Automatic Classification of Dayside Aurora in All-Sky Images Using a Multi-Level Texture Feature Representation

Abstract:

In this paper, we propose an aurora classification method using a multi-level feature representation aimed to capture both global and local texture information, and to reduce the feature space dimension substantially. First-order and second-order statistics are computed for an input image and its low-frequency scaled images at three lower levels obtained using wavelet decomposition. The features include gray level distribution, co-occurrence matrix features, and run-length matrix features. A support vector machine (SVM) classifier was trained and tested on a Chinese Arctic Yellow River Station dayside aurora image dataset. Classification performance was evaluated and compared with those of k-nearest neighbor (KNN) classifiers and back-propagation neural networks (BPNN). To explore the possibility of using a smaller feature space, we used a Minimum-Redundancy Max-Relevance feature selection strategy. The result shows that there is only indistinct performance decrease by reducing the feature vector from a total of 88 to the most discriminatory 38 features. This proves that our multi-level feature representation is very robust.

• Periodical: Advanced Materials Research (Volumes 341-342)
• Pages: 158-162
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.341-342.158
• Online since: September 2011
• Authors: Shen Miao Han, Zhen Sen Wu, Guang Li Wu, Jun Tan
• Keywords: Aurora, Grey Level Co-Occurrence Matrix, Run-Length, Support Vector Machine (SVM), Wavelet Decomposition

References

[1] Ayukawa, M. and Makita, K., Characteristics of polar cap aurora, Journal of Atmospheric and Terrestrial Physics 58, 1885-1894 (1996).

DOI: 10.1016/0021-9169(95)00180-8

[2] Fu, R., Li, J., Gao, X. and Jian, Y., Automatic Aurora Images Classification Algorithm Based on Separated Texture, Proc. International Conference on Robotics and Biomimetics, (2009).

DOI: 10.1109/robio.2009.5420722

[3] Wang, Q., Liang, J., Hu, Z.J., Hua, H.H., Zhao, H., Hu, H.Q., Gao, X. and Yang, H., Spatial texture based automatic classification of dayside aurora in all-sky images, Journal of Atmospheric and Solar-Terrestrial Physics 72, 498–508 (2010).

DOI: 10.1016/j.jastp.2010.01.011

[4] Syrjäsuo, M.T. and E.F. Donovan, Analysis of Aurora Images: Detection and Tracking, Geophysica 38(1-2), 3-14 (2002).

[5] Syrjäsuo, M.T. and Donovan, E.F., Automatic classification of aurora images in substorm studies, Int. Conf. Substorms 8, 309–313 (2006).

[6] Syrjäsuo, M.T. and Donovan, E.F., Diurnal aurora occurrence statistics obtained via machine vision, Annales Geophysicae 22, 1103–1113 (2004).

DOI: 10.5194/angeo-22-1103-2004

[7] Wang Y., Gao X., Fu R. and Jian Y., Dayside corona aurora classification based on X-gray level aura matrices, Proc. of the ACM International Conference on Image and Video Retrieval, 282–287 (2010).

DOI: 10.1145/1816041.1816083

[8] Mallat, S. and Zhong, S., Characterization of signals from multiscale edges, IEEE Transactions on Pattern Analysis and Machine Intelligence 14(7), 710-732 (1992).

DOI: 10.1109/34.142909

[9] Haralick, R.M., Shanmugam, K., Dinstein, I., Textural Features for Image Classification, IEEE Transactions on Systems, Man, and Cybernetics 3(6), 610-621 (1973).

DOI: 10.1109/tsmc.1973.4309314

[10] Tang, X., Texture information in run-length matrices, IEEE transactions on image processing 7(11), 1602-9 (1998).

DOI: 10.1109/83.725367

[11] Galloway, M., Texture analysis using gray level run lengths, Computer Graphics and Image Processing 4(2), 172-179 (1975).

DOI: 10.1016/s0146-664x(75)80008-6

[12] Vapnik, V., The Nature of Statistical Learning Theory, Springer, N.Y., (1998).

[13] Duda, R.O., Hart, P.E. and Stork, D.G., Pattern Classification, 2nd edition, Wiley-Interscience, Chapter 4, 18-21 (2000).

[14] Russell, S.J. and Norvig, P., Artificial Intelligence: A Modern Approach, 2rd edition, Prentice Hall, 578, (1995).

[15] Peng, H., Long, F. and Ding, C., Feature selection based on mutual information: criteria of max-dependency, max-relevance, and min-redundancy, IEEE transactions on pattern analysis and machine intelligence 27(8), 1226-38 (2005).

DOI: 10.1109/tpami.2005.159