Development of a neural network model for spatial data analysis

Objectives. The paper aimed to develop and validate a neural network model for spatial data analysis. The advantage of the proposed model is the presence of a large number of degrees of freedom allowing its flexible configuration depending on the specific problem. This development is part of the knowledge base of a deep machine learning model repository including a dynamic visualization subsystem based on adaptive web interfaces allowing interactive direct editing of the architecture and topology of neural network models.

Methods. The presented solution to the problem of improving the accuracy of spatial data analysis and classification is based on a geosystem approach for analyzing the genetic homogeneity of territorial-adjacent entities of different scales and hierarchies. The publicly available EuroSAT dataset used for initial validation of the proposed methodology is based on Sentinel-2 satellite imagery for training and testing machine learning models aimed at classifying land use/land cover systems. The ontological model of the repository including the developed model is decomposed into domains of deep machine learning models, project tasks and data, thus providing a comprehensive definition of the formalizing area of knowledge. Each stored neural network model is mapped to a set of specific tasks and datasets. Results. Model validation for the EuroSAT dataset algorithmically extended in terms of the geosystem approach allows classification accuracy to be improved under training data shortage within 9% while maintaining the accuracy of ResNet50 and GoogleNet deep learning models.

Conclusions. The implemention of the developed model into the repository enhances the knowledge base of models for spatial data analysis as well as allowing the selection of efficient models for solving problems in the digital economy.

1. Saleh H., Alexandrov D., Dzhonov A. Uberisation business model based on blockchain for implementation decentralized application for lease/rent lodging. In: Rocha A., Serrhini M., (Eds.). Information Systems and Technologies to Support Learning (EMENA-ISTL 2018). Smart Innovation, Systems and Technologies. International Conference Europe Middle East & North Africa. Springer, Cham. 2018;111:225-232. https://doi.org/10.1007/978-3-030-03577-8_26

2. Sigov A.S., Tsvetkov V.Ya., Rogov I.E. Methods for assessing testing difficulty in education sphere. Russ. Technol. J. 2021;9(6):64-72 (in Russ.). https://doi.org/10.32362/2500-316X-2021-9-6-64-72

3. Liu Y., Sangineto E., Bi W., Sebe N., Lepri B., Nadai M. Efficient training of visual transformers with small datasets. Advances in Neural Information Processing Systems. 2021;34:23818-23830. Available from URL: https://arxiv.org/pdf/2106.03746.pdf

4. Zanozin V.V., Karabaeva A.Z., Koneeva A.V., Makeeva E.V., Molokova V.G. Features of the horizontal structure of the central part of the Volga delta landscape. In. Geographic Sciences and Education: Proceedings of the XI All-Russian Conf. 2018. P. 161-163 (in Russ.).

5. Yamashkina E.O., Kovalenko S.M., Platonova O.V. Development of repository of deep neural networks for the analysis of geospatial data. IOP Conf. Ser.: Mater. Sci. Eng. 2021;1047(1).012124. https.//doi.org/10.1088/1757-899X/1047/1/012124

6. Weiss M., Jacob F., Duveiller G. Remote sensing for agricultural applications. A meta-review. Remote Sens. Environ. 2020;236(5).111402. https.//doi.org/10.1016/j.rse.2019.111402

7. Yamashkin S.A., Yamashkin A.A. Improving the efficiency of remote sensing data interpretation by analyzing neighborhood descriptors. Inzhenernye tekhnologii i sistemy = Engineering Technologies and Systems (Vestnik Mordovskogo universiteta = Mordovia University Bulletin). 2018; 28(3).352-365 (in Russ.). https:///doi.org/10.15507/0236-2910.028.201803.352-365

8. Ioffe S., Szegedy Ch. Batch Normalization: accelerating deep network training by reducing internal covariate shift. Preprint. March 2, 2015. Available from URL. https.//arxiv.org/abs/1502.03167

9. Yao Z., Cao Y., Zheng S., Huang G., Lin S. Cross-iteration Batch Normalization. In. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021.12331-12340. https.//doi.org/10.1109/CVPR46437.2021.01215

10. Jung W., Jung D., Kim B., Lee S., Rhee W., Ahn J.H. Restructuring Batch Normalization to accelerate CNN training. In. Proceedings of Machine Learning and Systems. 2019;1.14-26. Available from URL. https://mlsys.org/Conferences/2019/doc/2019/18.pdf

11. Chen Y., Dai X., Liu M., Chen D., Yuan L., Liu Z. Dynamic ReLU. In. Vedaldi A., Bischof H., Brox T., Frahm J.M. (Eds.). Computer Vision - ECCV 2020. ECCV 2020. Lecture Notes in Computer Science. Cham. Springer; 2020. V. 12364. P. 351-367. https.//doi.org/10.1007/978-3-030-58529-7_21

12. Gu J., et al. Recent advances in convolutional neural networks. Pattern Recognition. 2018;77.354-377. https://doi.org/10.1016/j.patcog.2017.10.013

13. Kozaev A., Saleh H., Alexandrov D. Simulation of emergency situations on main gas pipeline with MATLAB Simulink. In. 2019 Actual Problems of Systems and Software Engineering (APSSE). IEEE. 2019.63-68. https.//doi.org/10.1109/APSSE47353.2019.00015

14. Helber P., Bischke B., DengelA., Borth D. Introducing Eurosat. A novel dataset and deep learning benchmark for land use and land cover classification. In. IEEE International Symposium on Geoscience and Remote Sensing (IGARSS). 2018. 204-207. https.//doi.org/10.1109/IGARSS.2018.8519248

15. Phiri D., Simwanda M., Salekin S., Nyirenda V.R., Murayama Y., Ranagalage M. Sentinel-2 data for land cover/use mapping. a review. Remote Sens. 2020;12(14).2291. https.//doi.org/10.3390/rs12142291

16. Helber P., Bischke B., Dengel A., Borth D. EuroSAT. A novel dataset and deep learning benchmark for land use and land cover classification. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2019;12(7). 2217-2226. https.//doi.org/10.1109/JSTARS.2019.2918242

17. Szegedy Ch., Ioffe S., Vanhoucke V., Alemi A. Inception-v4, Inception-ResNet and the impact of residual connections on learning. In: Thirty-First AAAI Conference on Artificial Intelligence. 2017;31(1). https://doi.org/10.1609/aaai.v31i1.11231

18. Yamashkin S.A., Radovanovic M.M., Yamashkin A.A., Barmin A.N., Zanozin V.V., Petrovic M.D. Problems of designing geoportal interfaces. GeoJournal of Tourism and Geosites. 2019;24(1):88-101. https://doi.org/10.30892/gtg.24108-345

19. Soni A., Ranga V. API features individualizing of web services: REST and SOAP. Int. J. Innovative Technol. Exploring Eng. 2019;8(9S):664-671. https://doi.org/10.35940/ijitee.I1107.0789S19

20. Szegedy Ch., et al. Going deeper with convolutions. In: 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2015;1-9. https://doi.org/10.1109/CVPR.2015.7298594