An unsupervised representation learning approach for modelling forest landform characteristics and fire susceptibility assessment

doi:10.7523/j.ucas.2021.0057

Abstract

Abstract: Destructive wildfires have caused extraordinary losses in both economic and natural property worldwide with an even increasing frequency in recent decades. One practical approach in forest fire susceptibility prediction is using statistical learning methods to learn from historic data. Conventional methods use handcraft feature to reduce data dimension. With the continuous development of remote sensing technology, the difficulty of obtaining high-precision gridded multi-dimensional forest landform information is constantly decreasing. It is difficult to make full use of such data through handcraft features, which limits the performance of conventional methods when applied in the real world. This paper introduces a novel approach to model forest geographic information through deep representation learning, which is, leveraging deep convolutional neural network and state-of-the-art representation learning methods to extract feature embedding for a given area of interest. Fire susceptibility assessment experiments are used to evaluate the proposed method and compare the unsupervised learning and its supervised counterpart to show its effectiveness.

Key words: representation learning, convolutional neural network (CNN), deep clustering, forest fire risks

CLC Number:

ZHUANG Zijun, YUAN Xiaobing, PEI Jun, WANG Guohui, LIU Jianpo. An unsupervised representation learning approach for modelling forest landform characteristics and fire susceptibility assessment[J]. Journal of University of Chinese Academy of Sciences, 2023, 40(2): 227-239.

References

[1] Wehner M, Arnold J, Knutson T, et al. Chapter 8:droughts, floods, and wildfire. Climate science special report:Fourth national climate assessment (NCA4), Volume I[R]. U.S. Global Change Research Program, 2017.DOI:10.7930/JOCJ8BNN.
[2] Ngoc Thach N, Bao-Toan Ngo D, Xuan-Canh P, et al. Spatial pattern assessment of tropical forest fire danger at Thuan Chau area (Vietnam) using GIS-based advanced machine learning algorithms:a comparative study[J]. Ecological Informatics, 2018, 46:74-85.DOI:10.1016/j.ecoinf.2018.05.009.
[3] Pettinari M L, Chuvieco E. Fire behavior simulation from global fuel and climatic information[J]. Forests, 2017, 8(6):179.DOI:10.3390/f8060179.
[4] Finney M A. FARSITE:fire Area Simulator-model development and evaluation[R]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, 1998.
[5] Tymstra C, Bryce R, Wotton B, et al. Development and structure of Prometheus:the Canadian wildland fire growth simulation model[R/OL]. (2013-04-03)[2021-04-06]. http://publications.gc.ca/pub?id=9.619969&sl=1.
[6] Wotton B M, Martell D L, Logan K A. Climate change and people-caused forest fire occurrence in Ontario[J]. Climatic Change, 2003, 60(3):275-295.DOI:10.1023/A:1026075919710.
[7] Koutsias N, Martínez-Fernández J, Allgöwer B. Do factors causing wildfires vary in space? evidence from geographically weighted regression[J]. GIScience & Remote Sensing, 2010, 47(2):221-240.DOI:10.2747/1548-1603.47.2.221.
[8] Conedera M, Torriani D, Neff C, et al. Using Monte Carlo simulations to estimate relative fire ignition danger in a low-to-medium fire-prone region[J]. Forest Ecology and Management, 2011, 261(12):2179-2187.DOI:10.1016/j.foreco.2010.08.013.
[9] Lautenberger C. Wildland fire modeling with an Eulerian level set method and automated calibration[J]. Fire Safety Journal, 2013, 62:289-298.DOI:10.1016/j.firesaf.2013.08.014.
[10] Bisquert M, Caselles E, Sánchez J M, et al. Application of artificial neural networks and logistic regression to the prediction of forest fire danger in Galicia using MODIS data[J]. International Journal of Wildland Fire, 2012, 21(8):1025-1029.DOI:10.1071/wf11105.
[11] Satir O, Berberoglu S, Donmez C. Mapping regional forest fire probability using artificial neural network model in a Mediterranean forest ecosystem[J]. Geomatics, Natural Hazards and Risk, 2016, 7(5):1645-1658.DOI:10.1080/19475705.2015.1084541.
[12] Ghorbanzadeh O, Blaschke T, Gholamnia K, et al. Forest fire susceptibility and risk mapping using social/infrastructural vulnerability and environmental variables[J]. Fire, 2019, 2(3):50.DOI:10.3390/fire2030050.
[13] Pourghasemi H R, Gayen A, Lasaponara R, et al. Application of learning vector quantization and different machine learning techniques to assessing forest fire influence factors and spatial modelling[J]. Environmental Research, 2020, 184:109321.DOI:10.1016/j.envres.2020.109321.
[14] LeCun Y, Bengio Y, Hinton G. Deep learning[J]. Nature, 2015, 521(7553):436-444.DOI:10.1038/nature14539.
[15] Zhang G L, Wang M, Liu K. Forest fire susceptibility modeling using a convolutional neural network for Yunnan Province of China[J]. International Journal of Disaster Risk Science, 2019, 10(3):386-403.DOI:10.1007/s13753-019-00233-1.
[16] Hodges J L, Lattimer B Y. Wildland fire spread modeling using convolutional neural networks[J]. Fire Technology, 2019, 55(6):2115-2142.DOI:10.1007/s10694-019-00846-4.
[17] Zhang C, Pan X, Li H P, et al. A hybrid MLP-CNN classifier for very fine resolution remotely sensed image classification[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 140:133-144.DOI:10.1016/j.isprsjprs.2017.07.014.
[18] Liu T, Abd-Elrahman A. Deep convolutional neural network training enrichment using multi-view object-based analysis of Unmanned Aerial systems imagery for wetlands classification[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 139:154-170.DOI:10.1016/j.isprsjprs.2018.03.006.
[19] Wang Y, Fang Z C, Hong H Y. Comparison of convolutional neural networks for landslide susceptibility mapping in Yanshan County, China[J]. Science of the Total Environment, 2019, 666:975-993.DOI:10.1016/j.scitotenv.2019.02.263.
[20] Landfire. Forest canopy cover layer, forest canopy height layer, etc[EB/OL]. US Department of the Interior, Geological Survey, (2019)[2020-06-23]. http://landfire.cr.usgs.gov/viewer/.
[21] Bengio Y, Courville A, Vincent P. Representation learning:a review and new perspectives[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8):1798-1828.DOI:10.1109/TPAMI.2013.50.
[22] Pan S J, Yang Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(10):1345-1359.DOI:10.1109/TKDE.2009.191.
[23] Kolesnikov A, Zhai X H, Beyer L. Revisiting self-supervised visual representation learning[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 15-20, 2019, Long Beach, CA, USA. IEEE, 2020:1920-1929.DOI:10.1109/CVPR.2019.00202.
[24] Donahue J, Krähenbühl P, Darrell T. Adversarial feature learning[EB/OL]. arXiv:1605.09782. (2017-04-03)[2021-04-06]. https://arxiv.org/abs/1605.09782.
[25] Le Q V. Building high-level features using large scale unsupervised learning[C]//2013 IEEE International Conference on Acoustics, Speech and Signal Processing. May 26-31, 2013, Vancouver, BC, Canada. IEEE, 2013:8595-8598.DOI:10.1109/ICASSP.2013.6639343.
[26] Lee H, Grosse R, Ranganath R, et al. Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations[C]//Proceedings of the 26th Annual International Conference on Machine Learning-ICML'09. June 14-18, 2009, Montreal, Quebec, Canada. New York:ACM Press, 2009:609-616.DOI:10.1145/1553374.1553453.
[27] Kingma D P, Welling M. Auto-encoding variational bayes[EB/OL]. arXiv:1312.6114. (2014-05-01)[2021-04-06]. https://arxiv.org/abs/1312.6114.
[28] Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[EB/OL]. arXiv:1406.2661. (2014-06-10)[2021-04-06]. https://arxiv.org/abs/1406.2661.
[29] Doersch C, Gupta A, Efros A A. Unsupervised visual representation learning by context prediction[C]//2015 IEEE International Conference on Computer Vision (ICCV). December 7-13, 2015, Santiago, Chile. IEEE, 2016:1422-1430.DOI:10.1109/ICCV.2015.167.
[30] Noroozi M, Favaro P. Unsupervised learning of visual representations by solving jigsaw puzzles[M]//Computer Vision-ECCV 2016. Cham:Springer International Publishing,2016:69-84. DOI:10.1007/978-3-319-46466-4_5.
[31] Larsson G, Maire M, Shakhnarovich G. Learning representations for automatic colorization[M]//Computer Vision-ECCV 2016. Cham:Springer International Publishing,2016:577-593. DOI:10.1007/978-3-319-46493-0_35.
[32] Pathak D, Krähenbühl P, Donahue J, et al. Context encoders:feature learning by inpainting[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016:2536-2544.DOI:10.1109/CVPR.2016.278.
[33] Zhang R, Isola P, Efros A A. Split-brain autoencoders:Unsupervised learning by cross-channel prediction[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). July 21-26, 2017, Honolulu, HI, USA. IEEE, 2017:645-654.DOI:10.1109/CVPR.2017.76.
[34] Noroozi M, Pirsiavash H, Favaro P. Representation learning by learning to count[C]//2017 IEEE International Conference on Computer Vision (ICCV). October 22-29, 2017, Venice, Italy. IEEE, 2017:5899-5907.DOI:10.1109/ICCV.2017.628.
[35] Gidaris S, Singh P, Komodakis N. Unsupervised representation learning by predicting image rotations[EB/OL]. arXiv:1803.07728. (2018-03-21)[2021-04-06]. https://arxiv.org/abs/1803.07728.
[36] Chen X, Fan H, Girshick R, et al. Improved baselines with momentum contrastive learning[EB/OL]. arXiv:2003.04297. (2020-03-09)[2021-04-06]. https://arxiv.org/abs/2003.04297.
[37] Gillies S, Ward B, Petersen A, et al. Rasterio:geospatial raster I/O for Python programmers[EB/OL]. (2013)[2020-06-21] https://github. com/mapbox/rasterio.
[38] Scott J H, Burgan R E. Standard fire behavior fuel models:a comprehensive set for use with Rothermel's surface fire spread model[R]. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, 2005.
[39] Zhan X H, Xie J H, Liu Z W, et al. Online deep clustering for unsupervised representation learning[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 13-19, 2020, Seattle, WA, USA. IEEE, 2020:6687-6696.DOI:10.1109/CVPR42600.2020.00672.
[40] He K M, Fan H Q, Wu Y X, et al. Momentum contrast for unsupervised visual representation learning[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). June 13-19, 2020, Seattle, WA, USA. IEEE, 2020:9726-9735.DOI:10.1109/CVPR42600.2020.00975.
[41] Oord A Van Den, Li Y, Vinyals O. Representation learning with contrastive predictive coding[EB/OL]. arXiv:1807.03748. (2019-01-22)[2021-04-06]. https://arxiv.org/abs/1807.03748.
[42] Janocha K, Czarnecki W M. On loss functions for deep neural networks in classification[EB/OL]. arXiv:1702.05659. (2017-02-18)[2021-04-06]. https://arxiv.org/abs/1702.05659.
[43] Pan H Y, Han H, Shan S G, et al. Mean-variance loss for deep age estimation from a face[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. June 18-23, 2018, Salt Lake City, UT, USA. IEEE, 2018:5285-5294.DOI:10.1109/CVPR.2018.00554.
[44] He K M, Zhang X Y, Ren S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016:770-778.DOI:10.1109/CVPR.2016.90.