Vladimir Lukin, Galina Proskura, Irina Vasilieva


The subject of this study is the pixel-by-pixel controlled classification of multichannel satellite images distorted by additive white Gaussian noise. The paper aim is to study the effectiveness of various methods of image classification in a wide range of signal-to-noise ratios; an F-measure is used as a criterion for recognition efficiency. It is a harmonic mean of accuracy and completeness: accuracy shows how much of the objects identified by the classifier as positive are positive; completeness shows how much of the positive objects were allocated by the classifier. Tasks: generate random valuesof the brightness of the noise components, ensuring their compliance with the accepted probabilistic model; implement the procedures of element-wise controlled classification according to the methods of support vectors, logistic regression, neural network based on a multilayer perceptron for images distorted by noise; evaluate and analyze the results of objects bezel-wise classification of noisy images; investigate the effect of noise variance on classification performance. The following results are obtained. Algorithms of pixel-by-pixel controlled classification are implemented. A comparative analysis of classification efficiency in noisy images is performed. Conclusions are drawn. It is shown that all classifiers provide the best results for classes that mainly correspond to areal objects (Water, Grass) while heterogeneous objects (Urban and, especially, Bushes) are recognized in the worst way; classifiers based on the support vector machine and logistic regression show low recognition accuracy of extended objects, such as a narrow river (that belongs to the wide class of "water"). The presence of noise in the image leads to a significant increase in the number of recognition errors, which mainly appear as isolated points on the selected segments, that is, incorrectly classified pixels. In this case, the best value of the classification quality indicator is achieved using neural networks based on a multilayer perceptron.


classification; additive white Gaussian noise; pixel-by-pixel controlled classification; the probability of correct classification

Full Text:



Abramov, N. S., Makarov, D. A., Talalaev, A. A., Fralenko, V. P. Sovremennye metody intellektual'noi obrabotki dannykh DZZ [Modern methods for intelligent processing of Earth remote sensing data]. Programmnye sistemy : teoriya i prilozheniya – Program Systems : Theory and Applications, 2018, vol. 9, no. 4(39), pp. 417-442. (In Russian). doi: 10.25209/ 2079-3316-2018-9-4-417-442.

Meola, J., Eismann, M. T., Moses, R. L., Ash J. N. Modeling and estimation of signal-dependent noise in hyperspectral imagery. J. Appl. Opt., 2011, vol. 50, pp. 3829-3846.

Lukin, V. V., Ponomarenko, N. N., Egiazarian, K., Astola, J. Adaptive DCT-based filtering of images corrupted by spatially correlated noise. Proc. SPIE Conference Image Processing: Algorithms and Systems VI., 2008, vol. 6812. 12 p.

Afanas'ev, A. A., Zamyatin A. V. Gibridnye metody avtomatizirovannoi identifikatsii izmenenii landshaftnogo pokrova po dannym distantsionnogo zondirovaniya Zemli v usloviyakh shumov [Hybrid methods of automated identification of changes in land-scape cover by Earth remote sensing data in the condi-tions of noises]. Komp'yuternaya optika – Computer optics, 2017, vol. 41, no. 3, pp. 431-440. (In Russian). doi: 10.18287/2412-6179-2017-41-3-431-440.

Proskura, G. A., Lukin, V. V., Vasilyeva, I. K. Improvement of Multichannel Image Classification by Combining Elementary Classifiers. Proc. Internat. Scientific-Practical Conf. on Problems of Infocommunications. Science and Technology, Kyev, Ukraine, 8-11 Oct. 2019. 5 p.

Chih-Wei Hsu, Chih-Chung Chang, Chih-Jen Lin. A practical guide to support vector classification. Available at: /papers /guide/guide.pdf (аccessed 9.10.2019).

Bach, F., Moulines, E. Non-Asymptotic Analysis of Stochastic Approximation Algorithms for Machine Learning. Proc. 25th Annual Conf. on Advances in Neural Information Processing Systems 24, Granada, Spain, 12-14 Dec. 2011, pp. 451-459.

Rong-En Fan, Kai-Wei Chang, Cho-Jui Hsieh, Xiang-Rui Wang, Chih-Jen Lin A library for large linear classification. Journal of Machine Learning Research, 2008, no. 9, pp. 1871 – 1874.

Venkat, N., Gudivada, C.R., Computational Analysis and Understanding of Natural Languages: Principles, Methods and Applications. Handbook of Statistics, 2018., vol. 38, pp. 2-515.

Alsmadi, M. S., Omar, K. B., Noah, S. A. Back Propagation Algorithm: The Best Algorithm Among the Multi-layer Perceptron Algorithm. IJCSNS International Journal of Computer Science and Network Security, 2004, vol. 9, no. 4, pp. 378-383.

Kussul, N., Lavreniuk, M., Skakun, S., Shelestov, A. Deep learning classification of land cover and crop types using remote sensing data. IEEE Geoscience and Remote Sensing Letters, 2017, vol. 14(5), pp. 778-782. doi: 10.1109/LGRS.2017. 2681128.



  • There are currently no refbacks.