Vol. 38, issue 05, article # 9

Abstract:

This paper considers the problem of classifying cloud images, which are complex texture structures with heterogeneous characteristics. Traditional image analysis methods do not always adequately classify such images, and modern deep learning methods require large amount of data and computational resources. The research focuses on evaluating the feasibility of developing a hybrid method combining traditional statistical approaches to texture description and state-of-the-art deep learning techniques. It was hypothesised that the high-level features extracted by a neural network during training can be insufficiently sensitive to subtle local differences in cloud formations. The hybrid approach was implemented and analysed; low-level texture features were extracted from the images before being analysed by the neural network. However, the test results showed that this technique did not improve the classification quality and turned out to be less effective in terms of accuracy compared to the use of unprocessed images. The results of this work can be of interest to specialists in of Earth remote sensing data analysis, meteorology, and development of new texture image analysis methods.

Keywords:

image classification, texture characteristics, image processing, neural network

Figures:

References:

1. Hassaballah M., Abdelmgeid A.A., Alshazly H.A. Image features detection, description, and matching // Image Feature Detectors and Descriptors: Foundations and Applications. Cham: Springer, 2016. P. 11–45. DOI: 10.1007/978-3-319-28854-3_2.
2. Lei B.J., Hendriks E.A., Reinders M.J.T. On feature extraction from images. URL: https://www.academia.edu/656131/On_Feature_Extraction_from_Images (last access: 04.04.2025).
3. Bloshchinskiy V.D., Kramareva L.S., Shamilova Yu.A. Detektirovanie oblachnogo pokrova s ispol'zovaniem neironnoi seti po dannym pribora MSU-GS kosmicheskogo apparata «Arktika-M» N 1 // Optika atmosf. i okeana. 2024. V. 37, N 2. P. 99–104. DOI: 10.15372/AOO20240202; Bloshchinskiy V.D., Kramareva L.S., Shamilova Yu.A. Cloud Cover Detection using a neural network based on MSU-GS instrument data of Arktika-M No. 1 Satellite // Atmos. Ocean. Opt. 2024. V. 37, N 3. P. 394–399.
4. Astafurov V.G., Skorokhodov A.V., Kur’yanovich K.V., Mitrofanenko Ya.K. Harakteristiki razlichnykh tipov oblachnosti nad prirodnymi zonami Zapadnoi Sibiri po sputnikovym dannym MODIS // Optika atmosf. i okeana. 2020. V. 33, N 4. P. 266–271. DOI: 10.15372/AOO20200404; Astafurov V.G., Skorokhodov A.V., Kur’yanovich K.V., Mitrofanenko Ya.K. Parameters of different cloud types over the natural zones of Western Siberia according to MODIS satellite data // Atmos. Ocean. Opt. 2020. V. 33, N 5. P. 512–518.
5. Tassov K.L., Bekasov D.E. Obrabotka perekrytii v zadachakh otslezhivaniya ob"ektov v videopotoke // Inzhenernyi zhurnal: nauka i innovatsii. 2013. N 6. P. 1–27. DOI: 10.18698/2308-6033-2013-6-1099.
6. Ravi R., Yadhukrishna S.V., Prithviraj R. A face expression recognition using CNN & LBP // 2020 Fourth International Conference on Computing Methodologies and Communication (ICCMC). New York: IEEE, 2020. P. 684–689. DOI: 10.1109/ICCMC48092. 2020. ICCMC-000127.
7. Hamdi M. Ebrahim M.S., Jadhav M.E., Olayah F., Awaji B., Alalayah K.M. Hybrid models based on fusion features of a CNN and handcrafted features for accurate histopathological image analysis for DiagNsing malignant lymphomas // DiagNstics. 2023. V. 13, N 13. P. 2258. DOI: 10.3390/diagNstics13132258.
8. Gurunathan A., Krishnan B.A. Hybrid CNN-GLCM classifier for detection and grade classification of brain tumor // Brain Imaging Behav. 2022. V. 16, N 3. P. 1410–1427. DOI: 10.1007/s11682-021-00598-2.
9. Haralick R.M., Shanmugam K., Dinstein I.H. Textural features for image classification // IEEE Trans. Syst. Man Cybern. 1973. V. SMC-3, N 6. P. 610–621. DOI: 10.1109/TSMC.1973.4309314.
10. Nurtanio I., Zainuddin Z., Setiadi B.H. Cloud classification based on images texture features // IOP Conf. Ser.: Mater. Sci. Eng. 2019. V. 676, N 1. P. 012015. DOI: 10.1088/1757-899X/676/1/012015.
11. Ameur Z., Ameur S., Adane A., Sauvageot H., Bara K. Cloud classification using the textural features of Meteosat images // Int. J. Remote Sens. 2004. V. 25, N 21. P. 4491–4503. DOI: 10.1080/01431160410001735120.
12. Ojala T., Pietikainen M., Harwood D. Performance evaluation of texture measures with classification based on Kullback discrimination of distributions // Proc. 12th International Conference on Pattern Recognition.V. 1. 1994. P. 582–585. DOI: 10.1109/ICPR.1994.576366.
13. Petruk V., Samorodov A.V., Spiridonov I.N. Primenenie lokal'nykh binarnykh shablonov k resheniyu zadachi raspoznavaniya lits // Vestn. Mosk. gos. tekhn. un-ta im. N.E. Baumana. Ser. Priborostroenie. 2011. N S. P. 58–63.
14. Wang Y., Cunzhao S., Wang C., Xiao B. Ground-based cloud classification by learning stable local binary patterns // Atmos. Res. 2018. V. 207. P. 74–89. DOI: 10.1016/j.atmosres.2018.02.023.
15. OikoNmou S., Kazantzidis A., EcoNmou G., Fotopoulos S. A local binary pattern classification approach for cloud types derived from all-sky imagers // Int. J. Remote Sens. 2019. V. 40, N 7. P. 2667–2682. DOI: 10.1080/01431161.2018.1530807.
16. Laws K.I. Rapid texture identification // Proc. SPIE. 1980. V. 238. P. 376–380. DOI: 10.1117/12.959169.
17. Mallat S.G. A theory for multiresolution signal decomposition: The wavelet representation // IEEE Trans. Pattern Anal. Mach. Intell. 1989. V. 11, N 7. P. 674–693. DOI: 10.1109/34.192463.
18. Histograms of oriented gradients. URL: https://courses.cs.duke.edu/fall17/compsci527/Ntes/hog.pdf (last access: 05.12.2024).
19. Haralik R.M. Statisticheskii i strukturnyi podkhody k opisaniyu tekstur // TIIER. 1979. V. 67, N 5. P. 98–120.
20. Gonzalez R.C., Wood R.E. Digital Image Processing, 4th ed. London: Pearson, 2017. 1192 p.
21. Unser M. Texture classification and segmentation using wavelet frames // IEEE Trans. Image Process. 1995. V. 4, N 11. P. 1549–1560. DOI: 10.1109/83.469936.
22. Turner M.R. Texture discrimination by Gabor functions // Biol. Cyber. 1986. V. 55, N 2. P. 71–82. DOI: 10.1007/BF00341922.
23. Recio J.A.R., Fernandez L.A.R., Fernández-Sarriá A. Use of Gabor filters for texture classification of digital images // Física de la Tierra. 2005. V. 17. P. 47.
24. Shleimovich M.P., Lyasheva S.A., Kirpichnikov A.P. Vychislenie priznakov izobrazhenii na osnove veivlet-preobrazovaniya // Vestn. tekhnol. un-ta. 2015. V. 18, N 18. P. 223–228.
25. Fralenko V.P. Metody teksturnogo analiza izobrazhenii, obrabotka dannykh distantsionnogo zondirovaniya Zemli // Programmnye sistemy: teoriya i prilozheniya. 2014. V. 5, N 4. P. 19–39.
26. ResNet (34, 50, 101): «ostatochnye» CNN dlya klassifikatsii izobrazhenii. URL: https://neurohive.io/ru/vidy-nejrosetej/resnet-34-50-101 (data obrashcheniya: 15.03.2023).
27. VGG16 – neiroset' dlya vydeleniya priznakov izobrazhenii. URL: https://neurohive.io/ru/vidy-nejrosetej/vgg16-model/ (data obrashcheniya: 15.03.2023).
28. Liang J. Image classification based on RESNET // J. Phys. Conf. Ser. 2020. V. 1634, N 1. P. 012110. DOI: 10.1088/1742-6596/1634/1/012110.
29. Very Deep Convolutional Networks for Large-Scale Image Recognition. URL: https://doi.org/10.48550/arXiv.1409.1556 (last access: 15.03.2023).
30. Improved U-net remote sensing classification algorithm based on multi-feature fusion perception // Remote Sens. 2022. V. 14, N 5. P. 1118. DOI: 10.3990/ rs14051118.
31. Khromov S.P., Petrosyan M.A. Meteorologiya i klimatologiya: ucheb. posobie. M.: MGU, 2012. 584 p.
32. TJNU ground-based remote sensing cloud database (TJNU-GRSCD). URL: https://github.com/shuangliutjnu / TJNU - Ground - based - Remote - Sensing - Cloud - Database (last access: 21.03.2025).
33. Cirrus Cumulus Stratus Nimbus (CCSN) Database. URL: https: // data-verse.harard.edu / dataset.xhtml?persistentId=doi:10.7910/DVN/CADDPD (last access: 21.03.2025).
34. Galileiskii V.P., Elizarov A.I., Kokarev D.V., Morozov A.M. Baza dannykh izobrazhenii oblachnogo polya nad gorodom Tomsk // Svidetel'stvo o gosudarstvennoi registratsii bazy dannykh N 2018620430 ot 14.03.2018. Pravoobladatel': Federal'noe gosudarstvennoe byudzhetnoe uchrezhdenie nauki Institut optiki atmosfery im. V.E. Zueva Sibirskogo otdeleniya Rossiiskoi akademii nauk (IOA SO RAN) (RU).