Journal Home Online First Current Issue Archive For Authors Journal Information 中文版

Engineering >> 2023, Volume 21, Issue 2 doi: 10.1016/j.eng.2021.11.021

Estimating Rainfall Intensity Using an Image-Based Deep Learning Model

a College of Civil Engineering and Architecture, Zhejiang University, China
b Department of Civil and Environmental Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China
c KWR Water Research Institute, Nieuwegein 3430 BB, The Netherlands
d Centre for Water Systems, University of Exeter, North Park Road, Exeter EX4 4QF, United Kingdom
e Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, Bangi 43600, Malaysia
f Department of Water Management, Delft University of Technology, Delft 2600 GA, The Netherlands

Received: 2021-07-09 Revised: 2021-09-27 Accepted: 2021-11-15 Available online: 2022-01-27

Next Previous


Urban flooding is a major issue worldwide, causing huge economic losses and serious threats to public safety. One promising way to mitigate its impacts is to develop a real-time flood risk management system; however, building such a system is often challenging due to the lack of high spatiotemporal rainfall data. While some approaches (i.e., ground rainfall stations or radar and satellite techniques) are available to measure and/or predict rainfall intensity, it is difficult to obtain accurate rainfall data with a desirable spatiotemporal resolution using these methods. This paper proposes an image-based deep learning model to estimate urban rainfall intensity with high spatial and temporal resolution. More specifically, a convolutional neural network (CNN) model called the image-based rainfall CNN (irCNN) model is developed using rainfall images collected from existing dense sensors (i.e., smart phones or transportation cameras) and their corresponding measured rainfall intensity values. The trained irCNN model is subsequently employed to efficiently estimate rainfall intensity based on the sensors' rainfall images. Synthetic rainfall data and real rainfall images are respectively utilized to explore the irCNN's accuracy in theoretically and practically simulating rainfall intensity. The results show that the irCNN model provides rainfall estimates with a mean absolute percentage error ranging between 13.5% and 21.9%, which exceeds the performance of other state-of-the-art modeling techniques in the literature. More importantly, the main feature of the proposed irCNN is its low cost in efficiently acquiring high spatiotemporal urban rainfall data. The irCNN model provides a promising alternative for estimating urban rainfall intensity, which can greatly facilitate the development of urban flood risk management in a real-time manner.


Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Fig. 9

Fig. 10

Fig. 11

Fig. 12


[ 1 ] Zheng F, Thibaud E, Leonard M, Westra S. Assessing the performance of the independence method in modeling spatial extreme rainfall. Water Resour Res 2015;51(9):7744–58. link1

[ 2 ] Gülbaz S, Kazezyılmaz-Alhan CM. Experimental investigation on hydrologic performance of lid with rainfall-watershed-bioretention system. J Hydrol Eng 2017;22(1):D4016003. link1

[ 3 ] Wahl T, Jain S, Bender J, Meyers SD, Luther ME. Increasing risk of compound flooding from storm surge and rainfall for major us cities. Nat Clim Chang 2015;5(12):1093–7. link1

[ 4 ] Zhou Q, Leng G, Su J, Ren Y. Comparison of urbanization and climate change impacts on urban flood volumes: importance of urban planning and drainage adaptation. Sci Total Environ 2019;658:24–33. link1

[ 5 ] Westra S, Fowler HJ, Evans JP, Alexander LV, Berg P, Johnson F, et al. Future changes to the intensity and frequency of short-duration extreme rainfall. Rev Geophys 2014;52(3):522–55. link1

[ 6 ] Archer DR, Fowler HJ. Characterising flash flood response to intense rainfall and impacts using historical information and gauged data in Britain. J Flood Risk Manag 2018;11:S121–33. link1

[ 7 ] Alves A, Vojinovic Z, Kapelan Z, Sanchez A, Gersonius B. Exploring trade-offs among the multiple benefits of green–blue–grey infrastructure for urban flood mitigation. Sci Total Environ 2020;703:134980. link1

[ 8 ] Morris ZB, Malone SM, Cohen AR, Weissburg MJ, Bras B. Impact of low-impact development technologies from an ecological perspective in different residential zones of the city of Atlanta, Georgia. Engineering 2018;4(2):194–9. link1

[ 9 ] Fletcher TD, Shuster W, Hunt WF, Ashley R, Butler D, Arthur S, et al. SUDS, LID, BMPS, WSUD and more–the evolution and application of terminology surrounding urban drainage. Urban Water J 2015;12(7):525–42. link1

[10] He M, Chen C, Zheng F, Chen Q, Zhang J, Yan H, et al. An efficient dynamic route optimization for urban flooding evacuation based on cellular automata. Comput Environ Urban Syst 2021;87:101622. link1

[11] Zhao Q, Yao Y, Yao W, Li Z. Real-time precise point positioning-based zenith tropospheric delay for precipitation forecasting. Sci Rep 2018;8(1):7939. link1

[12] Lee O, Sim I, Kim S. Application of the non-stationary peak-over-threshold methods for deriving rainfall extremes from temperature projections. J Hydrol 2020;585:124318. link1

[13] Yao L, Wei W, Chen L. How does imperviousness impact the urban rainfall– runoff process under various storm cases? Ecol Indic 2016;60:893–905. link1

[14] Berg P, Moseley C, Haerter JO. Strong increase in convective precipitation in response to higher temperatures. Nat Geosci 2013;6(3):181–5. link1

[15] Wasko C, Sharma A. Steeper temporal distribution of rain intensity at higher temperatures within Australian storms. Nat Geosci 2015;8(7):527–9. link1

[16] Muller CL, Chapman L, Johnston S, Kidd C, Illingworth S, Foody G, et al. Crowdsourcing for climate and atmospheric sciences: current status and future potential. Int J Climatol 2015;35(11):3185–203. link1

[17] Moron V, Barbero R, Fowler HJ, Mishra V. Storm types in India: linking rainfall duration, spatial extent and intensity. Philos Trans Royal Soc Math Phys Eng Sci 2021;379(2195):20200137. link1

[18] Breinl K, Müller-Thomy H, Blöschl G. Space–time characteristics of areal reduction factors and rainfall processes. J Hydrometeorol 2020;21(4): 671–89. link1

[19] Berkhahn S, Fuchs L, Neuweiler I. An ensemble neural network model for realtime prediction of urban floods. J Hydrol 2019;575:743–54. link1

[20] Yang L, Tian F, Niyogi D. A need to revisit hydrologic responses to urbanization by incorporating the feedback on spatial rainfall patterns. Urban Clim 2015;12:128–40. link1

[21] Jamali B, Bach PM, Deletic A. Rainwater harvesting for urban flood management–an integrated modelling framework. Water Res 2020;171:115372. link1

[22] Xia X, Liang Q, Ming X, Hou J. An efficient and stable hydrodynamic model with novel source term discretization schemes for overland flow and flood simulations. Water Resour Res 2017;53(5):3730–59. link1

[23] Powers JG, Klemp JB, Skamarock WC, Davis CA, Dudhia J, Gill DO, et al. The weather research and forecasting model: overview, system efforts, and future directions. Bull Am Meteorol Soc 2017;98(8):1717–37. link1

[24] Shen M, Chen J, Zhuan M, Chen H, Xu CY, Xiong L. Estimating uncertainty and its temporal variation related to global climate models in quantifying climate change impacts on hydrology. J Hydrol 2018;556:10–24. link1

[25] Lazri M, Labadi K, Brucker JM, Ameur S. Improving satellite rainfall estimation from msg data in northern Algeria by using a multi-classifier model based on machine learning. J Hydrol. 2020;584:124705. link1

[26] Villarini G, Mandapaka PV, Krajewski WF, Moore RJ. Rainfall and sampling uncertainties: a rain gauge perspective. J Geophys Res Atmos 2008;113(D11): D11102. link1

[27] Thorndahl S, Einfalt T, Willems P, Nielsen JE, ten Veldhuis MC, ArnbjergNielsen K, et al. Weather radar rainfall data in urban hydrology. Hydrol Earth Syst Sci 2017;21(3):1359–80. link1

[28] Yong B, Hong Y, Ren LL, Gourley JJ, Huffman GJ, Chen X, et al. Assessment of evolving TRMM-based multisatellite real-time precipitation estimation methods and their impacts on hydrologic prediction in a high latitude basin. J Geophys Res Atmos 2012;117(D9):D09108. link1

[29] Tan ML, Yang X. Effect of rainfall station density, distribution and missing values on swat outputs in tropical region. J Hydrol 2020;584:124660. link1

[30] Voormansik T, Cremonini R, Post P, Moisseev D. Evaluation of the dualpolarization weather radar quantitative precipitation estimation using longterm datasets. Hydrol Earth Syst Sci 2021;25(3):1245–58. link1

[31] Zorzetto E, Marani M. Downscaling of rainfall extremes from satellite observations. Water Resour Res 2019;55(1):156–74. link1

[32] Zheng F, Tao R, Maier HR, See L, Savic D, Zhang T, et al. Crowdsourcing methods for data collection in geophysics: state of the art, issues, and future directions. Rev Geophys 2018;56(4):698–740. link1

[33] Hut R, de Jong S, van de Giesen N. Using umbrellas as mobile rain gauges: prototype demonstration. EGU Gen Assem Conf Abstr 2014:16418. link1

[34] Jiang S, Babovic V, Zheng Y, Xiong J. Advancing opportunistic sensing in hydrology: a novel approach to measuring rainfall with ordinary surveillance cameras. Water Resour Res 2019;55(4):3004–27. link1

[35] Yuan Q, Shen H, Li T, Li Z, Li S, Jiang Y, et al. Deep learning in environmental remote sensing: Achievements and challenges. Remote Sens Environ 2020;241:111716. link1

[36] Reichstein M, Camps-Valls G, Stevens B, Jung M, Denzler J, Carvalhais N, et al. Deep learning and process understanding for data-driven Earth system science. Nature 2019;566(7743):195–204. link1

[37] Ham YG, Kim JH, Luo JJ. Deep learning for multi-year ENSO forecasts. Nature 2019;573(7775):568–72. link1

[38] Pyo J, Park LJ, Pachepsky Y, Baek SS, Kim K, Cho KH. Using convolutional neural network for predicting cyanobacteria concentrations in river water. Water Res 2020;186:116349. link1

[39] Raissi M, Yazdani A, Karniadakis GE. Hidden fluid mechanics: learning velocity and pressure fields from flow visualizations. Science 2020;367(6481): 1026–30. link1

[40] Guo Z, Leitão JP, Simões NE, Moosavi V. Data-driven flood emulation: speeding up urban flood predictions by deep convolutional neural networks. J Flood Risk Manag 2021;14(1):e12684. link1

[41] LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015;521(7553):436–44. link1

[42] He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proc IEEE Conf Comput Vis Pattern Recognit 2016:770–8. link1

[43] Lecun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proc IEEE 1998;86(11):2278–324. link1

[44] Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 2012;25:1097–105. link1

[45] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. 2014. arXiv:1409.1556.

[46] Rawat W, Wang Z. Deep convolutional neural networks for image classification: a comprehensive review. Neural Comput 2017;29(9):2352–449. link1

[47] Aloysius N, Geetha MA. A review on deep convolutional neural networks. In: Proceedings of 2017 International Conference on Communication and Signal Processing (ICCSP); 2017 Apr 6–8; Chennai, India. New York City: IEEE; 2017. link1

[48] Canziani A, Paszke A, Culurciello E. An analysis of deep neural network models for practical applications. 2016. arXiv:1605.07678.

[49] Gupta A, Matta P, Pant B. Graph neural network: Current state of art, challenges and applications. Mater Today Proc 2021;46:10927–32. link1

[50] Deng J, Dong W, Socher R, Li LJ, Li K, Li FF. Imagenet: a large-scale hierarchical image database. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition; 2009 Jun 20–25; Miami, FL, USA. New York City: IEEE; 2009. p. 248–55. link1

[51] Zhang H, Patel VM. Density-aware single image de-raining using a multistream dense network. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition; 2018 Jun 19–21; Salt Lake City, UT, USA. New York City: IEEE; 2018. p. 695–704. link1

[52] How to add rain to a photo with photoshop [Internet]. San Jose: Photoshop; 2021 [cited 2021 Jul 8]. Available from: https://www. link1

[53] Smith LN, Topin N. Super-convergence: very fast training of neural networks using large learning rates. In: Proceedings of Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications; 2019 Apr 14– 18; Baltimore, MD, USA. Bellingham: Society of Photo-Optical Instrumentation Engineers; 2019. p. 2019. link1

[54] Mu L, Zheng F, Tao R, Zhang Q, Kapelan Z. Hourly and daily urban water demand predictions using a long short-term memory based model. J Water Resour Plan Manage 2020;146(9):05020017. link1

[55] Wang C, Hou J, Miller D, Brown I, Jiang Y. Flood risk management in sponge cities: the role of integrated simulation and 3D visualization. Int J Disaster Risk Reduct 2019;39:101139. link1

Related Research