All Issue

2025 Vol.58, Issue 12S-4

Special Issue: Blue-Green-Grey 도시홍수 방어

FLO-SRNet: A deep-learning super-resolution model for 1D–2D urban flood simulations

FLO-SRNet: 1D–2D 연계 도시 침수 모의 초해상화 모형

Hyeonjin Choia
,
Hyuna Woob
,
Jieun Pyoc
,
Hyungon Ryud
,
Dong Sop Rheee
,
Seong Jin Nohf*
최 현진a
,
우 현아b
,
표 지은c
,
유 현곤d
,
이 동섭e
,
노 성진f*
aPh.D. Course Student, Department of Civil Engineering, Kumoh National Institute of Technology, Gumi, Korea
bPh.D. Course Student, Department of Civil Engineering, Kumoh National Institute of Technology, Gumi, Korea
cUndergraduate Student, Department of Civil Engineering, Kumoh National Institute of Technology, Gumi, Korea
dKorea Lead, NVIDIA AI Technology Center, Seoul, Korea
eResearch Fellow, Department of Hydro Science and Engineering Research, Korea Institute of Civil Engineering and Building Technology, Goyang, Korea
fAssociate Professor, Department of Civil Engineering, Kumoh National Institute of Technology, Gumi, Korea
a국립금오공과대학교 토목공학과 박사과정
b국립금오공과대학교 토목공학과 박사과정
c국립금오공과대학교 토목공학과 학사과정
dNVIDIA AI Technology Center Korea Lead
e한국건설기술연구원 수자원하천연구본부 연구위원
f금오공과대학교 토목공학과 부교수
*Corresponding Author
31 December 2025. pp. 1589-1603
PDF XML Citation
Abstract

Climate change-related rainfall intensification together with ongoing urban development continues to elevate urban flood risk, making the rapid production of high-resolution inundation maps a practical necessity. However, physics-based models at fine grid spacing demand heavy computation, which constrains near-real-time use. Therefore, we propose a deep learning–based super-resolution (SR) approach that converts low-resolution outputs of 1D–2D coupled urban flood simulations to high-resolution maps. We develop FLO-SRNet, an enhanced ResUNet architecture with GAN fine-tuning, and evaluate it in the Oncheon-cheon catchment. From historical events, we construct 1,000 training pairs (2014 event) and 300 validation pairs (2021 event) to ensure temporal independence. Evaluation considers conventional image metrics (PSNR, SSIM, RMSE) as well as perceptual similarity (LPIPS) and flood-classification accuracy (F1-Score). Relative to a ResUNet baseline, FLO-SRNet improves PSNR/SSIM by 23.0%/12.0% with RMSE/LPIPS reductions of 47.1%/78.9% at ×2 upscaling, and delivers 10.5%/13.9% PSNR/SSIM gains with 28.6%/60.7% RMSE/LPIPS reductions at ×4 upscaling. Classification improves by 85.0% (mild) and 50.0% (severe) at ×2 upscaling, and 144.4% (mild) and 76.0% (severe) at ×4 upscaling. The benefit of GAN fine-tuning depends on generator quality, with FLO-SRNet+GAN yielding the best overall performance. This study provides a systematic validation of SR for outputs from 1D–2D coupled urban flood models, offering a practical pathway to near-real-time flood prediction and disaster response systems.

Keywords
Urban flood
Deep learning
Super-Resolution
High-Resolution
Oncheon-cheon catchment

기후변화와 도시화로 인한 도시 홍수 위험이 증가하면서 고해상도 침수지도의 신속한 생성이 핵심 과제로 대두되고 있다. 반면, 기존 물리 기반 모형은 고해상도 해석 시 연산 비용이 증가하여 준실시간 예측에 제약이 있다. 본 연구는 1D-2D 연계 도시 침수 물리 모의 결과를 대상으로 딥러닝 기반 초해상화 기술을 적용하여 저해상도 침수지도를 고해상도로 변환하는 방법론을 제시하였다. ResUNet 기반 아키텍처를 개선하고 GAN fine-tuning 기법을 연계한 FLO-SRNet 모형을 개발하고 부산 온천천 유역에 대해 적용성을 평가하였다. 온천천 유역의 과거 사상 중 2014년 호우 사상에서 1,000쌍의 학습 데이터를, 2021년 호우 사상에서 300쌍의 검증 데이터를 구축하고, 기존 화질 평가(PSNR, SSIM)와 함께 지각적 유사도(LPIPS) 및 침수 영역 분류 정확도(F1-Score) 등 침수 특성 평가 체계를 적용하였다. 연구 결과, FLO-SRNet은 기존 ResUNet 대비 ×2 업스케일링에서 PSNR 23.0%, SSIM 12.0% 향상, RMSE 47.1%, LPIPS 78.9% 감소, ×4 업스케일링에서 PSNR 10.5%, SSIM 13.9% 향상, RMSE 28.6%, LPIPS 60.7% 감소를 달성하였다. 침수 분류 성능은 ×2 업스케일링에서 경미 침수 85.0%, 심각 침수 50.0% 향상, ×4 업스케일링에서 경미 침수 144.4%, 심각 침수 76.0% 향상을 보였다. GAN fine-tuning은 생성자 모형의 기본 성능에 따라 차별화된 효과를 보였으며, FLO-SRNet+GAN이 모든 평가 지표에서 우수함을 확인하였다. 본 연구는 복잡한 도시 배수 체계가 반영된 1D-2D 연계 물리 모형 결과에 대한 초해상화 성능을 검증하였으며, 실시간 침수 예측 및 재해 대응 시스템 구축을 위한 실용적 기술을 제공할 것으로 기대된다.

키워드
도시 침수
딥러닝
초해상화
고해상도
온천천 유역
References
1

Akkala, A., Boubrahimi, S.F., Hamdi, S.M., Hosseinzadeh, P., and Nassar, A. (2025). “Spatio-temporal graph neural networks for streamflow prediction in the upper Colorado Basin.” Journal of Hydrology, Vol. 12, No. 3, 60.

10.3390/hydrology12030060
2

Andreadis, K.M., Wing, O.E.J., Colven, E., Gleason, C.J., Bates, P.D., and Brown, C.M. (2022). “Urbanizing the floodplain: Global changes of imperviousness in flood-prone areas.” Environmental Research Letters, Vol. 17, No. 10, 104024.

10.1088/1748-9326/ac9197
3

Bentivoglio, R., Isufi, E., Jonkman, S.N., and Taormina, R. (2025). “Multi-scale hydraulic graph neural networks for flood modelling.” Natural Hazards and Earth System Sciences, Vol. 25, No. 1, pp. 335-351.

10.5194/nhess-25-335-2025
4

Bermúdez, M., Ntegeka, V., Wolfs, V., and Willems, P. (2018). “Development and comparison of two fast surrogate models for urban pluvial flood simulations.” Water Resources Management, Vol. 32, No. 8, pp. 2801-2815.

10.1007/s11269-018-1959-8
5

Bhattarai, Y., Bista, S., Talchabhadel, R., Duwal, S., and Sharma, S. (2024). “Rapid prediction of urban flooding at street-scale using physics-informed machine learning-based surrogate modeling.” Total Environment Advances, Vol. 12, 200116.

10.1016/j.teadva.2024.200116
6

Cheng, J., Kuang, Q., Shen, C., Liu, J., Tan, X., and Liu, W. (2020). “ResLap: Generating high-resolution climate prediction through image super-resolution.” IEEE Access, Vol. 8, pp. 39623-39634.

10.1109/ACCESS.2020.2974785
7

Choi, H., Lee, S., Woo, H., and Noh, S.J. (2023a). “High-resolution urban flood modeling using cellular automata-based WCA2D in the Oncheon-cheon catchment in Busan, South Korea.” Journal of Civil and Environmental Engineering Research, Vol. 43, No. 5, pp. 587-599.

8

Choi, H., Lee, S., Woo, H., Kim, M., and Noh, S.J. (2023b). “Applying deep learning based super-resolution technique for high-resolution urban flood analysis.” Journal of Korea Water Resources Association, Vol. 56, No. 10, pp. 641-653.

9

Choi, H., Woo, H., Kim, M., Ryu, H., Lee, J.-H., Lee, S., and Noh, S.J. (2025). “FLO-SR: Deep learning-based urban flood super-resolution model.” Journal of Hydrology, Vol. 661, pp. 133529.

10.1016/j.jhydrol.2025.133529
10

Choi, S., and Kim, Y. (2022). “Rad-cGAN v1.0: Radar-based precipitation nowcasting model with conditional generative adversarial networks for multiple dam domains.” Geoscientific Model Development, Vol. 15, No. 15, pp. 5967-5985.

10.5194/gmd-15-5967-2022
11

Choi, S., and Kim, Y. (2025). “Analyzing the predictive power of GAN-based precipitation prediction model against a process-based model: a case study of the Yeongsangang River basin in South Korea.” Journal of Korea Water Resources Association, Vol. 58, No. 1, pp. 53-66.

10.3741/JKWRA.2025.58.1.53
12

Conde, M.V., Choi, U.-J., Burchi, M., and Timofte, R. (2022). “Swin2SR: SwinV2 transformer for compressed image super-resolution and restoration.” Proceedings of the European Conference on Computer Vision (ECCV 2022) Workshops, Springer, Tel Aviv, Israel, Vol. 13802, pp. 669-687.

10.1007/978-3-031-25063-7_42
13

Contreras, M.T., Gironás, J., and Escauriaza, C. (2020). “Forecasting flood hazards in real time: A surrogate model for hydrometeorological events in an Andean watershed.” Natural Hazards and Earth System Sciences, Vol. 20, No. 12, pp. 3261-3277.

10.5194/nhess-20-3261-2020
14

Demiray, B.Z., Sit, M., and Demir, I. (2021). “D-SRGAN: DEM super-resolution with generative adversarial networks.” SN Computer Science, Vol. 2, No. 1, 48.

10.1007/s42979-020-00442-2
15

do Lago, C.A.F., Giacomoni, M.H., Bentivoglio, R., Taormina, R., Gomes, M.N., and Mendiondo, E.M. (2023). “Generalizing rapid flood predictions to unseen urban catchments with conditional generative adversarial networks.” Journal of Hydrology, Vol. 618, 129276.

10.1016/j.jhydrol.2023.129276
16

Donnelly, J., Daneshkhah, A., and Abolfathi, S. (2024). “Physics-informed neural networks as surrogate models of hydrodynamic simulators.” Science of The Total Environment, Vol. 912, 168814.

10.1016/j.scitotenv.2023.168814
17

Gao, W., Liao, Y., Chen, Y., Lai, C., He, S., and Wang, Z. (2024). “Enhancing transparency in data-driven urban pluvial flood prediction using an explainable CNN model.” Journal of Hydrology, Vol. 645, 132228.

10.1016/j.jhydrol.2024.132228
18

Golla, S., Murukesh, M., and Kumar, P. (2024). “Comparative assessment of image super-resolution techniques for spatial downscaling of gridded rainfall data.” SN Computer Science, Vol. 5, No. 3, 312.

10.1007/s42979-024-02653-3
19

He, J., Zhang, L., Xiao, T., Wang, H., and Luo, H. (2023). “Deep learning enables super-resolution hydrodynamic flooding process modeling under spatiotemporally varying rainstorms.” Water Research, Vol. 239, 120057.

10.1016/j.watres.2023.120057
20

Henonin, J., Russo, B., Mark, O., and Gourbesville, P. (2013). “Real-time urban flood forecasting and modelling - A state of the art.” Journal of Hydroinformatics, Vol. 15, No. 3, pp. 717-736.

10.2166/hydro.2013.132
21

Hwang, S.-G., and Lee, J.-H. (2023). “Super-resolution technique of underwater image based on lightweight convolutional neural network for marine accident cause analysis.” Journal of Korean Institute of Intelligent Systems, Vol. 33, No. 2, pp. 127-132.

10.5391/JKIIS.2023.33.2.127
22

Ivanov, V.Y., Xu, D., Dwelle, M.C., Sargsyan, K., Wright, D.B., Katopodes, N., Kim, J., Tran, V.N., Warnock, A., and Fatichi, S. et al. (2021). “Breaking down the computational barriers to real-time urban flood forecasting.” Geophysical Research Letters, Vol. 48, No. 20, e2021GL093585.

10.1029/2021GL093585
23

Jia, Y., Ge, Y., Chen, Y., Li, S., Heuvelink, G.B.M., and Ling, F. (2019). “Super-resolution land cover mapping based on the convolutional neural network.” Remote Sensing, Vol. 11, No. 15, 1815.

10.3390/rs11151815
24

Jian, J., He, S., Liu, W., Liu, S., and Guo, L. (2025). “A refined method for the simulation of catchment rainfall-runoff based on satellite-precipitation downscaling.” Journal of Hydrology, Vol. 653, 132795.

10.1016/j.jhydrol.2025.132795
25

Jo, W., Park, S., Kim, Y., Jung, S., and Sim, C. (2024). “Super-resolution model based on wavelet domain loss for improving wind speed forecasting accuracy.” The Transactions of the Korea Information Processing Society, Vol. 13, No. 12, pp. 710-718.

26

Karimanzira, D. (2024). “Mass conservative time-series GAN for synthetic extreme flood-event generation: Impact on probabilistic forecasting models.” Stats, Vol. 7, No. 3, pp. 808-826.

10.3390/stats7030049
27

Kim, H., Lim, C., and Tak, S. (2024). “Improved recall of plant disease detection model using image super resolution.” Journal of KIISE, Vol. 51, No. 2, pp. 125-130.

10.5626/JOK.2024.51.2.125
28

Kim, Y., and Hong, S. (2022). “Very short-term rainfall prediction using ground radar observations and conditional generative adversarial networks.” IEEE Transaction on Geoscience and Remote Sensing, Vol. 60, pp. 1-8.

10.1109/TGRS.2021.3108812
29

Kwon, O.-S. (2023). “Vehicle detection algorithm using super resolution based on deep residual dense block for remote sensing images.” Journal of Broadcast Engineering, Vol. 28, No. 1, pp. 124-131.

10.5909/JBE.2023.28.1.124
30

Ledig, C., Theis, L., Huszar, F., Caballero, J., Cunningham, A., Acosta, A., Aitken, A., Tejani, A., Totz, J., and Wang, Z. et al. (2017). “Photo-realistic single image super-resolution using a generative adversarial network.” Proceeding of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Honolulu, HI, U.S., pp. 102-114.

10.1109/CVPR.2017.19
31

Lee, J.-H., Lee, S., Kim, B., Choi, H., and Noh, S.J. (2025a). “Evaluating the effects of spatial resolution on 2D pluvial flood modeling in urban built environments.” Journal of Flood Risk Management, Vol. 18, No. 3, e70105.

10.1111/jfr3.70105
32

Lee, J.-W., Yoon, S.-W., and Lee, K.-C. (2025b). “Transformer-based deep learning models for ERS SAR image super-resolution.” Korean Journal of Remote Sensing, Vol. 41, No. 1, pp. 143-152.

10.7780/kjrs.2025.41.1.12
33

Liang, J., Cao, J., Sun, G., Zhang, K., Van Gool, L., and Timofte, R. (2021). “SwinIR: Image restoration using swin transformer.” Proceeding of the 2021 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), IEEE, Montreal, BC, Canada, pp. 1833-1844.

10.1109/ICCVW54120.2021.00210
34

Lim, B., Son, S., Kim, H., Nah, S., and Lee, K.M. (2017). “Enhanced deep residual networks for single image super-resolution.” Proceeding of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), IEEE, Honolulu, HI, U.S., pp. 1132-1140.

10.1109/CVPRW.2017.151
35

Liu, Q., Du, M., Wang, Y., Deng, J., Yan, W., Qin, C., Liu, M., and Liu, J. (2024). “Global, regional and national trends and impacts of natural floods, 1990-2022.” Bulletin of the World Health Organization, Vol. 102, No. 6, pp. 410-420.

10.2471/BLT.23.29024338812801PMC11132161
36

Maranzoni, A., D’Oria, M., and Rizzo, C. (2023). “Quantitative flood hazard assessment methods: A review.” Journal of Flood Risk Management, Vol. 16, No. 1, e12855.

10.1111/jfr3.12855
37

Noh, S.J., Lee, J.-H., Lee, S., Kawaike, K., and Seo, D.-J. (2018). “Hyper-resolution 1D-2D urban flood modelling using LiDAR data and hybrid parallelization.” Environmental Modelling & Software, Vol. 103, pp. 131-145.

10.1016/j.envsoft.2018.02.008
38

Piadeh, F., Behzadian, K., and Alani, A.M. (2022). “A critical review of real-time modelling of flood forecasting in urban drainage systems.” Journal of Hydrology, Vol. 607, 127476.

10.1016/j.jhydrol.2022.127476
39

Pianforini, M., Dazzi, S., Pilzer, A., and Vacondio, R. (2024). “Real-time flood maps forecasting for dam-break scenarios with a transformer-based deep learning model.” Journal of Hydrology, Vol. 635, 131169.

10.1016/j.jhydrol.2024.131169
40

Ren, H., Pang, B., Zhao, G., Yu, H., Tian, P., and Xie, C. (2025). “Incorporating dynamic drainage supervision into deep learning for accurate real-time flood simulation in urban areas.” Water Research, Vol. 270, 122816.

10.1016/j.watres.2024.122816
41

Saharia, C., Ho, J., Chan, W., Salimans, T., Fleet, D.J., and Norouzi, M. (2023). “Image super-resolution via iterative refinement.” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 45, No. 4, pp. 4713-4726.

42

Sara, U., Akter, M., and Uddin, M.S. (2019). “Image quality assessment through FSIM, SSIM, MSE and PSNR-A comparative study.” Journal of Computer and Communications, Vol. 7, No. 3, pp. 8-18.

10.4236/jcc.2019.73002
43

Song, W., Guan, M., Guo, K., and Yu, D. (2025). “Rapid flood inundation mapping by integrating deep learning-based image super-resolution with coarse-grid hydrodynamic modeling.” Engineering Applications of Computational Fluid Mechanics, Vol. 19, No. 1, 2481115.

10.1080/19942060.2025.2481115
44

Wang, J., Yue, Z., Zhou, S., Chan, K.C.K., and Loy, C.C. (2023). “Exploiting diffusion prior for real-world image super-resolution.” International Journal of Computer Vision, Vol. 132, pp. 5929-5949.

10.1007/s11263-024-02168-7
45

Wang, X., Yi, J., Guo, J., Song, Y., Lyu, J., Xu, J., Yan, W., Zhao, J., Cai, Q., and Min, H. (2022). “A review of image super-resolution approaches based on deep learning and applications in remote sensing.” Remote Sensing, Vol. 14, No. 21, 5423.

10.3390/rs14215423
46

Wang, Z., Bovik, A., Sheikh, H., and Simoncelli, E. (2004). “Image quality assessment: From error visibility to structural similarity.” IEEE Transactions on Image Processing, Vol. 13, pp. 600-612.

10.1109/TIP.2003.819861
47

Yang, F., Ding, W., Zhao, J., Song, L., Yang, D., and Li, X. (2024). “Rapid urban flood inundation forecasting using a physics-informed deep learning approach.” Journal of Hydrology, Vol. 643, 131998.

10.1016/j.jhydrol.2024.131998
48

Yin, Z., Saadati, Y., Hu, B., Leon, A.S., Amini, M.H., and McDaniel, D. (2024). “Fast high-fidelity flood inundation map generation by super-resolution techniques.” Journal of Hydroinformatics, Vol. 26, pp. 319-336.

10.2166/hydro.2024.228
49

Yoon, S., Shin, H., and Heo, J.-Y. (2023). “Radar-based rainfall prediction using generative adversarial network.” Journal of Korea Water Resources Association, Vol. 56, No. 8, pp. 471-484.

10.3741/JKWRA.2023.56.8.471
50

Zhang, Q., Li, C., Wen, D., Kang, J., Chen, T., Zhang, B., Hu, Y., Yin, J., and Slater, L. (2025). “Global South shows higher urban flood exposures than the Global North under current and future scenarios.” Communications Earth & Environment, Vol. 6, No. 1, 594.

10.1038/s43247-025-02585-7
51

Zhang, Y., Li, K., Li, K., Wang, L., Zhong, B., and Fu, Y. (2018a). “Image super-resolution using very deep residual channel attention networks.” Proceedings of the European Conference on Computer Vision (ECCV 2018), Springer, Munich, Germany, Vol. 11211, pp. 294-310.

10.1007/978-3-030-01234-2_18
52

Zhang, Z., Liu, Q., and Wang, Y. (2018b). “Road Extraction by Deep Residual U-Net.” IEEE Geoscience and Remote Sensing Letters, Vol. 15, No. 5, pp. 749-753.

10.1109/LGRS.2018.2802944
Information
  • Publisher :KOREA WATER RESOURECES ASSOCIATION
  • Publisher(Ko) :한국수자원학회
  • Journal Title :Journal of Korea Water Resources Association
  • Journal Title(Ko) :한국수자원학회 논문집
  • Volume : 58
  • No :12
  • Pages :1589-1603
  • Received Date : 2025-09-26
  • Revised Date : 2025-10-21
  • Accepted Date : 2025-10-24
  • DOI :https://doi.org/10.3741/JKWRA.2025.58.S-4.1589
Journal Informaiton Journal of Korea Water Resources Association Journal of Korea Water Resources Association
Online Submission
Archives
  • scopus
  • KCI-수자원
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close