Advancing debris flow detection based on deep learning model and high-resolution images

Liem Ngo Van; Hieu Nguyen; Bac Dang Kinh; Linh Giang Tuan; Bao Dang Van; Hieu Do Trung; Hieu Nguyen Minh; Vu Dang Nguyen; Duc Dao Minh

doi:10.15625/2615-9783/23027

Author affiliations

Authors

Ngo Van Liem VNU University of Science, Vietnam National University, Hanoi, Vietnam
Nguyen Hieu VNU University of Science, Vietnam National University, Hanoi, Vietnam
Dang Kinh Bac VNU University of Science, Vietnam National University, Hanoi, Vietnam
Giang Tuan Linh 1-VNU University of Science, Vietnam National University, Hanoi, Vietnam; 2-VNU Institute of Vietnamese Studies and Development Science (VNU-IVIDES), Vietnam National University, Hanoi, Vietnam
Dang Van Bao VNU University of Science, Vietnam National University, Hanoi, Vietnam
Do Trung Hieu VNU University of Science, Vietnam National University, Hanoi, Vietnam
Nguyen Minh Hieu VNU University of Science, Vietnam National University, Hanoi, Vietnam
Dang Nguyen Vu VNU University of Science, Vietnam National University, Hanoi, Vietnam
Dao Minh Duc Institute of Earth Sciences, Vietnam Academy of Science and Technology, Hanoi, Vietnam

DOI:

https://doi.org/10.15625/2615-9783/23027

Keywords:

Deep learning, U-Net; U2-Net, debris flows, Vietnam

Abstract

Debris flow inventory is an essential task for scientists and managers to mitigate danger to humans, especially in mountainous areas. However, rapid land use and cover change, as well as technological limitations, make it a challenging task. Monitoring debris-flow efforts, especially in hilly places with limited transportation and technology, may improve management to minimize damage caused by this hazard. This work assesses U-shaped deep learning architectures, focusing on the roles of image size, optimization procedures, and data quality in debris flow trace identification using U-Net and U2-Net. While new debris flows can be detected through machine learning modeling, the U-Net model, combined with the Adam optimizer and an input size of 64×64, has been proven to be efficient, accurate, and stable. Small debris traces that can be used for planning debris thickness maps were easily identified in Worldview-2 and UAV images but not in the medium-resolution remote sensing data. When applied to Bat Xat district, Vietnam, the models identified that the distribution of debris flows is not uniform and depends on natural factors, such as rainfall and human-interpolated factors, including the construction of structures. The study also establishes the need to continually assess and incorporate big data for enhanced debris flow hazard assessment and mitigation. Further developments should focus on the effective use of multi-spectral and large-scale topographic data to strengthen disaster risk identification and provide recommendations for disaster risk reduction.

Downloads

Download data is not yet available.

References

Abdallah A., Hamada M., Nurseitov D., 2020. Attention-Based Fully Gated CNN-BGRU for Russian Handwritten Text. Journal of Imaging, 6, 141. https://doi.org/10.3390/jimaging6120141.

Alom M.Z., Taha T.M., Yakopcic C., Westberg S., Sidike P., Nasrin M.S., Hasan M., Van Essen B.C., Awwal A.A.S., Asari V.K., 2019. A state-of-the-art survey on deep learning theory and architectures. Electronics (Switzerland), 8, 1–67. https://doi.org/10.3390/electronics8030292.

Boehm V., Ji Leong W., Bal Mahesh R., Prapas I., Kalaitzis F., Ganju S., Ramos-Pollan R., n.d. Deep Learning for Rapid Landslide Detection using Synthetic Aperture Radar (SAR) Datacubes. 1–8.

Borrelli L., Antronico L., Gullà G., Sorriso-Valvo G.M., 2014. Geology, geomorphology and dynamics of the 15 February 2010 Maierato landslide (Calabria, Italy). Geomorphology, 208, 50–73. https://doi.org/10.1016/j.geomorph.2013.11.015.

Brockerhoff E.G., Jactel H., Parrotta J.A., Ferraz S.F.B., 2013. Forest Ecology and Management Role of eucalypt and other planted forests in biodiversity conservation and the provision of biodiversity-related ecosystem services. Forest Ecology and management, 301, 43–50. https://doi.org/10.1016/j.foreco.2012.09.018.

Carrión-Mero P., Montalván-Burbano N., Morante-Carballo F., Quesada-Román A., Apolo-Masache B., 2021. Worldwide research trends in landslide science. International Journal of Environmental Research and Public Health, 18. https://doi.org/10.3390/ijerph18189445.

Casagli N., Intrieri E., Tofani V., Gigli G., Raspini F., 2023. Landslide detection, monitoring and prediction with remote-sensing techniques. Nature Reviews Earth and Environment, 4, 51–64. https://doi.org/10.1038/s43017-022-00373-x.

D'Amato D., Rekola M., Wan M., Cai D., Toppinen A., 2017. Effects of industrial plantations on ecosystem services and livelihoods: Perspectives of rural communities in China. Land Use Policy, 63, 266–278. https://doi.org/10.1016/j.landusepol.2017.01.044.

Dang K.B., Dang V.B., Bui Q.T., Nguyen V.V., Pham T.P.N., Ngo V.L., 2020. A Convolutional Neural Network for Coastal Classification Based on ALOS and NOAA Satellite Data. IEEE Access, 8, 11824–11839. https://doi.org/10.1109/ACCESS.2020.2965231.

Dang K.B., Dang V.B., Ngo V.L., Vu K.C., Nguyen H., Nguyen D.A., Nguyen T.D.L., Pham T.P.N., Giang T.L., Nguyen H.D., Hieu Do T., 2022. Application of deep learning models to detect coastlines and shorelines. Journal of Environmental Management 320, 115732.

Dang K.B., Giang T.L., Dang V.B., Phan T.T., Truong Q.H., Ngo V.L., Do T.H., Dang N.V., Forino G., 2024a. Deep learning models integrating multi-sensor and -temporal remote sensing to monitor landslide traces in Vietnam. International Journal of Disaster Risk Reduction, 105, 104391. https://doi.org/10.1016/j.ijdrr.2024.104391.

Dang K.B., Giang T.L., Dang V.B., Phan T.T., Truong Q.H., Ngo V.L., Do T.H., Dang N.V., Forino G., 2024b. Deep learning models integrating multi-sensor and -temporal remote sensing to monitor landslide traces in Vietnam. International Journal of Disaster Risk Reduction, 105, 104391. https://doi.org/10.1016/j.ijdrr.2024.104391.

Dang K.B., Nguyen C.Q., Tran Q.C., Nguyen H., Nguyen T.T.T.T., Nguyen D.A., Tran T.H., Bui P.T., Giang T.L., Lenh T.A., Ngo V.L., Yasir M., Nguyen T.T.T.T., Ngo H.H., Quan C., Cuong Q., Nguyen H., Thuy T., Hao H., 2024c. Comparison between U-shaped structural deep learning

models to detect landslide traces. Science of the Total Environment, 912, 169113. https://doi.org/10.1016/j.scitotenv.2023.169113.

Dao D.B., Dang V.B., Nguyen H., 2008. Towards authentic warning of high-risk areas for some common natural disasters in Vietnam. Proceedings of the Third International Conference on Vietnam Studies. Hanoi National University Publishing House, 441–461.

Diakogiannis F.I., Waldner F., Caccetta P., Wu C., 2020. ResUNet-a: A deep learning framework for semantic segmentation of remotely sensed data. ISPRS Journal of Photogrammetry and Remote Sensing, 162, 94–114. https://doi.org/10.1016/j.isprsjprs.2020.01.013.

Dong Z., An S., Zhang J., Yu J., Li J., Xu D., 2022. L-Unet: A Landslide Extraction Model Using Multi-Scale Feature Fusion and Attention Mechanism. Remote Sensing, 14, 1–13. https://doi.org/10.3390/rs14112552.

Falbel D., Allaire J., François Tang, Y. Bijl, W. Van Der, Studer M., Keydana S., 2019. R Interface to "Keras."

Gallo F., Lavé J., 2014. Evolution of a large landslide in the High Himalaya of central Nepal during the last half-century. Geomorphology, 223, 20–32. https://doi.org/10.1016/j.geomorph.2014.06.021.

Ghorbanzadeh O., Shahabi H., Crivellari A., Homayouni S., Blaschke T., Ghamisi P., 2022. Landslide detection using deep learning and object-based image analysis. Landslides, 19, 929–939. https://doi.org/10.1007/s10346-021-01843-x.

Giang N.C., Chien N.Q., Khac D.V., 2024. Estimation of greenhouse gas emission due to open burning of rice straw using Sentinel data. Vietnam Journal of Earth Sciences, 46, 381–398.

Giang T.L., Dang K.B., Bui Q.T., 2023. Coastline and shoreline change assessment in sandy coasts based on machine learning models and high-resolution satellite images. Vietnam Journal of Earth Sciences, 45, 251–270.

Gulli A., Pal S., 2017. Deep Learning with Keras- Implement neural networks with Keras on Theano and TensorFlow. Packt Publishing Ltd., Birmingham, UK.

Handwerger A., Jones S., Amatya P., Kerner H., Kirschbaum D., Huang M.-H., 2021. Strategies for landslide detection using open-access synthetic aperture radar backscatter change in Google Earth Engine. Natural Hazards and Earth System Sciences Discussions, 1–35.

https://doi.org/10.1016/j.jenvman.2022.115732.

Huang H., Lin L., Tong R., Hu H., Zhang Q., Iwamoto Y., Han X., Chen Y.W., Wu J., 2020. UNet 3+: A Full-Scale Connected UNet for Medical Image Segmentation. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings 2020-May, 1055–1059. https://doi.org/10.1109/ICASSP40776.2020.9053405.

Hung P. Van, Son P.Q., Dung N. Van, 2015. The study evaluated arming of risk of lanslide in Hoa Binh and Son La reservoir hydropower area on the basis of analyzing high-resolution remote sensing and geographic information systems. Vietnam Journal of Earth Sciences, 37, 193–203.

Iglovikov V., Mushinskiy S., Osin V., 2017. Satellite Imagery Feature Detection using Deep Convolutional Neural Network: A Kaggle Competition. ArXivLabs - Computer Vision and Pattern Recognition, v1, 6.

Iverson R.M., George D.L., Allstadt K., Reid M.E., Collins B.D., Vallance J.W., Schilling S.P., Godt J.W., Cannon C.M., Magirl C.S., Baum R.L., Coe J.A., Schulz W.H., Bower J.B., 2015. Landslide mobility and hazards: Implications of the 2014 Oso disaster. Earth and Planetary Science Letters, 412, 197–208. https://doi.org/10.1016/j.epsl.2014.12.020.

Jiang H., Zou Q., Zhu Y., Li Y., Zhou B., Zhou W., Yao S., Dai X., Yao H., Chen S., 2024. Deep learning prediction of rainfall-driven debris flows considering the similar critical thresholds within comparable background conditions. Environmental Modelling and Software, 179. https://doi.org/10.1016/j.envsoft.2024.106130.

Ju Y., Xu Q., Jin S., Li W., Su Y., Dong X., Guo Q., 2022. Loess Landslide Detection Using Object Detection Algorithms in Northwest China. Remote Sensing, 14. https://doi.org/10.3390/rs14051182.

Kean J.W., McCoy S.W., Tucker G.E., Staley D.M., Coe J.A., 2013. Runoff-generated debris flows: Observations and modeling of surge initiation, magnitude, and frequency. Journal of Geophysical Research: Earth Surface, 118, 2190–2207. https://doi.org/10.1002/jgrf.20148.

Keller E.A., Bean G., Best D., 2015. Fluvial geomorphology of a boulder-bed, debris-flow - Dominated channel in an active tectonic environment. Geomorphology, 243, 14–26. https://doi.org/10.1016/j.geomorph.2015.04.012.

Kirschbaum D., Stanley T., Zhou Y., 2015. Spatial and temporal analysis of a global landslide catalog. Geomorphology, 249, 4–15. https://doi.org/10.1016/j.geomorph.2015.03.016.

Kuschel E., Tolle F., Klaus V., Laa U., Prokop A., Friedt J.M., Bernard E., Zangerl C., 2024. Meteorological factors control debris slides and debris flows in a high-Arctic glacier basin (Ny-Ålesund, Svalbard). Geomorphology, 467. https://doi.org/10.1016/j.geomorph.2024.109492.

Lee S., An H., Kim M., Lee D., Lee J., 2024. Debris flows analysis through quantitative evaluation of soil depth distribution under limited data. Catena, 246. https://doi.org/10.1016/j.catena.2024.108379.

Li C.J., Hu Y.X., Fan G., Zhu Q.Y., Liu D.R., Zhou J.W., 2024. Exploring debris flow deposit morphology in river valleys: Insights from physical modeling experiments. Engineering Geology, 332, 1–15. https://doi.org/10.1016/j.enggeo.2024.107465.

Liu B., Li Y., Li G., Liu A., 2019. A spectral feature based convolutional neural network for classification of sea surface oil spill. ISPRS International Journal of Geo-Information, 8. https://doi.org/10.3390/ijgi8040160.

McKean J., Roering J., 2004. Objective landslide detection and surface morphology mapping using high-resolution airborne laser altimetry. Geomorphology, 57, 331–351. https://doi.org/10.1016/S0169-555X(03)00164-8.

Megown R.A, Webster M., Jacobs S., 1999. Using Landsat TM Imagery to Estimate LAI in a Eucalyptus Plantation, 1–13.

Meinhardt M., Fink M., Tünschel H., 2015. Landslide susceptibility analysis in central Vietnam based on an incomplete landslide inventory: Comparison of a new method to calculate weighting factors by means of bivariate statistics. Geomorphology, 234, 80–97. https://doi.org/10.1016/j.geomorph.2014.12.042.

Nava L., Bhuyan K., Meena S.R., Monserrat O., Catani F., 2022. Rapid Mapping of Landslides on SAR Data by Attention U-Net. Remote Sensing, 14, 1–15. https://doi.org/10.3390/rs14061449.

Nguyen Cong G., Nguyen Quyet C., Dang Vu K., 2024. Estimation of greenhouse gas emission due to open burning of rice straw using Sentinel data. Vietnam Journal of Earth Sciences, 46(3), 381–398. https://doi.org/10.15625/2615-9783/20716.

Nguyen H.D., Pham V.T., Nguyen Q.-H., Bui Q.-T., 2025. Soil salinity prediction using satellite-based variables and machine learning: Case study in Tra Vinh province, Mekong Delta, Vietnam. Vietnam Journal of Earth Sciences, 47, 1–19.

Nguyen Thanh H., et al., 2025. Forest cover change mapping based on Deep Neuron Network, GIS, and High-resolution Imagery. Vietnam Journal of Earth Sciences, 48. https://doi.org/10.15625/2615-9783/22192.

Nhu V.H., Hoang N.D., Nguyen H., Ngo P.T.T., Thanh Bui T., Hoa P.V., Samui P., Tien Bui D., 2020. Effectiveness assessment of Keras based deep learning with different robust optimization algorithms for shallow landslide susceptibility mapping at tropical area. Catena, 188. https://doi.org/10.1016/j.catena.2020.104458.

Pham T.T., Dang K.B., Giang T.L., Hoang T.H.N., Le V.H., Ha H.N., 2024. Deep learning models for monitoring landscape changes in a UNESCO Global Geopark. Journal of Environmental Management, 354, 120497. https://doi.org/10.1016/j.jenvman.2024.120497.

Pham V.T., Le H.L., Tran T.N., Nguyen Q.P., Phan T.T., Dinh T.Q., Dao M.D., Nguyen C.L., Nguyen H.C., 2023. Mechanism and numerical simulation of a rapid deep-seated landslide in Van Hoi reservoir, Vietnam. Vietnam Journal of Earth Sciences, 45, 357–373.

Pierson T., 2005. Distinguishing between debris flows and floods from field evidence in small watersheds. Water, 4.

Pouliot D., Latifovic R., Pasher J., Duffe J., 2019. Assessment of convolution neural networks for wetland mapping with landsat in the central Canadian boreal forest region. Remote Sensing, 11. https://doi.org/10.3390/rs11070772.

Pour A.B., Hashim M., 2017. Application of Landsat-8 and ALOS-2 data for structural and landslide hazard mapping in Kelantan, Malaysia, 1285–1303.

Qin X., Zhang Z., Huang C., Dehghan M., Zaiane O.R., Jagersand M., 2020. U2-Net: Going deeper with nested U-structure for salient object detection. Pattern Recognition, 106. https://doi.org/10.1016/j.patcog.2020.107404.

Ren D., 2015. Storm-triggered landslides in warmer climates. Storm-Triggered Landslides in Warmer Climates, 1–365. https://doi.org/10.1007/978-3-319-08518-0.

Sánchez-Núñez J.M., Macías J.L., Saucedo R., Zamorano J.J., Novelo D., Mendoza M.E., Torres-Hernández J.R., 2015. Geomorphology, internal structure and evolution of alluvial fans at Motozintla, Chiapas, Mexico. Geomorphology, 230, 1–12. https://doi.org/10.1016/j.geomorph.2014.10.003.

Seale C., Redfern T., Chatfield P., Luo C., Dempsey K., 2022. Coastline detection in satellite imagery: A deep learning approach on new benchmark data. Remote Sensing of Environment, 278, 113044. https://doi.org/10.1016/j.rse.2022.113044.

Sivapriya M.S., Suresh S., 2023. ViT-DexiNet: a vision transformer-based edge detection operator for small object detection in SAR images. International Journal of Remote Sensing, 44, 7057–7084. https://doi.org/10.1080/01431161.2023.2277167.

Stoian A., Poulain V., Inglada J., Poughon V., Derksen D., 2019. Land cover maps production with high resolution satellite image time series and convolutional neural networks: Adaptations and limits for operational systems. Remote Sensing, 11, 1–26. https://doi.org/10.3390/rs11171986.

Studer M., Falbel M.D., 2019. Keras Package Manual. Su B., Li Y., Han Z., Ma Y., Wang W., Ruan B., Guo W., Xie W., Tan S., 2024. Topography-based and vectorized algorithm for extracting physical quantities in 3D-SPH form and its application in debris-flow entrainment modeling. Engineering Geology, 340. https://doi.org/10.1016/j.enggeo.2024.107693.

Thao V.B., Huong N.T.T., 2022. Assessing the morphological characteristics of stream basins on the formation of flash floods in the mountainous region of Northern Vietnam. Journal of Irrigation Science and Technology, 70, 1–16.

Tram T., Bac K., Linh T., Huyen T., Hoang N., Le V.H., 2024. Deep learning models for monitoring landscape changes in a UNESCO Global Geopark. Journal of Environmental Management, 354, 120497. https://doi.org/10.1016/j.jenvman.2024.120497.

Tran Anh T., Pham Viet H., Tran Thi T., Nguyen Thi Anh N., Nguyen Van D., Pham Tuan H., Tran Van P., 2024. Landslide susceptibility in Phuoc Son, Quang Nam: A deep learning approach. Vietnam Journal of Earth Sciences, 47(1), 39–57. https://doi.org/10.15625/2615-9783/21658.

Tran T.H., 2013. Relationship between geomorphology and landslide in Lao Cai province. VNU Journal of Science, 3, 35–44.

Van Thang N., Sato G., Wakai A., Hung H.V., Ducmanh N., Kimura T., Yamasaki T., Tosa S., Hayashi K., Watanabe A., Ozaki T., Asai N., Kitamura N., 2021. Landslide investigation results in SAPA town, Lao CAI province, Vietnam in December 2019. Journal of Disaster Research, 16, 547–555. https://doi.org/10.20965/JDR.2021.P0547.

Varma Byrraju S., 2019. Landslide Detection Using Remote Sensing Methods A Review of Current Techniques. (Master's thesis). Retrieved from https://scholarcommons.sc.edu/etd/5105.

Vu B.T., Nguyen T.T.H., 2023. Impact mechanisms and morphological characteristics of Nam Pam debris flow - flash flood catchment. Ministry of Science and Technology, Vietnam, 65, 58–66. https://doi.org/10.31276/vjst.65(6).58-66.

Wang L., Yang Y., Min R., Chakradhar S., 2017. Accelerating Deep Neural Network Training with Inconsistent Stochastic Gradient Descent. Computer Science, Machine Learning, Cornell University, 3, 12.

Wen S., Yuan Y., Chen J., 2023. A Vision Detection Scheme Based on Deep Learning in a Waste Plastics Sorting System. Applied Sciences (Switzerland), 13. https://doi.org/10.3390/app13074634.

Yang H., Liu J., Sun H., You Y., Zhao W., Yang D., 2024. Evolution characteristics of post-fire debris flow in Xiangjiao gully, Muli County. Catena, 246. https://doi.org/10.1016/j.catena.2024.108353.

Yang S., Mei G., Zhang Y., 2024. Susceptibility analysis of glacier debris flow by investigating glacier changes based on remote sensing imagery and deep learning: A case study. Natural Hazards Research, 4, 539–549. https://doi.org/10.1016/j.nhres.2023.12.013.

Yasir M., Liu S., Pirasteh S., Xu M., Sheng H., Wan J., de Figueiredo F.A.P., Aguilar F.J., Li J., 2024a. YOLOShipTracker: Tracking ships in SAR images using lightweight YOLOv8. International Journal of Applied Earth Observation and Geoinformation, 134, 104137. https://doi.org/10.1016/j.jag.2024.104137.

Yasir M., Shanwei L., Mingming X., Jianhua W., Nazir S., Islam Q.U., Dang K.B., 2024b. SwinYOLOv7: Robust ship detection in complex synthetic aperture radar images. Applied Soft Computing, 160, 111704. https://doi.org/10.1016/j.asoc.2024.111704.

Yousefi S., Imaizumi F., Takayama S., 2025. Spatial distribution and transport characteristics of debris flow sediment using high resolution UAV images in the Ohya debris flow fan. Geomorphology, 469, 109533. https://doi.org/10.1016/j.geomorph.2024.109533.

Zhang J., Cheng H., Wu S., Wu G., Tuo R., Liu W., Xinglong Feng, Li Z., 2024. Risk analysis of underground debris flows in mines based on a coupled weighted Bayesian network. International Journal of Disaster Risk Reduction, 114. https://doi.org/10.1016/j.ijdrr.2024.104922.

Zhang L., Zhang J., Ming Z., Li H., Chen R., Jia Y., 2024. Effects of changing spatial scale on debris-flow hazard assessment: A case study in the Dadu River basin, China. Science of the Total Environment, 954. https://doi.org/10.1016/j.scitotenv.2024.176482.

Zhang P., Liu X., Shu H., 2023. Hazard assessment of debris flow by using FLO-2D and hazard matrix: a case study of Qingshui Gully in the southern Gansu Province, China. Desalination and Water Treatment, 315, 650–662. https://doi.org/10.5004/dwt.2023.30108.

Zhou W., Zhou Y., Liu R., Yin H., Nie H., 2025. Predictive modeling of river blockage severity from debris flows: Integrating statistical and machine learning approaches with insights from Sichuan Province, China. Catena, 248. https://doi.org/10.1016/j.catena.2024.108576.