Restoring the values of geo-fields using a combination of kernel smoothing methods and artificial neural networks models

Geographic Information Systems
UDC: 
52.08:004.62:681.518.3
DOI: 
10.22389/0016-7126-2022-990-12-57-64
1 Gvozdev O.G.
2 Materuhin A.V.
3 Maiorov A.A.
Year: 
2022
№: 
12
990
Pages: 
57-64

Moscow State University of Geodesy and Cartography (MIIGAiK)

1, 
2, 
3, 
Abstract:
The purpose of the study, the results of which are described in the article, was to improve solving the matter of the geo-fields’ values restoring based on processing high-intensity spatial-temporal data streams received from a highly mobile geo-sensors network. Previously, the authors proposed an original approach to solving this task, which means applying the kernel smoothing methods, the nuclear function for which is determined automatically, using discrete stochastic optimization, in particular, the annealing simulation method. The idea of a new approach proposed by the authors is as follows: to apply kernel smoothing methods using an automatic adaptive technique for determining the parameters of the function to build a preliminary geo-field model, and then one based on a neural network to refine it. That enables removing the main limitation of the applicability of the previous solution based on the assumption of all points’ local homogeneity. The method described in this article does not use this assumption and can be applied in situations where it is violated. The article contains a description of the computational experiments carried out as well as a discussion of the outcome obtained. The results of computational experiments showed a clear advantage of the new technology.
The results were obtained as a part of the state task of the Ministry of Science and Higher Education of the Russian Federation (No. FSFE-2022-0002)
References: 
1.   Bondarev I.N., Materukhin A.V., Gvozdev O.G. Ispol’zovanie kletochnykh avtomatov dlya imitatsionnogo modelirovaniya rasprostraneniya zagryazneniya atmosfernogo vozdukha v usloviyakh megapolisa. URL: https://clck.ru/ahW3D (accessed: 15.11.2021).
2.   Gvozdev O.G., Maiorov A.A., Materuhin A.V. (2021) Method of restoring the geofield values based on data from a highly mobile geosensors network using an automatic adaptive technique for determining the parameters of the local regression kernel. Geodezia i Kartografia, 82(12), pp. 23-33. (In Russian). DOI: 10.22389/0016-7126-2021-978-12-23-33.
3.   Savel'eva E.A., Dem'yanov V.V. Geostatistika: teoriya i praktika. Pod red. R.V. Arutyunyana. M.: Nauka, 2010, 327 p.
4.   Kopnov M. V., Markov N. G. Vosstanovlenie dvumernykh geopolei metodami geostatistiki. Problemy informatiki, 2011, no. 2, pp. 36–43.
5.   Kanevskii M. F., Dem'yanov V. V., Savel'eva E. A., Chernov S. Yu. Osnovnye ponyatiya i elementy geostatistiki. Problemy okruzhayushchei sredy i prirodnykh resursov, 1999, no. 11, pp. 15–21.
6.   Aliev K.-A., Sevastopolsky A., Kolos M., Ulyanov D., Lempitsky V. (2020) Neural Point-Based Graphics. Lecture Notes in Computer Science. Lecture Notes in Computer Science, LNIP, no. 12367, DOI: 10.1007/978-3-030-58542-6_42.
7.   Byon Y.-J., Ha J. S., Cho C.-S., Kim T.-Y., Yeun C. Y. (2017) Real-Time Transportation Mode Identification Using Artificial Neural Networks Enhanced with Mode Availability Layers: A Case Study in Dubai. Applied Sciences, Volume 7, no. 923, DOI: 10.3390/app7090923.
8.   Chen G., Li Y., Sun G., Zhang Y. (2017) Application of Deep Networks to Oil Spill Detection Using Polarimetric Synthetic Aperture Radar Images. Applied Sciences, Volume 7, no. 968, DOI: 10.3390/app7100968.
9.   Dtissibe F. Y., Ari A. A. A., Titouna C., Thiare O., Gueroui A. M. (2020) Flood forecasting based on an artificial neural network scheme. Natural Hazards, no. 104 (2), pp. 1211–1237. DOI: 10.1007/s11069-020-04211-5.
10.   Feng W., Li J., Cai H., Luo X., Zhang J. (2022) Neural Points: Point Cloud Representation with Neural Fields for Arbitrary Upsampling. Conference on Computer Vision and Pattern Recognition (New Orleans, USA). pp. 18612–18621. DOI: 10.1109/CVPR52688.2022.01808.
11.   Goel A., Goel A. K., Kumar A. (2022) The role of artificial neural network and machine learning in utilizing spatial information. Spatial Information Research, DOI: 10.1007/s41324-022-00494-x.
12.   Hwang J.-I., Chae S.-H., Kim D., Jung H.-S. (2017) Application of Artificial Neural Networks to Ship Detection from X-Band Kompsat-5 Imagery. Applied Sciences, Volume 7, no. 961, DOI: 10.3390/app7090961.
13.   Ibtehaz N., Rahman M. S. (2020) MultiResUNet: Rethinking the U-Net architecture for multimodal biomedical image segmentation. Neural Networks, Volume 121, pp. 74–87. DOI: 10.1016/j.neunet.2019.08.025.
14.   Kadavi P. R., Lee W.-J., Lee C.-W. (2017) Analysis of the Pyroclastic Flow Deposits of Mount Sinabung and Merapi Using Landsat Imagery and the Artificial Neural Networks Approach. Applied Sciences, Volume 7, no. 935, DOI: 10.3390/app7090935.
15.   Kia M. B., Pirasteh S., Pradhan B., Mahmud A. R., Sulaiman W. N. A., Moradi A. (2012) An artificial neural network model for flood simulation using GIS: Johor River Basin, Malaysia. Environmental Earth Sciences, no. 67, pp. 251–264. DOI: 10.1007/s12665-011-1504-z.
16.   Kingma D. P., Ba J. (2015) Adam: A Method for Stochastic Optimization. URL: arxiv.org/abs/1412.6980 (accessed: 10.10.2022). DOI: 10.48550/arXiv.1412.6980.
17.   Kwon S.-K., Jung H.-S., Baek W.-K., Kim D. (2017) Classification of Forest Vertical Structure in South Korea from Aerial Orthophoto and Lidar Data Using an Artificial Neural Network. Applied Sciences, no. 7, DOI: 10.3390/app7101046.
18.   Lu X., Rudi A., Borgonovo E., Rosasco L. (2019) Faster Kriging: Facing High-Dimensional Simulators. Operations Research, Volume 68, no. 1, DOI: 10.1287/opre.2019.1860.
19.   Zaheer M., Kottur S., Ravanbhakhsh S., Póczos B., Salakhutdinov R., Smola A. J. (2017) Deep Sets. 31st Conference on Neural Information Processing Systems. pp. 3394–3404.
20.   Materukhin A. V., Gvozdev O. G., Maiorov A. A., Sokolova O. D. (2019) Simulation of Spatio-Temporal Data Streams from Geosensors Located on Mobile Objects, 15th International Asian School-Seminar Optimization Problems of Complex Systems. pp. 179–182. DOI: 10.1109/OPCS.2019.8880188.
21.   Newell A., Yang K., Deng J. (2016) Stacked Hourglass Networks for Human Pose Estimation. Lecture Notes in Computer Science. DOI: 10.1007/978-3-319-46484-8_29.
22.   Oh H.-J., Lee S. (2017) Shallow Landslide Susceptibility Modeling Using the Data Mining Models Artificial Neural Network and Boosted Tree. Applied Sciences, no. 7, DOI: 10.3390/app7101000.
23.   Rombach R., Blattmann A., Lorenz D., Esser P., Ommer B. (2022) High-resolution image synthesis with latent diffusion models. Conference on Computer Vision and Pattern Recognition. pp. 10684–10695.
24.   Ronneberger O., Fischer P., Brox T. (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015. Lecture Notes in Computer Science. DOI: 10.1007/978-3-319-24574-4_28.
25.   Soelch M., Akhundov A., van der Smagt P., Bayer J. (2019) On Deep Set Learning and the Choice of Aggregations. Artificial Neural Networks and Machine Learning – ICANN 2019: Theoretical Neural Computation. Lecture Notes in Computer Science, no. 11727, DOI: 10.1007/978-3-030-30487-4_35.
26.   Srinivasan B. V., Duraiswami R., Murtugudde R. (2010) Efficient kriging for real-time spatiotemporal interpolation. Proceedings of the 20th Conference on Probability and Statistics in the Atmospheric Sciences. pp. 228–235.
27.   Tan Q., Huang Y., Hu J., Zhou P., Hu J. (2021) Application of artificial neural network model based on GIS in geological hazard zoning. Neural Computing and Applications, no. 33, pp. 591–602. DOI: 10.1007/s00521-020-04987-4.
28.   Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A. N., Kaiser L., Polosukhin I. (2017) Attention is all you need. Proceedings of the 31st International Conference on Neural Information Processing Systems (New York, USA). pp. 6000–6010.
29.   Vigsnes M., Kolbjørnsen O., Hauge V. L. (2017) Fast and Accurate Approximation to Kriging Using Common Data Neighborhoods. Mathematical Geosciences, no. 49, pp. 619–634.
30.   Wang Yu-An, Chen Yun-Nung (2020) What Do Position Embeddings Learn? An Empirical Study of Pre-Trained Language Model Positional Encoding. Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP). pp. 6840–6849. DOI: 10.18653/v1/2020.emnlp-main.555.
Citation:
Gvozdev O.G., 
Materuhin A.V., 
Maiorov A.A., 
(2022) Restoring the values of geo-fields using a combination of kernel smoothing methods and artificial neural networks models. Geodesy and cartography = Geodezia i Kartografia, 83(12), pp. 57-64. (In Russian). DOI: 10.22389/0016-7126-2022-990-12-57-64
Publication History
Received: 01.12.2022
Accepted: 05.12.2022
Published: 20.01.2023

Content

2022 December DOI:
10.22389/0016-7126-2022-990-12