Abstract
The technique of a search for images of cloudiness of various types from MODIS satellite images based on a comparison with archive data of observations on the network of meteorological stations is presented. Based on an expert estimate, 14 types of cloudiness possessing a unique structure on images recorded with a spatial resolution of 250 m are identified. Images of cloudiness of these types and results of investigations of their texture parameters found based on the statistical gray-level co-occurrences matrix (GLCM) approach are presented. For the indicated cloudiness types, characteristic texture features or their combinations are determined. To classify the cloudiness based on information on the texture parameters, it is proposed to use the neural network based on the three-layer perceptron. The modified method of adaptive tuning of the learning rate of the neural network is described. Results of cloudiness classification and their reliability are discussed.
Similar content being viewed by others
REFERENCES
Astafurov, V.G. and Skorokhodov, A.V., Segmentation of satellite images of cloudiness according to texture signs using neural network technologies, Issled. Zemli Kosmosa, 2011, no. 6, pp. 10–20.
Astafurov, V.G., Rasskazchikova, T.M., and Skorokhodov, A.V., Interpretation of data of cloud remote sensing from space in the visible range of the spectrum, Russ. Phys. J., 2012, vol. 55, no. 3, pp. 323–329.
Bankert, R.L. and Wade, R.H., Optimization of an instance-based goes cloud classification algorithm, J. Appl. Meteorol. Climatol., 2007, vol. 46, pp. 36–49.
Bankert, R.L., Rabindra, P., and Sengupta, S.K., A Probabilistic Neural Network Approach to Cloud Classification, Monterey, Calif.: Naval Oceanographic and Atmospheric Research Laboratory, 1992, Tech. Note NOARL-TN-173.
Bankert, R.L., Mitrescu, C., Miller, S.W., and Wade, R.H., Comparison of GOES cloud classification algorithms employing explicit and implicit physics, J. Appl. Meteorol. Climatol., 2009, vol. 48, pp. 1411–1421.
Callan, R., Essence of Neural Networks, Upper Saddle River, N.J.: Prentice Hall, 1998; Moscow: Vil’yams, 2001.
Davis, S.M., Landgrebe, D.A., Phillips, T.L., Swain, P.H., Hoffer, R.M., Lindenlaub, J.C., and Silva, L.F., Remote Sensing: The Quantitative Approach, New York: McGraw-Hill, 1978; Moscow: Nedra, 1983.
Galushkin, A.I., Teoriya neironnykh setei (Theory of Neural Networks), vol. 1, Moscow: IPRZhR, 2000.
German, M.A., Sputnikovaya meteorologiya (Satellite Meteorology), Leningrad: Gidrometeoizdat, 1975.
Gill, P., Murray, W., and Wright, M., Practical Optimization, Emerald Group Publishing Limited, 1982; Moscow: Mir, 1985.
Haikin, S., Neural Networks: A Comprehensive Foundation, Upper Saddle River, N.J.: Prentice Hall, 1994; Moscow: Vil’yams, 2008.
Haralick, R.M. and Bosley, R., Texture features for image classification, Proc. of the Third ERTS Symp. 10–14 December 1973, Washington, D.C., 1973, pp. 1929–1969.
Kod dlya sostavleniya gidrometeorologicheskikh radiogramm na sudakh KN-01 (KN-01 Code for Composing Hydrometeorological Radiograms on Ships), Leningrad: Gidrometizdat, 1981.
Kolodnikova, N.V., Review of texture indicators for pattern recognition problems, Dokl. Tomsk. Gos. Univ. Sist, Upr. Radioelektron., 2004, no. 1, pp. 113–124.
Miller, S.W. and Emery, W.J., An automatic neural-network cloud classifier for use over land and ocean surface, J. Appl. Meteorol., 1997, vol. 36, pp. 1346–1362.
Neironnye seti. STATISTICA Neural Networks: Metodologiya i tekhnologii sovremennogo analiza dannykh (Neural Networks. STATISTICA Neural Networks: Methodology and Technologies of Modern Data Analysis), Borovikov, V.P., Ed., Moscow: Goryachaya liniya–Telekom, 2008.
Oblaka i oblachnaya atmosfera. Spravochnik (Clouds and Cloudy Atmosphere. A Handbook), Mazin, I.P. and Khrgian, A.Kh., Eds., Leningrad: Gidrometeoizdat, 1989.
Osowski, S., Sieci neuronowe do przetwarzania informacji, Warszawa: Oficyna Wydawnicza Politechniki Warszawskiej, 2000; Moscow: Finansy i statistika, 2002.
Potapov, A.A., Fraktaly v radiofizike i radiolokatsii: Topologiya vyborki (Fractals in Radiophysics and Radar Direction Finding: Topology of Sampling), Moscow: Universitetskaya kniga, 2005.
Skorokhodov, A.V. and Tungusova, A.V., Comparative analysis of gradient methods of minimization in the multilayer perceptron learning problem, Dokl. Tomsk. Gos. Univ. Sist, Upr. Radioelektron., 2013, no. 2, pp. 98–102.
Volkova, E.V. and Uspenskii, A.B., Estimates for cloud cover parameters during daylight hours according to data from the METEOSAT-8 geostationary meteorological satellite, Sovrem. Probl. Distantsionnogo Zondirovaniya Zemli Kosmosa, 2008, vol. 5, no. 1, pp. 441–450.
Welch, R.M., Kuo, K.S., Sengupta, S.K., and Chen, D.W., Cloud field classification based upon high spatial resolution texture feature (I): Gray-level cooccurence matrix approach, J. Geophys. Res., 1988, vol. 93, pp. 12663–12681.
ACKNOWLEDGMENTS
We are grateful to T.M. Rasskazchikova for participating in peer inspections on the determination of cloudiness types with a unique image texture.
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated by A. Nikol’skii
Rights and permissions
About this article
Cite this article
Astafurov, V.G., Skorokhodov, A.V. Application of Neural Network Technologies for the Classification of Cloudiness by Texture Parameters of MODIS High-Resolution Images. Izv. Atmos. Ocean. Phys. 55, 1012–1021 (2019). https://doi.org/10.1134/S000143381909007X
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S000143381909007X