СОВРЕМЕННЫЕ МЕТОДЫ ОБРАБОТКИ ГИПЕРСПЕКТРАЛЬНЫХ ИЗОБРАЖЕНИЙ: СИСТЕМНЫЙ АНАЛИЗ, АЛГОРИТМЫ И ПЕРСПЕКТИВЫ ПРИМЕНЕНИЯ В СТРОИТЕЛЬНОЙ ДИАГНОСТИКЕ
Аннотация
Актуальность исследования обусловлена ростом интереса к гиперспектральным изображениям как инструменту неразрушающего контроля строительных материалов и конструкций, а также недостаточной систематизацией современных методов их обработки. Цель работы состоит в системном анализе алгоритмов обработки HSI, выявлении их преимуществ, ограничений и перспектив применения в задачах строительной диагностики. В исследовании рассмотрены особенности гиперспектральных данных, включая высокую размерность, шумы, калибровочные погрешности, атмосферные искажения и дефицит размеченных выборок. Показана эволюция подходов от классических методов машинного обучения и ручного конструирования признаков к глубоким нейронным сетям. Проанализированы методы снижения размерности, классификаторы kNN, байесовские модели, логистическая регрессия, Random Forest, SVM и MLP, а также способы учета спектрально-пространственного контекста. Особое внимание уделено современным архитектурам глубокого обучения: 1D-, 2D- и 3D-CNN, RNN, LSTM/GRU, гибридным CNN–RNN-моделям, трансформерам и CNN–Transformer-схемам. Отдельно рассмотрены transfer learning, semi-supervised learning, self-supervised learning, few-shot learning, meta-learning и domain adaptation как способы преодоления ограниченности разметки. Выполнено сопоставление подходов по требованиям к данным, вычислительной сложности, устойчивости к шумам, способности учитывать спектральные и пространственные зависимости. Показано, что наиболее перспективными для строительной диагностики являются гибридные модели, сочетающие локальные признаки сверток, глобальный контекст внимания и возможность дообучения на малых специализированных выборках. Работа обобщает современное состояние области и формирует основу для выбора методов обнаружения дефектов, оценки влажности, коррозии и деградации строительных конструкций. Сформулированные выводы могут использоваться при проектировании экспериментальных протоколов и выборе архитектур для последующих прикладных исследований в области мониторинга зданий.
Список литературы
1. Goetz A.F.H., Vane G., Solomon J.E., Rock B.N. Imaging Spectrometry for Earth Remote Sensing, Sci-ence, 1985, Vol. 228, No. 4704, pp. 1147-1153. DOI: 10.1126/science.228.4704.1147.
2. Ahmad M., Distefano S., Khan A.M., Mazzara M., Li C., Li H., Aryal J., Ding Y., Vivone G., Hong D. A comprehensive survey for hyperspectral image classification: The evolution from conventional to transformers and Mamba models, Neurocomputing, 2024, Vol. 644, Art. 130428. DOI: 10.1016/j.neucom.2025.130428.
3. Hong D., Han Z., Yao J., Gao L., Zhang B., Plaza A., Chanussot J. SpectralFormer: Rethinking Hyper-spectral Image Classification with Transformers, IEEE Transactions on Geoscience and Remote Sens-ing, 2022, Vol. 60, pp. 1-15, Art. 5518615. DOI: 10.1109/TGRS.2021.3130716.
4. Li S., Strauss A., Grba D., Granzner M., Täubling-Fruleux B., Zimmermann T. Hyperspectral Signa-tures for Detecting the Concrete Hydration Process Using Neural Networks, Infrastructures, 2025, Vol. 10, No. 7, Art. 172. DOI: 10.3390/infrastructures10070172.
5. Cheng M.F., Mukundan A., Karmakar R., Valappil M.A.E., Jouhar J., Wang H.-C. Modern Trends and Recent Applications of Hyperspectral Imaging: A Review, Technologies, 2025, Vol. 13, Art. 170. DOI: 10.3390/technologies13050170.
6. Grill J.B., Strub F., Altché F., Tallec C., Richemond P.H., Buchatskaya E., Doersch C., Avila Pires B., Guo Z.D., Gheshlaghi Azar M., Piot B., Kavukcuoglu K., Munos R., Valko M. Bootstrap your own la-tent: A new approach to self-supervised learning, arXiv, 2020. DOI: 10.48550/arXiv.2006.07733.
7. Bioucas-Dias J.M., Plaza A., Camps-Valls G., Scheunders P., Nasrabadi N.M., Chanussot J. Hyper-spectral Remote Sensing Data Analysis and Future Challenges, IEEE Geoscience and Remote Sensing Magazine, 2013, Vol. 1, No. 2, pp. 6-36. DOI: 10.1109/MGRS.2013.2244672.
8. Karantzalos K., Karakizi C., Kandylakis Z., Antoniou G. HyRANK Hyperspectral Satellite Dataset I [Dataset]. Zenodo, 2018. DOI: 10.5281/zenodo.1222202.
9. Li H., Cui J., Zhang X., Han Y., Cao L. Dimensionality Reduction and Classification of Hyperspectral Remote Sensing Image Feature Extraction, Remote Sensing, 2022, Vol. 14, No. 18, Art. 4579. DOI: 10.3390/rs14184579.
10. Roweis S.T., Saul L. K. Nonlinear Dimensionality Reduction by Locally Linear Embedding, Science, 2000, Vol. 290, No. 5500, pp. 2323-2326. DOI: 10.1126/science.290.5500.2323.
11. McInnes L., Healy J., Saul N., Großberger L. UMAP: Uniform Manifold Approximation and Projec-tion, Journal of Open Source Software, 2018, Vol. 3, No. 29, Art. 861. DOI: 10.21105/joss.00861.
12. Chen S., Hu Y., Xu S., Li L., Cheng Y. Classification of hyperspectral remote sensing imagery by k-nearest-neighbor simplex based on adaptive C-mutual proportion standard deviation metric, Journal of Applied Remote Sensing, 2014, Vol. 8, No. 1, Art. 083578. DOI: 10.1117/1.JRS.8.083578.
13. Cao F., Yang Z., Ren J., Ling W.-K., Zhao H., Marshall S. Extreme Sparse Multinomial Logistic Re-gression: A Fast and Robust Framework for Hyperspectral Image Classification, Remote Sensing, 2017, Vol. 9, No. 12, Art. 1255. DOI: 10.3390/rs9121255.
14. Goel P.K., Prasher S.O., Patel R.M., Landry J.-A., Bonnell R.B., Viau A.A. Classification of hyperspec-tral data by decision trees and artificial neural networks to identify weed stress and nitrogen status of corn, Computers and Electronics in Agriculture, 2003, Vol. 39, No. 2, pp. 67-93. DOI: 10.1016/S0168-1699(03)00020-6.
15. Ham J., Chen Y., Crawford M.M., Ghosh J. Investigation of the Random Forest Framework for Classi-fication of Hyperspectral Data, IEEE Transactions on Geoscience and Remote Sensing, 2005, Vol. 43, No. 3, pp. 492-501. DOI: 10.1109/TGRS.2004.842481.
16. Mountrakis G., Im J., Ogole C. Support vector machines in remote sensing: A review, ISPRS Journal of Photogrammetry and Remote Sensing, 2011, Vol. 66, No. 3, pp. 247-259. DOI: 10.1016/j.isprsjprs.2010.11.001.
17. Melgani F., Bruzzone L. Classification of Hyperspectral Remote Sensing Images with Support Vector Machines, IEEE Transactions on Geoscience and Remote Sensing, 2004, Vol. 42, No. 8, pp. 1778-1790. DOI: 10.1109/TGRS.2004.831865.
18. Subramanian S., Ramakrishnan S. Methodology for hyperspectral image classification using novel neu-ral network, Proc. SPIE, 1997, Vol. 3071: Algorithms for Multispectral and Hyperspectral Imagery III. DOI: 10.1117/12.280589.
19. Plaza J., Plaza A., Perez R., Martinez P. On the use of small training sets for neural network-based characterization of mixed pixels in remotely sensed hyperspectral images, Pattern Recognition, 2009, Vol. 42, pp. 3032-3045. DOI: 10.1016/j.patcog.2009.04.008.
20. Merenyi E., Farrand W. H., Taranik J.V., Minor T.B. Classification of hyperspectral imagery with neu-ral networks: comparison to conventional tools, EURASIP Journal on Advances in Signal Processing, 2014, Art. 71. DOI: 10.1186/1687-6180-2014-71.
21. Wang L., Shi C., Diao C., Ji W., Yin D. A survey of methods incorporating spatial information in image classification and spectral unmixing, International Journal of Remote Sensing, 2016, Vol. 37,
No. 16, pp. 3870-3910. DOI: 10.1080/01431161.2016.1204032.
22. Benediktsson J.A., Palmason J.A., Sveinsson J.R. Classification of hyperspectral data from urban areas based on extended morphological profiles, IEEE Transactions on Geoscience and Remote Sensing, 2005, Vol. 43, No. 3, pp. 480-491. DOI: 10.1109/TGRS.2004.842478.
23. Li J., Bioucas-Dias J.M., Plaza A. Spectral-Spatial Hyperspectral Image Segmentation Using Subspace Multinomial Logistic Regression and Markov Random Fields, IEEE Transactions on Geoscience and Remote Sensing, 2012, Vol. 50, No. 3, pp. 809-823. DOI: 10.1109/TGRS.2011.2162649.
24. Fauvel M., Benediktsson J.A., Chanussot J., Sveinsson J.R. Spectral and Spatial Classification of Hyper-spectral Data Using SVMs and Morphological Profiles, IEEE Transactions on Geoscience and Remote Sensing, 2008, Vol. 46, No. 11, pp. 3804-3814. DOI: 10.1109/TGRS.2008.922034.
25. Green A.A., Berman M., Switzer P., Craig M.D. A transformation for ordering multispectral data in terms of image quality with implications for noise removal, IEEE Transactions on Geoscience and Re-mote Sensing, 1988, Vol. 26, No. 1, pp. 65-74. DOI: 10.1109/36.3001.
26. Cheriyadat A., Bruce L.M. Why principal component analysis is not an appropriate feature extraction method for hyperspectral data, Proc. IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2003, Vol. 6, pp. 3420-3422. DOI: 10.1109/IGARSS.2003.1294808.
27. Camps-Valls G., Tuia D., Bruzzone L., Benediktsson J.A. Advances in hyperspectral image classifica-tion: Earth monitoring with statistical learning methods, IEEE Signal Processing Magazine, 2014, Vol. 31, No. 1, pp. 45-54. DOI: 10.1109/MSP.2013.2279179.
28. Rasti B., Hong D., Hang R., Ghamisi P., Kang X., Chanussot J., Benediktsson J.A. Feature Extraction for Hyperspectral Imagery: The Evolution from Shallow to Deep: Overview and Toolbox, IEEE Geosci-ence and Remote Sensing Magazine, 2020, Vol. 8, No. 4, pp. 60-88. DOI: 10.1109/MGRS.2020.2979764.
29. Zhou J., Zeng S., Xiao Z., Zhou J., Li H., Kang Z. An Enhanced Spectral Fusion 3D CNN Model for Hyperspectral Image Classification, Remote Sensing, 2022, Vol. 14, No. 21, Art. 5334. DOI: 10.3390/rs14215334.
30. Hu W., Huang Y., Wei L., Zhang F., Li H. Deep Convolutional Neural Networks for Hyperspectral Image Classification, Journal of Sensors, 2015, Vol. 2015, Art. 258619. DOI: 10.1155/2015/258619.
31. Yue J., Zhao W., Mao S., Liu H. Spectral-spatial classification of hyperspectral images using deep con-volutional neural networks, Remote Sensing Letters, 2015, Vol. 6, No. 6, pp. 468-477. DOI: 10.1080/2150704X.2015.1047045.
32. Mou L., Ghamisi P., Zhu X.X. Deep Recurrent Neural Networks for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sensing, 2017, Vol. 55, No. 7, pp. 3639-3655. DOI: 10.1109/TGRS.2016.2636241.
33. Hochreiter S., Schmidhuber J. Long Short-Term Memory, Neural Computation, 1997, Vol. 9, No. 8, pp. 1735-1780. DOI: 10.1162/neco.1997.9.8.1735.
34. Liu Q., Zhou F., Hang R., Yuan X. Bidirectional-Convolutional LSTM Based Spectral-Spatial Feature Learning for Hyperspectral Image Classification, Remote Sensing, 2017, Vol. 9, No. 12, Art. 1330. DOI: 10.3390/rs9121330.
35. Zhou F., Hang R., Liu Q., Yuan X. Hyperspectral Image Classification Using Spectral-Spatial LSTMs, Neurocomputing, 2019, Vol. 328, pp. 39-47. DOI: 10.1016/j.neucom.2018.02.105.
36. Wu H., Prasad S. Convolutional Recurrent Neural Networks for Hyperspectral Data Classification, Remote Sensing, 2017, Vol. 9, No. 3, Art. 298. DOI: 10.3390/rs9030298.
37. Hu W.S., Li H.C., Pan L., Li W., Tao R., Du Q. Spatial-Spectral Feature Extraction via Deep Con-vLSTM Neural Networks for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sensing, 2020, Vol. 58, No. 6, pp. 4237-4250. DOI: 10.1109/TGRS.2019.2961947.
38. Qi W., Zhang X., Wang N., Zhang M., CenY. A Spectral-Spatial Cascaded 3D Convolutional Neural Network with a Convolutional Long Short-Term Memory Network for Hyperspectral Image Classifica-tion, Remote Sensing, 2019, Vol. 11, No. 20, Art. 2363. DOI: 10.3390/rs11202363.
39. Yin J., Qi C., Chen Q., Qu J. Spatial-Spectral Network for Hyperspectral Image Classification: A 3-D CNN and Bi-LSTM Framework, Remote Sensing, 2021, Vol. 13, No. 12, Art. 2353. DOI: 10.3390/rs13122353.
40. Farooque G., Xiao L., Yang J., Sargano A.B. Hyperspectral Image Classification via a Novel Spectral–Spatial 3D ConvLSTM-CNN, Remote Sensing, 2021, Vol. 13, No. 21, Art. 4348. DOI: 10.3390/rs13214348.
41. Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A.N., Kaiser Ł., Polosukhin I. At-tention Is All You Need, arXiv, 2017. DOI: 10.48550/arXiv.1706.03762.
42. Dosovitskiy A., Beyer L., Kolesnikov A., Weissenborn D., Zhai X., Unterthiner T., Dehghani M., Mind-erer M., Heigold G., Gelly S., Uszkoreit J., Houlsby N. An Image is Worth 16×16 Words: Transformers for Image Recognition at Scale, arXi, 2020. DOI: 10.48550/arXiv.2010.11929.
43. He J., Zhao L., Yang H., Zhang M., Li W. HSI-BERT: Hyperspectral Image Classification Using the Bidirectional Encoder Representation from Transformers, IEEE Transactions on Geoscience and Re-mote Sensing, 2020, No. 58 (1), pp. 165-178. DOI: 10.1109/TGRS.2019.2934760.
44. Yu H., Xu Z., Zheng K., Hong D., Yang H., Song M. MSTNet: A Multilevel Spectral-Spatial Transform-er Network for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sensing, 2022, No. 60, pp. 1-13. DOI: 10.1109/TGRS.2022.3186400.
45. Yang X., Cao W., Lu Y., Zhou Y. Hyperspectral Image Transformer Classification Networks, IEEE Transactions on Geoscience and Remote Sensing, 2022, Vol. 60, pp. 1-15. DOI: 10.1109/TGRS.2022.3171551.
46. Yuan D., Yu D., Qian Y., Xu Y., Liu Y. S2Former: Parallel Spectral-Spatial Transformer for Hyperspec-tral Image Classification, Electronics, 2023, Vol. 12, No. 18, Art. 3937. DOI: 10.3390/electronics12183937.
47. Wang D., Zhang J., Du B., Zhang L., Tao D. DCN-T: Dual Context Network with Transformer for Hyperspectral Image Classification, IEEE Transactions on Image Processing, 2023, No. 32, pp. 2536-2551. DOI: 10.1109/TIP.2023.3270104.
48. Zhang B., Chen Y., Rong Y., Xiong S., Lu X. MATNet: A Combining Multi-Attention and Transformer Network for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sens-ing, 2023, No. 61, pp. 1-15. DOI: 10.1109/TGRS.2023.3254523.
49. Sun L., Zhao G., Zheng Y., Wu Z. Spectral-Spatial Feature Tokenization Transformer for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sensing, 2022, No. 60, pp. 1-14. DOI: 10.1109/TGRS.2022.3144158.
50. Xu R., Dong X.-M., Li W., Peng J., Sun W., Xu Y. DBCTNet: Double-Branch Convolution–Transformer Network for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sens-ing, 2024, No. 62, pp. 1-15. DOI: 10.1109/TGRS.2024.3368141.
51. Qi W., Huang C., Wang Y., Zhang X., Sun W., Zhang L. Global-Local Three-Dimensional Convolution-al Transformer Network for Hyperspectral Image Classification, IEEE Transactions on Geoscience and Remote Sensing, 2023, No. 61, pp. 1-20. DOI: 10.1109/TGRS.2023.3272885.
52. Wang X.Y., Hu Q., Cheng Y.S., Ma J.Y. Hyperspectral Image Super-Resolution Meets Deep Learning: A Survey and Perspective, IEEE/CAA Journal of Automatica Sinica, 2023, Vol. 10, No. 8, pp. 1668-1691. DOI: 10.1109/JAS.2023.123681.
53. He X., Chen Y., Lin Z. Spatial-Spectral Transformer for Hyperspectral Image Classification, Remote Sensing, 2021, Vol. 13, No. 3, Art. 498. DOI: 10.3390/rs13030498.
54. Wang W., Liu L., Zhang T., Shen J., Wang J., Li J. Hyper-ES2T: Efficient Spatial–Spectral Transformer for the Classification of Hyperspectral Remote Sensing Images, International Journal of Applied Earth Observation and Geoinformation, 2022, Vol. 113, Art. 103005. DOI: 10.1016/j.jag.2022.103005.
55. Ji Y., Shi J., Zhang Y., Yang H., Zong Y., Xu L. HyFormer: Hybrid Grouping-Aggregation Transformer and Wide-Spanning CNN for Hyperspectral Image Super-Resolution, Remote Sensing, 2023, Vol. 15, No. 17, Art. 4131. DOI: 10.3390/rs15174131.
56. Arya S., Dubey S.R., Moorthi S.M., Dhar D., Singh S.K. 3D-CmT: 3D-CNN Meets Transformer for Hyperspectral Image Classification, In: Cho M.; Laptev I.; Tran D.; Yao A.; Zha H. B. (eds), Computer Vision – ACCV 2024 Workshops: Lecture Notes in Computer Science, Vol. 15483. Singapore: Springer, 2025. DOI: 10.1007/978-981-96-2644-1_23.
57. Marmanis D., Datcu M., Esch T., Stilla U. Deep Learning Earth Observation Classification Using ImageNet Pretrained Networks, IEEE Geoscience and Remote Sensing Letters, 2016, Vol. 13, No. 1, pp. 105-109. DOI: 10.1109/LGRS.2015.2499239.
58. Zhang H., Li Y., Jiang Y., Wang P., Shen Q., Shen C. Hyperspectral Classification Based on Lightweight 3-D-CNN with Transfer Learning, IEEE Transactions on Geoscience and Remote Sensing, 2019, Vol. 57, No. 8, pp. 5813-5828. DOI: 10.1109/TGRS.2019.2902568.
59. Yang J., Zhao Y., Chan J. C.-W. Learning and Transferring Deep Joint Spectral–Spatial Features for Hyperspectral Classification, IEEE Transactions on Geoscience and Remote Sensing, 2017, Vol. 55, No. 8, pp. 4729-4742. DOI: 10.1109/TGRS.2017.2698503.
60. Wang H., Cheng Y., Wang X. A Novel Hyperspectral Image Classification Method Using Class-Weighted Domain Adaptation Network, Remote Sensing, 2023, Vol. 15, No. 4, Art. 999. DOI: 10.3390/rs15040999.
61. Penatti O.A.B., Nogueira K., Santos J.A. dos. Do Deep Features Generalize from Everyday Objects to Remote Sensing and Aerial Scenes Domains?, 2015 IEEE Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW), 2015, pp. 44-51. DOI: 10.1109/CVPRW.2015.7301382.
62. Chen T., Kornblith S., Norouzi M., Hinton G. A Simple Framework for Contrastive Learning of Visual Representations, arXiv, 2020. arXiv:2002.05709. DOI: 10.48550/arXiv.2002.05709.
63. Xu J., Tang H., Ren Y., Peng L., Zhu X., He L. Multi-level Feature Learning for Contrastive Multi-view Clustering, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 16030-16039. DOI: 10.1109/CVPR52688.2022.01558.
64. Wang Y.; Albrecht C.M.; Ait Ali Braham N.; Mou L.; Zhu X.X. Self-Supervised Learning in Remote Sensing: A Review, IEEE Geoscience and Remote Sensing Magazine, 2022, Vol. 10, No. 4, pp. 213-247. DOI: 10.1109/MGRS.2022.3198244.
65. Ayuba D. L.; Guillemaut J.-Y.; Marti-Cardona B.; Mendez Maldonado O. SpecBPP: A Self-Supervised Learning Approach for Hyperspectral Representation and Soil Organic Carbon Estimation, arXiv, 2025. arXiv:2507.19781. DOI: 10.48550/arXiv.2507.19781.
66. Liu B., Yu X., Zhang P. A Semi-Supervised Convolutional Neural Network for Hyperspectral Image Classification, Remote Sensing Letters, 2017, Vol. 8, No. 9, pp. 839-848. DOI: 10.1080/2150704X.2017.1331053.
67. Wang L., Hao S., Wang Q., Wang Y. Semi-Supervised Classification for Hyperspectral Imagery Based on Spatial-Spectral Label Propagation, ISPRS Journal of Photogrammetry and Remote Sensing, 2014, Vol. 97, pp. 123-137. DOI: 10.1016/j.isprsjprs.2014.08.016.
68. Zhang D., Ren Y., Liu C., Han Z., Wang J. Weighted Contrastive Prototype Network for Few-Shot Hyperspectral Image Classification with Noisy Labels, Remote Sensing, 2024, Vol. 16, No. 18, Art. 3527. DOI: 10.3390/rs16183527.
69. Ait Ali Braham N., Wang Y., Albrecht C.M., Mou L., Zhu X.X. Self-Supervised Learning for Few-Shot Hyperspectral Image Classification, arXiv, 2022. arXiv:2206.12117. DOI: 10.48550/arXiv.2206.12117.
70. Kumar V., Singh R.S., Rambabu M., Dua Y. Deep learning for hyperspectral image classification:
A survey, Computer Science Review, 2024, Art. 100658. DOI: 10.1016/j.cosrev.2024.100658.
