Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
153 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Estimating Uncertainty in Landslide Segmentation Models (2311.11138v2)

Published 18 Nov 2023 in cs.CV

Abstract: Landslides are a recurring, widespread hazard. Preparation and mitigation efforts can be aided by a high-quality, large-scale dataset that covers global at-risk areas. Such a dataset currently does not exist and is impossible to construct manually. Recent automated efforts focus on deep learning models for landslide segmentation (pixel labeling) from satellite imagery. However, it is also important to characterize the uncertainty or confidence levels of such segmentations. Accurate and robust uncertainty estimates can enable low-cost (in terms of manual labor) oversight of auto-generated landslide databases to resolve errors, identify hard negative examples, and increase the size of labeled training data. In this paper, we evaluate several methods for assessing pixel-level uncertainty of the segmentation. Three methods that do not require architectural changes were compared, including Pre-Threshold activations, Monte-Carlo Dropout and Test-Time Augmentation -- a method that measures the robustness of predictions in the face of data augmentation. Experimentally, the quality of the latter method was consistently higher than the others across a variety of models and metrics in our dataset.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (35)
  1. Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE transactions on pattern analysis and machine intelligence, 39(12):2481–2495, 2017.
  2. Deeplab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected crfs. IEEE transactions on pattern analysis and machine intelligence, 40(4):834–848, 2017.
  3. Automated landslides detection for mountain cities using multi-temporal remote sensing imagery. Sensors, 18(3):821, 2018.
  4. Natural disaster hotspots: A global risk analysis. World Bank Disaster Risk Management Series, 5:1–132, 01 2005.
  5. Anthropogenic contribution to global occurrenceof heavy-precipitation andhigh-temperature extremes. Nature Climate Change, 5(6):560–564, 2015.
  6. Learning dynamics from kinematics: Estimating 2d foot pressure maps from video frames. arXiv preprint arXiv:1811.12607, 2018.
  7. Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In international conference on machine learning, pages 1050–1059, 2016.
  8. Evaluation of different machine learning methods and deep-learning convolutional neural networks for landslide detection. Remote Sensing, 11(2):196, 2019.
  9. Alex Graves. Practical variational inference for neural networks. In Advances in neural information processing systems, pages 2348–2356, 2011.
  10. Summary metadata - landslide inventories across the united states, 2019.
  11. What uncertainties do we need in bayesian deep learning for computer vision? In I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, editors, Advances in Neural Information Processing Systems 30, pages 5574–5584. Curran Associates, Inc., 2017.
  12. Simple and scalable predictive uncertainty estimation using deep ensembles. In Advances in neural information processing systems, pages 6402–6413, 2017.
  13. Landslide inventory mapping from bitemporal images using deep convolutional neural networks. IEEE Geoscience and Remote Sensing Letters, 16(6):982–986, 2019.
  14. Accurate uncertainty estimation and decomposition in ensemble learning. In Advances in Neural Information Processing Systems, pages 8950–8961, 2019.
  15. A new rainfall-induced deep learning strategy for landslide susceptibility prediction. In AGU Fall Meeting Abstracts, volume 2021, pages NH35E–0504, 2021.
  16. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3431–3440, 2015.
  17. David JC MacKay. A practical bayesian framework for backpropagation networks. Neural computation, 4(3):448–472, 1992.
  18. A simple baseline for bayesian uncertainty in deep learning. In Advances in Neural Information Processing Systems, pages 13132–13143, 2019.
  19. An efficient deep learning mechanism for cross-region generalization of landslide events. In AGU Fall Meeting Abstracts, volume 2020, pages NH030–0010, 2020.
  20. Patchrefinenet: Improving binary segmentation by incorporating signals from optimal patch-wise binarization. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 1361–1372, 2024.
  21. Constructing a large-scale landslide database across heterogeneous environments using task-specific model updates. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 15:4349–4370, 2022.
  22. Cloud-based interactive database management suite integrated with deep learning-based annotation tool for landslide mapping. In AGU Fall Meeting Abstracts, volume 2020, pages NH030–0011, 2020.
  23. Comparison of reinforcement learning algorithms applied to the cart-pole problem. In 2017 international conference on advances in computing, communications and informatics (ICACCI), pages 26–32. IEEE, 2017.
  24. Threshnet: Segmentation refinement inspired by region-specific thresholding. arXiv preprint arXiv:2211.06560, 2, 2022.
  25. Radford M Neal. Bayesian learning for neural networks, volume 118. Springer Science & Business Media, 2012.
  26. Utilizing an interactive ai-empowered web portal for landslide labeling for establishing a landslide database in washington state, usa. In EGU General Assembly Conference Abstracts, pages EGU21–13974, 2021.
  27. Dave Petley. An analysis of fatal landslides, and the resultant deaths, in 2017, Apr 2018.
  28. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical image computing and computer-assisted intervention, pages 234–241. Springer, 2015.
  29. Epistemic and aleatoric uncertainty in modeling. In The Mechanics of Jointed Structures, pages 593–603. Springer, 2018.
  30. Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15:1929–1958, 2014.
  31. Climate change 2013: the physical science basis: final draft underlying scientific-technical assessment: Working Group I contribution to the IPCC fifth assessment report. WMO, IPCC Secretariat, 2013.
  32. Landslide geohazard assessment with convolutional neural networks using sentinel-2 imagery data. In IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium, pages 9646–9649. IEEE, 2019.
  33. Aleatoric uncertainty estimation with test-time augmentation for medical image segmentation with convolutional neural networks. Neurocomputing, 338:34–45, 2019.
  34. Obtaining calibrated probability estimates from decision trees and naive bayesian classifiers. In Icml, volume 1, pages 609–616. Citeseer, 2001.
  35. A rapid and realistic 3d stratigraphic model generator conditioned on reference well log data. In Second EAGE Digitalization Conference and Exhibition, volume 2022, pages 1–5. European Association of Geoscientists & Engineers, 2022.
Citations (3)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now