Application of Machine Learning to Classification of Volcanic Deformation in Routinely Generated InSAR Data

Recent improvements in the frequency, type, and availability of satellite images mean it is now feasible to routinely study volcanoes in remote and inaccessible regions, including those with no ground-based monitoring. In particular, Interferometric Synthetic Aperture Radar data can detect surface deformation, which has a strong statistical link to eruption. However, the data set produced by the recently launched Sentinel-1 satellite is too large to be manually analyzed on a global basis. In this study, we systematically process >30,000 short-term interferograms at over 900 volcanoes and apply machine learning algorithms to automatically detect volcanic ground deformation. We use a convolutional neutral network to classify interferometric fringes in wrapped interferograms with no atmospheric corrections. We employ a transfer learning strategy and test a range of pretrained networks, finding that AlexNet is best suited to this task. The positive results are checked by an expert and fed back for model updating. Following training with a combination of both positive and negative examples, this method reduced the number of interferograms to ∼100 which required further inspection, of which at least 39 are considered true positives. We demonstrate that machine learning can efficiently detect large, rapid deformation signals in wrapped interferograms, but further development is required to detect slow or small deformation patterns which do not generate multiple fringes in short duration interferograms. This study is the first to use machine learning approaches for detecting volcanic deformation in large data sets and demonstrates the potential of such techniques for developing alert systems based on satellite imagery.

[1]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  M. Simons,et al.  An InSAR‐based survey of volcanic deformation in the southern Andes , 2004 .

[3]  Xue-wen Chen,et al.  Big Data Deep Learning: Challenges and Perspectives , 2014, IEEE Access.

[4]  Paul R. Bierman,et al.  A Cosmogenic view of erosion, relief generation, and the age of faulting in southern Africa , 2014 .

[5]  Nantheera Anantrasirichai,et al.  SVM-based texture classification in Optical Coherence Tomography , 2013, 2013 IEEE 10th International Symposium on Biomedical Imaging.

[6]  Hojjat Adeli,et al.  A probabilistic neural network for earthquake magnitude prediction , 2009, Neural Networks.

[7]  T. Wright,et al.  Statistical comparison of InSAR tropospheric correction techniques , 2015 .

[8]  Vir V. Phoha,et al.  K-Means+ID3: A Novel Method for Supervised Anomaly Detection by Cascading K-Means Clustering and ID3 Decision Tree Learning Methods , 2007, IEEE Transactions on Knowledge and Data Engineering.

[9]  Tamsin A. Mather,et al.  On the lack of InSAR observations of magmatic deformation at Central American volcanoes , 2013 .

[10]  Sotiris B. Kotsiantis,et al.  Supervised Machine Learning: A Review of Classification Techniques , 2007, Informatica.

[11]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[12]  Barbara Orecchio,et al.  Seismogenic stress field estimation in the Calabrian Arc region (south Italy) from a Bayesian approach , 2016 .

[13]  C. W. Chen,et al.  Two-dimensional phase unwrapping with use of statistical models for cost functions in nonlinear optimization. , 2001, Journal of the Optical Society of America. A, Optics, image science, and vision.

[14]  Yoshua Bengio,et al.  Why Does Unsupervised Pre-training Help Deep Learning? , 2010, AISTATS.

[15]  Sergio M. Savaresi,et al.  Unsupervised learning techniques for an intrusion detection system , 2004, SAC '04.

[16]  Tamsin A. Mather,et al.  Applicability of InSAR to tropical volcanoes: insights from Central America , 2013 .

[17]  Matthew E. Pritchard,et al.  Global Volcano Monitoring: What Does It Mean When Volcanoes Deform? , 2017 .

[18]  Matthew Wilks,et al.  Evidence for cross rift structural controls on deformation and seismicity at a continental rift caldera , 2018 .

[19]  Pierre Baldi,et al.  Learning Activation Functions to Improve Deep Neural Networks , 2014, ICLR.

[20]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[21]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[22]  Jie Wang,et al.  Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery , 2014, Remote. Sens..

[23]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Antonio Pepe,et al.  Volcano Geodesy: Recent developments and future challenges , 2017 .

[25]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[26]  Juliet Biggs,et al.  Multiple inflation and deflation events at Kenyan volcanoes, East African Rift , 2009 .

[27]  S. Pascale,et al.  Evaluation of prediction capability of the artificial neural networks for mapping landslide susceptibility in the Turbolo River catchment (northern Calabria, Italy) , 2014 .

[28]  Falk Amelung,et al.  Precursory inflation of shallow magma reservoirs at west Sunda volcanoes detected by InSAR , 2012 .

[29]  Masakazu Matsugu,et al.  Subject independent facial expression recognition with robust face detection using a convolutional neural network , 2003, Neural Networks.

[30]  Matthew E. Pritchard,et al.  Synthesis of global satellite observations of magmatic and volcanic deformation: implications for volcano monitoring & the lateral extent of magmatic domains , 2018, Journal of Applied Volcanology.

[31]  Arthur L. Samuel,et al.  Some Studies in Machine Learning Using the Game of Checkers , 1967, IBM J. Res. Dev..

[32]  Wei Zhang,et al.  The application of decision tree to intensity change classification of tropical cyclones in western North Pacific , 2013 .

[33]  T. Wright,et al.  Multi-interferogram method for measuring interseismic deformation: Denali Fault, Alaska , 2007 .

[34]  Zhenhong Li,et al.  Interferometric synthetic aperture radar atmospheric correction using a GPS-based iterative tropospheric decomposition model , 2018 .

[35]  R. S. J. Sparks,et al.  Global link between deformation and volcanic eruption quantified by satellite imagery , 2014, Nature Communications.

[36]  Jan-Peter Muller,et al.  Interferometric synthetic aperture radar (InSAR) atmospheric correction: GPS, moderate resolution Imaging spectroradiometer (MODIS), and InSAR integration , 2005 .

[37]  G. Kane.  Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol 1: Foundations, vol 2: Psychological and Biological Models , 1994 .

[38]  Demetris Stathakis,et al.  Neural networks as a tool for constructing continuous NDVI time series from AVHRR and MODIS , 2008 .

[39]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[40]  Basil Tikoff,et al.  Dynamics of a large, restless, rhyolitic magma system at Laguna del Maule, southern Andes, Chile , 2014 .

[41]  Nguyen Quoc Thanh,et al.  Spatial prediction of rainfall-induced landslides for the Lao Cai area (Vietnam) using a hybrid intelligent approach of least squares support vector machines inference model and artificial bee colony optimization , 2017, Landslides.

[42]  A. Hooper,et al.  Volcanology: lessons learned from synthetic aperture radar imagery , 2014 .

[43]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[44]  J. Sartohadi,et al.  Lessons learned from the 2010 evacuations at Merapi volcano , 2013 .

[45]  Christopher R. J. Kilburn,et al.  Volcanoes of the World , 1997 .

[46]  Elias Lewi,et al.  Pulses of deformation reveal frequently recurring shallow magmatic activity beneath the Main Ethiopian Rift , 2011 .

[47]  Marie-Pierre Doin,et al.  Improving InSAR geodesy using Global Atmospheric Models , 2014 .

[48]  M. Simons,et al.  An InSAR‐based survey of volcanic deformation in the central Andes , 2004 .

[49]  C. Werner,et al.  Satellite radar interferometry: Two-dimensional phase unwrapping , 1988 .

[50]  Amir Hossein Alavi,et al.  Machine learning in geosciences and remote sensing , 2016 .

[51]  Nello Cristianini,et al.  An Introduction to Support Vector Machines and Other Kernel-based Learning Methods , 2000 .

[52]  R. Sparks,et al.  A statistical analysis of the global historical volcanic fatalities record , 2013, Journal of Applied Volcanology.

[53]  Xing Li,et al.  Magmatic architecture within a rift segment: Articulate axial magma storage at Erta Ale volcano, Ethiopia , 2017 .

[54]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[55]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[56]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[57]  Philipp Slusallek,et al.  Introduction to real-time ray tracing , 2005, SIGGRAPH Courses.

[58]  Matthew E. Pritchard,et al.  Surveying Volcanic Arcs with Satellite Radar Interferometry: The Central Andes, Kamchatka, and Beyond , 2004 .

[59]  Zhong Lu,et al.  Systematic assessment of atmospheric uncertainties for InSAR data at volcanic arcs using large-scale atmospheric models: Application to the Cascade volcanoes, United States , 2015 .

[60]  I. Yilmaz.  Comparison of landslide susceptibility mapping methodologies for Koyulhisar, Turkey: conditional probability, logistic regression, artificial neural networks, and support vector machine , 2010 .