Intelligent monitoring method for road inundation based on deep learning

被引：0
|
作者：
Bai G. [1 ]
Hou J. [1 ]
Han H. [1 ]
Xia J. [2 ]
Li B. [1 ]
Zhang Y. [1 ]
Wei Z. [1 ]
机构：
[1] State Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an
[2] State Key Laboratory of Water Resources and Hydropower Engineering Science, Wuhan University, Wuhan
来源：
Water Resources Protection | 2021年 / 37卷 / 05期
关键词：
Convolutional neural network; Deep learning; Image identification; Inundation monitoring; Urban inundation;
D O I：
10.3880/j.issn.1004-6933.2021.05.012
中图分类号：
学科分类号：
摘要：
To solve the problems including frequent urban inundation and unsafe, costly and inefficient traditional monitoring urban inundation method, a method of detecting urban road flood rapidly based on deep learning techniques is proposed. This method, based on convolutional neural network, can extract the features of puddles from the input accumulated water image data set. The water accumulation in Xi’an University of Technology is selected for verification. It is shown that the average recognition accuracy of the method for training and verification of the data set is 96． 1% and 90． 1% respectively, and automatic extraction of the puddle area in the image is realized accurately. So that the automatic identification of the inundation area and the automatic acquisition of the water area in the urban inundation monitoring image are realized. © The Author(s) (2021).
引用
收藏
页码：75 / 80
页数：5
相关论文
共 23 条
  • [1] HUANG Miansong, YANG Shaoxiong, QI Wenchao, Et al., Numerical simulation of urban waterlogging reduction effect in Guyuan sponge city [J], Water Resources Protection, 35, 5, pp. 13-18, (2019)
  • [2] LU Haiming, ZOU Ying, SUN Jinhua, Et al., Simulation of waterlogging control effect in Tiexinqiao experimental base based on SWMM[J], Water Resources Protection, 36, 1, pp. 58-65, (2020)
  • [3] WU Haichun, HUANG Guoru, Risk assessment of urban waterlogging based on PCSWMM model [J], Water Resources Protection, 32, 5, pp. 11-16, (2016)
  • [4] SONG Yun, LI Qiongfang, NIU Mingyuan, Et al., Rainstorm and waterlogging simulation in typical inundated districts of Nanjing based on SWMM, Advances in Science and Technology of Water Resources, 39, 6, pp. 56-61, (2019)
  • [5] HOU Jingming, LI Donglai, WANG Xiaojun, Et al., Effects of initial conditions of LID measures on runoff control at residential community scale [J], ·79· Advances in Water Science, 30, 1, pp. 45-55, (2019)
  • [6] HOU Jingming, Application of deep learning technology in urban waterlogging monitoring [J], China Flood & Drought Management, 29, 8, (2019)
  • [7] CHEN Jinshui, Method of water level data capturing based on video image recognition [J], Water Resources Informatization, 1, pp. 48-51, (2013)
  • [8] HUA Xia, WANG Xinqing, RUI Ting, Et al., A fast selfattention cascaded network for object detection in large scene remote sensing images[J], Applied Soft Computing, 94, (2020)
  • [9] ZANG Di, CHAI Zhenliang, ZHANG Junqi, Et al., Vehicle license plate recognition using visual attention model and deep learning[J], Journal of Electronic Imaging, 24, 3, (2015)
  • [10] MASSOLI F V, AMATO G, FALICHI F., Cross-resolution learning for face recognition [J], Image and Vision Computing, 99, (2020)
123