Developing a Machine Learning-Based Framework for Roadway Vulnerability and Impact Assessment Using Aerial Imagery

Abstract

This study developed a machine learning-based framework for assessing roadway vulnerability and impacts in hurricane-prone regions, utilizing remote sensing techniques. To quantify the immediate and consistent impacts of hurricanes on the roadway network, the study developed two key metrics: the road closure impact index (RCII) and the roadway vulnerability index (RVI). The RCII assesses the severity of roadway closures by analyzing detected bounding boxes from high-resolution aerial imagery, offering insight into the spatial extent and severity of disruptions caused by each storm. In contrast, the RVI evaluates the consistency of roadway closure patterns across multiple events, revealing vulnerabilities within the transportation infrastructure through geospatial analysis. Also, by leveraging aerial imagery, remote sensing technology, and advanced machine learning models, the study assessed the impacts of Hurricanes Idalia and Debby on Taylor County, Florida, effectively classifying county roadway conditions in their aftermath into three categories: open, partially closed, and fully closed. Findings indicate that Hurricane Idalia caused significant structural damage due to wind and storm surge, while Hurricane Debby led to prolonged flooding and subsequent road submersion. By comparing the impacts of these two hurricanes, the study highlights the critical role of integrating machine learning, geospatial analysis, and remote sensing for enhanced disaster preparedness and response strategies. Ultimately, this framework provides critical insights for improving infrastructure resilience and planning efforts in coastal communities vulnerable to extreme weather events.

Keywords

Geospatial analysis / Hurricane impact assessment / Machine learning / Roadway closure impact index / Roadway vulnerability

References

[1] Al Shafian, S., and D. Hu. 2024. Integrating machine learning and remote sensing in disaster management: A decadal review of post-disaster building damage assessment. Buildings 14(8): Article 2344.

[2] Antwi, R.B., S. Takyi, A. Karaer, E.E. Ozguven, M. Kimollo, R. Moses, M.A. Dulebenets, and T. Sando. 2024a. Automated geographic-information-system-based framework for detecting crosswalk changes from bi-temporal high-resolution aerial images. Transportation Research Record 2679(4): Article 03611981241292333.

[3] Antwi RB, Takyi S, Karaer A, Ozguven EE, Moses R, Dulebenets MA, Sando T. Detecting school zones on Florida’s public roadways using aerial images and artificial intelligence (AI2). Transportation Research Record, 2024, 2678: 622-636

[4] Buslaev, A., S. Seferbekov, V. Iglovikov, and A. Shvets. 2018. Fully convolutional network for automatic road extraction from satellite imagery. In Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 18–22 June 2018, Salt Lake City, UT, USA.

[5] Cristian, A. 2024. Post-disaster recovery and environmental monitoring using GIS and GeoAI: Building damage assessment insights from McDonough County. https://doi.org/10.13140/RG.2.2.25657.17760.

[6] Drejza, S., P. Bernatchez, G. Marie, and S. Friesinger. 2019. Quantifying road vulnerability to coastal hazards: Development of a synthetic index. Ocean and Coastal Management 181. https://doi.org/10.1016/j.ocecoaman.2019.104894.

[7] Dutta Roy, A., D.A. Karpowicz, I. Hendy, S.M. Rog, M.S. Watt, R. Reef, E.N. Broadbent, E.F. Asbridge, et al. 2024. Current status of remote sensing for studying the impacts of hurricanes on mangrove forests in the coastal United States. Remote Sensing 16(19): Article 3596.

[8] Ferguson, N., G. Boura, A. Gray, E. Patelli, and E. Tubaldi. 2023. Vulnerability of the Scottish road network to flooding. https://pure.strath.ac.uk/ws/portalfiles/portal/201611960/Ferguson-etal-2023-Vulnerability-of-the-Scottish-Road-Network-to-Flooding.pdf. Accessed 13 Oct 2024.

[9] Goodfellow I, Bengio Y, Courville A. Deep learning, 2016, The MIT Press

[10] Hassan, S.A., H.A. Amlan, N.E. Alias, M.A. Ab-Kadir, and N.S.A. Sukor. 2022. Vulnerability of road transportation networks under natural hazards: A bibliometric analysis and review. International Journal of Disaster Risk Reduction 83: Article 103393.

[11] Hauptman, L., D. Mitsova, and T.R. Briggs. 2024. Hurricane Ian damage assessment using aerial imagery and LiDAR: A case study of Estero Island, Florida. Journal of Marine Science and Engineering 12. https://doi.org/10.3390/jmse12040668.

[12] He, K., X. Zhang, S. Ren, and J. Sun. 2016. Deep residual learning for image recognition. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 27–30 June 2016, Las Vegas, NV, USA, 770–778.

[13] Jena, R., B. Pradhan, and G. Beydoun. 2020. Earthquake vulnerability assessment in Northern Sumatra Province by using a multi-criteria decision-making model. International Journal of Disaster Risk Reduction 46: Article 101518.

[14] Karaer, A., M. Chen, M. Gazzea, M. Ghorbanzadeh, T. Abichou, R. Arghandeh, and E.E. Ozguven. 2022. Remote sensing-based comparative damage assessment of historical storms and hurricanes in Northwestern Florida. International Journal of Disaster Risk Reduction 72: Article 102857.

[15] Kathuria, A. 2019. What’s new in YOLO v3? 2018. https://towardsdatascience.com/yolo-v3-object-detection-53fb7d3bfe6b. Accessed 9 Sept 2024.

[16] Li X, Zhang Z, Lv S, Pan M, Ma Q, Yu H. Road extraction from high spatial resolution remote sensing image based on multi-task key point constraints. IEEE Access, 2021, 9: 95896-95910

[17] Linardos V, Drakaki M, Tzionas P, Karnavas YL. Machine learning in disaster management: Recent developments in methods and applications. Machine Learning and Knowledge Extraction, 2022, 4(2): 446-473

[18] Liu, R., J. Wu, W. Lu, Q. Miao, H. Zhang, X. Liu, Z. Lu, and L. Li. 2024. A review of deep learning-based methods for road extraction from high-resolution remote sensing images. Remote Sensing 16(12): Article 2056.

[19] Loggins, R., and W. Wallace. 2015. Rapid assessment of hurricane damage and disruption to interdependent civil infrastructure systems. Journal of Infrastructure Systems 21: Article 4015005.

[20] Lu, Q.-C., and Z.-R. Peng. 2011. Vulnerability analysis of transportation network under scenarios of sea level rise. Transportation Research Record 2263. https://doi.org/10.3141/2263-19.

[21] Lu X, Zhong Y, Zheng Z, Liu Y, Zhao J, Ma A, Yang J. Multi-scale and multi-task deep learning framework for automatic road extraction. IEEE Transactions on Geoscience and Remote Sensing, 2019, 57: 9362-9377

[22] Ma, H., N. Lu, L. Ge, Q. Li, X. You, and X. Li. 2013. Automatic road damage detection using high-resolution satellite images and road maps. In Proceedings of the 2013 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 21–26 July 2013, Melbourne, VIC, Australia, 3718–3721.

[23] Mathias, J.D., S.S. Clark, N. Onat, and T.P. Seager. 2018. An integrated dynamical modeling perspective for infrastructure resilience. Infrastructures 3(2): Article 11.

[24] Mitsova, D., M. Escaleras, A. Sapat, A.M. Esnard, and A.J. Lamadrid. 2019. The effects of infrastructure service disruptions and socio-economic vulnerability on hurricane recovery. Sustainability 11(2): Article 516.

[25] Mostafavi, A., J. Padgett, L. Dueñas-Osorio, E. Sutley, T. Norton, H. Lester, H. Wang, S. Dong, et al. 2022. Hurricane Harvey infrastructure resilience investigation. https://doi.org/10.17603/ds2-gcrf-h607.

[26] Mukherjee, M., K. Abhinay, M.M. Rahman, S. Yangdhen, S. Sen, B.R. Adhikari, R. Nianthi, S. Sachdev, and R. Shaw. 2023. Extent and evaluation of critical infrastructure, the status of resilience and its future dimensions in South Asia. Progress in Disaster Science 17: Article 100275.

[27] Mustakim, M.M. 2023. Road damage detection based on deep learning. https://doi.org/10.13140/RG.2.2.22523.28967.

[28] National Oceanic and Atmospheric Administration. 2023. 2023 Hurricane Idalia images. https://noaa-eri-pds.s3.amazonaws.com/index.html#2023_Hurricane_Idalia/20230902a_RGB/. Accessed 23 Jun 2024.

[29] Nguyen, L., A. Slobodzian, C. Villiers, and S. Tsegaye. 2022. Interdependencies of lifelines: A case study of transportation infrastructure under hurricane impacts. In Proceedings of the Canadian Society of Civil Engineering Annual Conference 2021, ed. S. Walbridge, M. Nik-Bakht, K.T.W. Ng, M. Shome, M.S. Alam, A. el Damatty, and G. Lovegrove, 1–10. Singapore: Springer.

[30] Pouryari M, Ardakani A, Hassani N. A multi-criteria vulnerability of urban transportation systems analysis against earthquake considering topological and geographical method: A case study. Iranian Journal of Science and Technology, Transactions of Civil Engineering, 2021, 46: 2147-2160

[31] Redmon, J., and A. Farhadi. 2018. YOLOv3: An incremental improvement. https://arxiv.org/abs/1804.02767. Accessed 12 Jul 2023.

[32] Redmon, J., S. Divvala, R. Girshick, and A. Farhadi. 2016. You Only Look Once: Unified, real-time object detection. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition, 27–30 June 2016, Las Vegas, NV, USA, 779–788.

[33] Robinson AC. Geographic Information Systems (GIS) for disaster management, 2020, 2New York. Taylor and Francis

[34] Ruseruka C, Mwakalonge J, Comert G, Siuhi S, Perkins J. Road condition monitoring using vehicle built-in cameras and GPS sensors: A deep learning approach. Vehicles, 2023, 5: 931-948

[35] Shafique, A., G. Cao, Z. Khan, M. Asad, and M. Aslam. 2022. Deep learning-based change detection in remote sensing images: A review. Remote Sensing 14(4): Article 871.

[36] Singh P, Sinha VSP, Vijhani A, Pahuja N. Vulnerability assessment of urban road network from urban flood. International Journal of Disaster Risk Reduction, 2018, 28: 237-250

[37] Takyi, S., R.B. Antwi, E.E. Ozguven, L. Okine, and R. Moses. 2025. Towards sustainable and resilient infrastructure: Hurricane-induced roadway closure and accessibility assessment in Florida using machine learning. Sustainability 17: Article 3909.

[38] Tampa Free Press. 2024. Hurricane Debby leaves severe damage in Florida; Sen. Rubio and Rep. Cammack urge full federal assistance. https://www.tampafp.com/hurricane-debby-leaves-severe-damage-in-florida-sen-rubio-and-rep-cammack-urge-full-federal-assistance/#google_vignette. Accessed 14 Oct 2024.

[39] Tsang, S.H. 2018. Review: YOLOv3—You Only Look Once (object detection). https://towardsdatascience.com/ review-yolov3-you-only-look-once-object-detection-eab75d 7a1ba6. Accessed 25 Jun 2023.

[40] United States Census Bureau. 2022. https://data.census.gov/profile/Taylor_County,_Florida?g=050XX00US12123. Accessed 28 Oct 2024.

[41] Van Westen CJShroder J, Bishop MP. Remote sensing and GIS for natural hazards assessment and disaster risk management. Treatise on geomorphology, 2013, Amsterdam. Elsevier: 259-298

[42] Vetrivel, A., N. Kerle, M. Gerke, F. Nex, and G. Vosselman. 2016. Towards automated satellite image segmentation and classification for assessing disaster damage using data-specific features with incremental learning. In Proceedings of GEOBIA 2016: Solutions and synergies, 14–16 September 2016, Enschede, Netherlands.

[43] Yang L, Cervone G. Analysis of remote sensing imagery for disaster assessment using deep learning: A case study of flooding event. Soft Computing, 2019, 23: 13393-13408

[44] Yuan, F., W. Mobley, H. Farahmand, and Y. Xu. 2021. Predicting road flooding risk with machine learning approaches using crowdsourced reports and fine-grained traffic data. https://doi.org/10.48550/arXiv.2108.13265.

[45] Zhang, M., M. Xu, Z. Wang, and C. Lai. 2021. Assessment of the vulnerability of road networks to urban waterlogging based on a coupled hydrodynamic model. Journal of Hydrology 603: Article 127105.

[46] Zhong, H., and D. Liang. 2024. A study of road closure due to rainfall and flood zone based on logistic regression. International Journal of Disaster Risk Reduction 102: Article 104291.

[47] Zhou M, Sui H, Chen S, Liu J, Shi W, Chen X. Large-scale road extraction from high-resolution remote sensing images based on a weakly-supervised structural and orientational consistency constraint network. ISPRS Journal of Photogrammetry and Remote Sensing, 2022, 193: 234-251

[48] Zhu, Y.J., Y. Hu, and J.M. Collins. 2020. Estimating road network accessibility during a hurricane evacuation: A case study of hurricane Irma in Florida. Transportation Research Part D: Transport and Environment 83: Article 102334.

RIGHTS & PERMISSIONS

The Author(s)