Automatic Identification of Thaw Slumps Based on Neural Network Methods and Thaw Slumping Susceptibility

Abstract

Thaw slumping is a periglacial process that occurs on slopes in cold environments, where the ground becomes unstable and the surface slides downhill due to saturation with water during thawing. In this study, GaoFen-1 remote sensing and fused multi-source feature data were used to automatically map thaw slumping landforms in the Beilu River Basin of the Qinghai–Tibet Plateau. The bi-directional cascade network structure was used to extract edges at different scales, where an individual layer was supervised by labeled edges at its specific scale, rather than directly applying the same supervision to all convolutional neural network outputs. Additionally, we conducted a 5-year multi-scale feature analysis of small baseline subset interferometric synthetic aperture radar deformation, normalized difference vegetation index, and slope, among other features. Our study analyzed the performance and accuracy of three methods based on edge object supervised learning and three preconfigured neural networks, ResNet101, VGG16, and ResNet152. Through verification using site surveys and multi-data fusion results, we obtained the best ResNet101 model score of intersection over union of 0.85 (overall accuracy of 84.59%).The value of intersection over union of the VGG and ResNet152 are 0.569 and 0.773, respectively. This work provides a new insight for the potential feasibility of applying the designed edge detection method to map diverse thaw slumping landforms in larger areas with high-resolution images.

Keywords

Bi-directional cascade network / Remote sensing / SBAS-InSAR / Thaw slumping

References

[1] Aas KS, Martin L, Nitzbon J, Langer M, Westermann S. Thaw processes in ice-rich permafrost landscapes represented with laterally coupled tiles in a land surface model. The Cryosphere, 2019, 13(2): 591-609

[2] Anyamba A, Tucker CJ. Analysis of Sahelian vegetation dynamics using NOAA-AVHRR NDVI data from 1981–2003. Journal of Arid Environments, 2005, 63(3): 596-614

[3] Berardino P, Fornaro G, Lanari R, Sansosti E. A new algorithm for surface deformation monitoring based on small baseline differential SAR interferograms. IEEE Transactions on Geoscience and Remote Sensing, 2002, 40(1): 2375-2383

[4] Bo, Y., and I. Lane. 2015. Multi-task deep learning for image understanding. Paper presented at the 2014 6th International Conference of Soft Computing and Pattern Recognition (SoCPaR), 11–14 August 2014, Tunis, Tunisia.

[5] Eigen, D., and R. Fergus. 2014. Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture. In Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), 7–13 December 2015, Santiago, Chile.

[6] Farabet C, Couprie C, Najman L, Yann L. Learning hierarchical features for scene labeling. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2013, 35(8): 1915-1929

[7] Hu B, Yang B, Zhang X, Chen X, Wu Y. Time-series displacement of land subsidence in Fuzhou downtown, monitored by SBAS-InSAR technique. Journal of Sensors, 2019, 2019: 3162652

[8] Huang L, Liu L, Jiang J, Zhang T. Automatic mapping of thermokarst landforms from remote sensing images using deep learning: A case study in the northeastern Tibetan Plateau. Remote Sensing, 2018, 10(12): 2067

[9] Jolivet R, Lasserre C, Doin M-P, Guillaso S, Peltzer G, Dailu R, Sun J, Shen Z-K, Xu X. Shallow creep on the Haiyuan fault (Gansu, China) revealed by SAR interferometry. Journal of Geophysical Research: Solid Earth, 2012

[10] Koven CD, Ringeval B, Friedlingstein P, Ciais P, Cadule P, Khvorostyanov D, Krinner G, Tarnocai C. Permafrost carbon-climate feedbacks accelerate global warming. Proceedings of the National Academy of Sciences of the United States of America, 2011, 108(36): 14769-14774

[11] Li, H., L. Zhe, X. Shen, J. Brandt, and H. Gang. 2015. A convolutional neural network cascade for face detection. Paper presented at the 2015 IEEE Conference on Computer Vision & Pattern Recognition (CVPR), 7–12 June 2015, Boston, MA, USA.

[12] Li, X., Z. Liu, P. Luo, C.C. Loy, and X. Tang. 2017. Not all pixels are equal: Difficulty-aware semantic segmentation via deep layer cascade. Paper presented at the 2017 IEEE Conference on Computer Vision & Pattern Recognition (CVPR), 21–26 July 2017, Honolulu, HI, USA.

[13] Lian X, Pang Y, Han J, Pan J. Cascaded hierarchical atrous spatial pyramid pooling module for semantic segmentation. Pattern Recognition, 2021, 110: 107622

[14] Luo J, Niu F, Lin Z, Liu M, Yin G. Thermokarst lake changes between 1969 and 2010 in the Beilu River Basin, Qinghai–Tibet Plateau. China. Science Bulletin, 2015, 60(5): 556-564.

[15] Murthy, V.N., V. Singh, T. Chen, R. Manmatha, and D. Comaniciu. 2016. Deep decision network for multi-class image classification. Paper presented at the 2016 IEEE Conference on Computer Vision & Pattern Recognition (CVPR), 27–30 June 2016, Las Vegas, NV, USA.

[16] Niu F, Zhang L, Yu Q, Xie Q. Study on slope types and stability of typical slopes in permafrost regions of the Tibetan Plateau. Journal of Glaciology and Geocryology, 2002, 5: 608-613 (in Chinese)

[17] Nowozin, S. 2014. Optimal decisions from probabilistic models: The intersection-over-union case. Paper presented at the 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 23–28 June 2014, Columbus, OH, USA.

[18] Peng X, Zhang T, Frauenfeld OW, Wa Ng K, Luo J. Evaluation and quantification of surface air temperature over Eurasia based on CMIP5 models. Climate Research, 2018, 77(2): 167-180

[19] Pinheiro, P., and R. Collobert. 2013. Recurrent convolutional neural networks for scene parsing. http://arxiv.org/abs/1306.2795. https://doi.org/10.48550/arXiv.1306.2795.

[20] Qayyum A, Iftikhar Ahmad WM, Alassafi MO, Alghamdi R, Mazher M, Mazher M. Automatic segmentation using a hybrid dense network integrated with an 3D-Atrous spatial pyramid pooling module for computed tomography (CT) imaging. IEEE Access, 2020, 8: 169794-169803

[21] Tizzani P, Berardino P, Casu F, Euillades P, Manzo M, Ricciardi GP, Zeni G, Lanari R. Surface deformation of Long Valley caldera and Mono Basin, California, investigated with the SBAS-InSAR approach. Remote Sensing of Environment, 2007, 108(3): 277-289

[22] Wang C, Zhang Z, Zhang H, Zhang B, Tang Y, Wu Q. Active layer thickness retrieval of Qinghai–Tibet permafrost using the TerraSAR-X InSAR technique. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2018, 11(11): 4403-4413

[23] Zhang Z, Wang M, Wu Z, Liu X. Permafrost deformation monitoring along the Qinghai–Tibet plateau engineering corridor using InSAR observations with multi-sensor SAR datasets from 1997–2018. Sensors (Basel), 2019, 19(23): 5306

[24] Zhao L, Zou D, Hu G, Du E, Pang Q, Xiao Y, Li R, Sheng Y Changing climate and the permafrost environment on the Qinghai–Tibet (Xizang) Plateau. Permafrost and Periglacial Processes, 2020, 31(3): 396-405