{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,5]],"date-time":"2026-03-05T16:00:51Z","timestamp":1772726451830,"version":"3.50.1"},"reference-count":44,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2023,12,22]],"date-time":"2023-12-22T00:00:00Z","timestamp":1703203200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2022YFC2903404"],"award-info":[{"award-number":["2022YFC2903404"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>Despite demonstrating exceptional inversion production for synthetic data, the application of deep learning (DL) inversion methods to invert realistic magnetotelluric (MT) measurements, which are inevitably contaminated by noise in acquisition, poses a significant challenge. Hence, to facilitate DL inversion for realistic MT measurements, this work explores developing a noise-robust MT DL inversion method by generating targeted noisy training datasets and constructing a physics-informed neural network. Different from most previous works that only considered the noise of one fixed distribution and level, we propose three noise injection strategies and compare their combinations to mitigate the adverse effect of measurement noise on MT DL inversion results: (1) add synthetic relative noise obeying Gaussian distribution; (2) propose a multiwindow Savitzky\u2013Golay (MWSG) filtering scheme to extract potential and possible noise from the target field data and then introduce them into training data; (3) create an augmented training dataset based on the former two strategies. Moreover, we employ the powerful Swin Transformer as the backbone network to construct a U-shaped DL model (SwinTUNet), based on which a physics-informed SwinTUNet (PISwinTUNet) is implemented to further enhance its generalization ability. In synthetic examples, the proposed noise injection strategies demonstrate impressive inversion effects, regardless of whether they are contaminated by familiar or unfamiliar noise. In a field example, the combination of three strategies drives PISwinTUNet to produce considerably faithful reconstructions for subsurface resistivity structures and outperform the classical deterministic Occam inversions. The experimental results show that the proposed noise-robust DL inversion method based on the noise injection strategies and physics-informed DL architecture holds great promise in processing MT field data.<\/jats:p>","DOI":"10.3390\/rs16010062","type":"journal-article","created":{"date-parts":[[2023,12,24]],"date-time":"2023-12-24T20:48:37Z","timestamp":1703450917000},"page":"62","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":9,"title":["Physics-Informed Deep Learning Inversion with Application to Noisy Magnetotelluric Measurements"],"prefix":"10.3390","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1541-4153","authenticated-orcid":false,"given":"Wei","family":"Liu","sequence":"first","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education), Central South University, Changsha 410083, China"},{"name":"Hunan Key Laboratory of Nonferrous Resources and Geological Hazards Exploration, Central South University, Changsha 410083, China"},{"name":"School of Geosciences and Info-Physics, Central South University, Changsha 410083, China"}]},{"given":"He","family":"Wang","sequence":"additional","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education), Central South University, Changsha 410083, China"},{"name":"Hunan Key Laboratory of Nonferrous Resources and Geological Hazards Exploration, Central South University, Changsha 410083, China"},{"name":"School of Geosciences and Info-Physics, Central South University, Changsha 410083, China"}]},{"given":"Zhenzhu","family":"Xi","sequence":"additional","affiliation":[{"name":"Key Laboratory of Metallogenic Prediction of Nonferrous Metals and Geological Environment Monitoring (Ministry of Education), Central South University, Changsha 410083, China"},{"name":"Hunan Key Laboratory of Nonferrous Resources and Geological Hazards Exploration, Central South University, Changsha 410083, China"},{"name":"School of Geosciences and Info-Physics, Central South University, Changsha 410083, China"}]},{"given":"Liang","family":"Wang","sequence":"additional","affiliation":[{"name":"Hunan 5D Geosciences Co., Ltd., Changsha 410083, China"}]}],"member":"1968","published-online":{"date-parts":[[2023,12,22]]},"reference":[{"key":"ref_1","first-page":"295","article-title":"Determination of the electrical characteristics of the deep strata of the earth\u2019s crust","volume":"73","author":"Tikhonov","year":"1950","journal-title":"Dolk. Acad. Nauk. SSSR"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"605","DOI":"10.1190\/1.1437915","article-title":"Basic Theory of the Magnetotelluric Method of Geophysical Prospecting","volume":"18","author":"Cagniard","year":"1953","journal-title":"Geophysics"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1432","DOI":"10.1093\/gji\/ggt055","article-title":"Three-dimensional parallel distributed inversion of CSEM data using a direct forward solver","volume":"193","author":"Grayver","year":"2013","journal-title":"Geophys. J. Int."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"E401","DOI":"10.1190\/geo2015-0481.1","article-title":"3D inversion for multipulse airborne transient electromagnetic data","volume":"81","author":"Liu","year":"2016","journal-title":"Geophysics"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"289","DOI":"10.1190\/1.1442303","article-title":"Occam\u2019s inversion: A practical algorithm for generating smooth models from electromagnetic sounding data","volume":"52","author":"Constable","year":"1987","journal-title":"Geophysics"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"791","DOI":"10.1190\/1.1444778","article-title":"An efficient data-subspace inversion method for 2-D magnetotelluric data","volume":"65","author":"Siripunvaraporn","year":"2000","journal-title":"Geophysics"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"410","DOI":"10.1046\/j.1365-246x.2000.00007.x","article-title":"Three-dimensional magnetotelluric inversion using non-linear conjugate gradients","volume":"140","author":"Newman","year":"2002","journal-title":"Geophys. J. Int."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"365","DOI":"10.1111\/j.1365-246X.2008.03717.x","article-title":"Non-linear conjugate gradient inversion for global EM induction: Resolution studies","volume":"173","author":"Kelbert","year":"2008","journal-title":"Geophys. J. Int."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1751","DOI":"10.1093\/gji\/ggy071","article-title":"Efficient Hierarchical Trans-Dimensional Bayesian Inversion of Magnetotelluric Data","volume":"213","author":"Xiang","year":"2018","journal-title":"Geophys. J. Int."},{"key":"ref_10","first-page":"260","article-title":"Quantum genetic algorithm and its application in magnetotelluric data inversion","volume":"52","author":"Luo","year":"2009","journal-title":"Chin. J. Geophys."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"58","DOI":"10.1190\/tle37010058.1","article-title":"Deep-learning tomography","volume":"37","author":"Jennings","year":"2018","journal-title":"Lead. Edge"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"R869","DOI":"10.1190\/geo2018-0838.1","article-title":"Convolutional neural network for seismic impedance inversion","volume":"84","author":"Das","year":"2019","journal-title":"Geophysics"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Bai, P., Vignoli, G., Viezzoli, A., Nevalainen, J., and Vacca, G. (2020). (Quasi-) Real-Time Inversion of Airborne Time-Domain Electromagnetic Data via Artificial Neural Network. Remote Sens., 12.","DOI":"10.3390\/rs12203440"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1761","DOI":"10.1111\/1365-2478.13136","article-title":"Convolutional neural network inversion of airborne transient electromagnetic data","volume":"69","author":"Wu","year":"2021","journal-title":"Geophys. Prospect."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"817","DOI":"10.1093\/gji\/ggz204","article-title":"Deep learning electromagnetic inversion with convolutional neural networks","volume":"218","author":"Puzyrev","year":"2019","journal-title":"Geophys. J. Int."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"247","DOI":"10.1093\/gji\/ggaa161","article-title":"One-dimensional deep learning inversion of electromagnetic induction data using convolutional neural network","volume":"222","author":"Moghadas","year":"2020","journal-title":"Geophys. J. Int."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"105454","DOI":"10.1016\/j.cageo.2023.105454","article-title":"One-Dimensional Magnetotelluric Parallel Inversion Using a ResNet1D-8 Residual Neural Network","volume":"180","author":"Ling","year":"2023","journal-title":"Comput. Geosci."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"2482","DOI":"10.1007\/s11771-019-4188-2","article-title":"Nonlinear inversion for magnetotelluric sounding based on deep belief network","volume":"26","author":"Wang","year":"2019","journal-title":"J. Cent. South Univ."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"1047","DOI":"10.1007\/s11600-022-00773-z","article-title":"2D Inversion of Magnetotelluric Data Using Deep Learning Technology","volume":"70","author":"Liao","year":"2022","journal-title":"Acta Geophys."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"2001","DOI":"10.1093\/gji\/ggab074","article-title":"Robust Deep Learning Seismic Inversion with a Priori Initial Model Constraint","volume":"225","author":"Zhang","year":"2021","journal-title":"Geophys. J. Int."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"R87","DOI":"10.1190\/geo2019-0138.1","article-title":"A theory-guided deep-learning formulation and optimization of seismic waveform inversion","volume":"85","author":"Sun","year":"2020","journal-title":"Geophysics"},{"key":"ref_22","first-page":"5906611","article-title":"Physics-Driven Deep Learning Inversion for Direct Current Resistivity Survey Data","volume":"61","author":"Liu","year":"2023","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Liu, W., Wang, H., Xi, Z., Zhang, R., and Huang, X. (2022). Physics-Driven Deep Learning Inversion with Application to Magnetotelluric. Remote Sens., 14.","DOI":"10.3390\/rs14133218"},{"key":"ref_24","first-page":"86","article-title":"A Physics-Driven Deep-Learning Network for Solving Nonlinear Inverse Problems","volume":"61","author":"Jin","year":"2020","journal-title":"Petrophys.-SPWLA J. Form. Eval. Reserv. Descr."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1109\/21.155944","article-title":"Noise Injection into Inputs in Back-Propagation Learning","volume":"22","author":"Matsuoka","year":"1992","journal-title":"IEEE Trans. Syst. Man. Cyb."},{"key":"ref_26","unstructured":"Goodfellow, I., Bengio, Y., and Courville, A. (2023, July 28). Available online: https:\/\/www.deeplearningbook.org\/contents\/regularization.html."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"1540","DOI":"10.1109\/LGRS.2020.3005796","article-title":"Application of Sample-Compressed Neural Network and Adaptive-Clustering Algorithm for Magnetotelluric Inverse Modeling","volume":"18","author":"Liu","year":"2021","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., and Guo, B. (2021, January 10\u201317). Swin Transformer: Hierarchical Vision Transformer Using Shifted Windows. Proceedings of the 2021 IEEE\/CVF International Conference on Computer Vision (ICCV), Online.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Liu, Z., Hu, H., Lin, Y., Yao, Z., Xie, Z., Wei, Y., Ning, J., Cao, Y., Zhang, Z., and Dong, L. (2022, January 18\u201324). Swin Transformer V2: Scaling Up Capacity and Resolution. Proceedings of the 2022 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans, LA, USA.","DOI":"10.1109\/CVPR52688.2022.01170"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Bai, D., Lu, G., Zhu, Z., Zhu, X., Tao, C., Fang, J., and Li, Y. (2022). Prediction Interval Estimation of Landslide Displacement Using Bootstrap, Variational Mode Decomposition, and Long and Short-Term Time-Series Network. Remote Sens., 14.","DOI":"10.3390\/rs14225808"},{"key":"ref_31","unstructured":"Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, \u0141., and Polosukhin, I. (2017). Attention is all you need. arXiv."},{"key":"ref_32","unstructured":"Hendrycks, D., and Gimpel, K. (2016). Gaussian Error Linear Units (GELUs). arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Bai, D., Lu, G., Zhu, Z., Zhu, X., Tao, C., and Fang, J. (2022). Using Electrical Resistivity Tomography to Monitor the Evolution of Landslides\u2019 Safety Factors under Rainfall: A Feasibility Study Based on Numerical Simulation. Remote Sens., 14.","DOI":"10.3390\/rs14153592"},{"key":"ref_34","unstructured":"He, K., Zhang, X., Ren, S., and Sun, J. (July, January 26). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"509","DOI":"10.1007\/s12665-022-10637-w","article-title":"Using time series analysis and dual-stage attention-based recurrent neural network to predict landslide displacement","volume":"81","author":"Bai","year":"2022","journal-title":"Environ. Earth Sci."},{"key":"ref_36","unstructured":"Sheng, Z., Xie, S.Q., and Pan, C.Y. (2008). Probability Theory and Mathematical Statistic, Higher Education Press. [4th ed.]."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1627","DOI":"10.1021\/ac60214a047","article-title":"Smoothing and Differentiation of Data by Simplified Least Squares Procedures","volume":"36","author":"Savitzky","year":"1964","journal-title":"Anal. Chem."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"332","DOI":"10.1016\/j.rse.2004.03.014","article-title":"A Simple Method for Reconstructing a High-Quality NDVI Time-Series Data Set Based on the Savitzky\u2013Golay Filter","volume":"91","author":"Chen","year":"2004","journal-title":"Remote Sens. Environ."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1016\/j.dsp.2004.09.008","article-title":"Properties of Savitzky\u2013Golay Digital Differentiators","volume":"15","author":"Luo","year":"2005","journal-title":"Digit. Signal Process."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"570","DOI":"10.1021\/ac00205a007","article-title":"General Least-Squares Smoothing and Differentiation by the Convolution (Savitzky-Golay) Method","volume":"62","author":"Gorry","year":"1990","journal-title":"Anal. Chem."},{"key":"ref_41","first-page":"4505214","article-title":"Smooth Deep Learning Magnetotelluric Inversion Based on Physics-Informed Swin Transformer and Multiwindow Savitzky\u2013Golay Filter","volume":"61","author":"Liu","year":"2023","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_42","unstructured":"Kingma, D.P., and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"167","DOI":"10.1016\/j.cageo.2014.07.013","article-title":"MTpy: A Python Toolbox for Magnetotellurics","volume":"72","author":"Krieger","year":"2014","journal-title":"Comput. Geosci."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"457","DOI":"10.1111\/j.1365-246X.2004.02281.x","article-title":"The magnetotelluric phase tensor","volume":"158","author":"Caldwell","year":"2004","journal-title":"Geophys. J. Int."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/16\/1\/62\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T21:40:41Z","timestamp":1760132441000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/16\/1\/62"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,12,22]]},"references-count":44,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2024,1]]}},"alternative-id":["rs16010062"],"URL":"https:\/\/doi.org\/10.3390\/rs16010062","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,12,22]]}}}