380.
[73] Wildes, R.P. and Amabile, M.J. and Lanzillotto, A.M. and Leu, T.S. Recoveringestimates of fluid flow from image sequence data. Computer vision and Imageunderstanding,2000,80(2):246-266.
[74] Zhou, L. and Kambhamettu, C. and Goldgof, D.B. Fluid Structure and MotionAnalysis from Multi-spectrum2D Cloud Image Sequences. In Proceedings IEEEConference on Computer Vision and Pattern Recognition,2000,2:2744.
[75] Fitzpatrick, J.M. The existence of geometrical density-image transformationscorresponding to object motion. Computer Vision, Graphics, and ImageProcessing,1988,44(2):155-174.
[76] Verri, A. and Poggio, T. Against quantitative optical flow. Proceedings of the1stInternational Conference on Computer Vision London England,1987:171-180.
[77] Cao, C. and Newman, T.S. and Germany. G.Shape-based mechanisms forcontent-based retrieval of aurora images. In Proceedings of SPIE, theInternational Society for Optical Engineering,2006.
[78] Davis, N. The aurora watcher's handbook. University of Alaska Press Fairbanks,1992.
[79] Wallace, J.M. and Foss, J.F. The measurement of vorticity in turbulent flows.Annual Review of Fluid Mechanics,1995,27(1):469-514.
[80] Black, M.J. Robust incremental optical flow. Yale University, New Haven, CT,1992:280.
[81] Black, M.J. and Anandan, P. The robust estimation of multiple motions:parametric and piecewise-smooth flow fields. Computer Vision ImageUnderstanting,1996,63(1):75-104.
[82] Lucas, B.D. and Kanade, T. An iterative image registration technique with anapplication to stereo vision. International joint conference on artificialintelligence,1981,3:674-679.
[83] Horwitz, J.L. and Akasofu, S.I. The response of the dayside aurora to sharpnorthward and southward transitions of the interplanetary magnetic field and tomagnetospheric substorms. Journal of Geophysical Research,1977,82(19):2723-2734.
[84] Sandholt, PE and Farrugia, CJ and Cowley, SWH and Lester, M. Dayside auroralbifurcation sequence during By-dominated interplanetary magnetic field:Relationship with merging and lobe convection cells. Journal of GeophysicalResearch,2001,106(A8):15429.
[85] Drury, E.E. and Mende, S.B. and Frey, H.U. and Doolittle, J.H. SouthernHemisphere poleward moving auroral forms. Journal of Geophysical Research.,2004,108(A3):1114.
[86] Nelson, R. C. and Polana, R. Qualitative recognition of motion using temporaltexture. CVGIP: Image Understand,1992,56(1):78-89.
[87] Chetverikov, D., Péteri, R. A Brief Survey of Dynamic Texture Description andRecognition. Computer Recognition Systems,2005:17-26.
[88] Peh, C.H. and Cheong, L.F. Synergizing spatial and temporal texture. IEEE Transon Image Processing,2002,11(10):1179-1191.
[89] Fablet, R. and Bouthemy, P. Motion recognition using nonparametric imagemotion models estimated from temporal and multiscale co-occurrence statistics.IEEE Transactions on Pattern Analysis and Machine Intelligence,2003,25(12):1619-1624.
[90] Saisan P. and Doretto G. and Wu Y. and Soatto S. Dynamic texture recognition.In Proceedings of the2001IEEE Computer Society Conference on ComputerVision and Pattern Recognition,2001, pages II:58-63.
[91] Fazekas, S., Amiaz, T., Chetverikov, D., Kiryat, N. Dynamic texture detectionbased on motion analysis. International journal of computer vision,2009,82(1):48-63.
[92] Ghanem, B. and Ahuja, N. Extracting a fluid dynamic texture and thebackground from video. IEEE Conference on Computer Vision and PatternRecognition,2008:1-8.
[93]卢宗庆.运动图像分析中的光流计算方法研究.博士学位论文,西安电子科技大学，2007.
[94]高琴,刘立波,赵必强,万卫星.基于修正Kp指数的中纬区单站电离层暴预报方法研究.地球物理学进展,2009,6:1943-1950.
[95] Ho, T.K. and Basu, M. Complexity measures of supervised classificationproblems. IEEE Transactions on Pattern Analysis and Machine Intelligence,2002:289-300.
[96]李洁.基于自然计算的模糊聚类新算法研究.博士学位论文,西安电子科技大学，2004.
[97]赵恒.数据挖掘中聚类若干问题研究.博士学位论文,西安电子科技大学，2005.
[98] Anderberg, M.R. Cluster analysis for applications. Academic Press,1973.
[99] Chen, M.S. and Han, J. and Yu, P.S. Data mining: An overview from a databaseperspective. IEEE Transactions on Knowledge and data Engineering,1996,8(6):866-883.
[100] Rastogi, R. and Shim, K. Mining optimized association rules with categoricaland numeric attributes. IEEE Transactions on Knowledge and Data Engineering,2002:29-50.
[101] Mahdavi, M. and Chehreghani, M.H. and Abolhassani, H. and Forsati, R. Novelmeta-heuristic algorithms for clustering web documents. Applied Mathematicsand Computation,2008,201(1-2),441-451.
[102] Zhou, D. and Bousquet, O. and Lai, T.N. and Weston, J. and Scholkopf, B.Learning with Local and Global Consistency. Advances in neural informationprocessing systems,2004,16:321.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |