A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale
Due to the frequent and sudden occurrence of urban waterlogging, targeted and rapid risk monitoring is extremely important for urban management. To improve the efficiency and accuracy of urban waterlogging monitoring, a real-time determination method of urban waterlogging based on computer vision te...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/15/6/1696 |
| _version_ | 1797609130373939200 |
|---|---|
| author | Haocheng Huang Xiaohui Lei Weihong Liao Haichen Li Chao Wang Hao Wang |
| author_facet | Haocheng Huang Xiaohui Lei Weihong Liao Haichen Li Chao Wang Hao Wang |
| author_sort | Haocheng Huang |
| collection | DOAJ |
| description | Due to the frequent and sudden occurrence of urban waterlogging, targeted and rapid risk monitoring is extremely important for urban management. To improve the efficiency and accuracy of urban waterlogging monitoring, a real-time determination method of urban waterlogging based on computer vision technology was proposed in this study. First, city images were collected and then identified using the ResNet algorithm to determine whether a waterlogging risk existed in the images. Subsequently, the recognition accuracy was improved by image augmentation and the introduction of an attention mechanism (SE-ResNet). The experimental results showed that the waterlogging recognition rate reached 99.50%. In addition, according to the actual water accumulation process, real-time images of the waterlogging area were obtained, and a threshold method using the inverse weight of the time interval (T-IWT) was proposed to determine the times of the waterlogging occurrences from the continuous images. The results showed that the time error of the waterlogging identification was within 30 s. This study provides an effective method for identifying urban waterlogging risks in real-time. |
| first_indexed | 2024-03-11T05:57:12Z |
| format | Article |
| id | doaj.art-533adbf0af17476bb90c1c9d9efd2d76 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-11T05:57:12Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-533adbf0af17476bb90c1c9d9efd2d762023-11-17T13:40:43ZengMDPI AGRemote Sensing2072-42922023-03-01156169610.3390/rs15061696A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block ScaleHaocheng Huang0Xiaohui Lei1Weihong Liao2Haichen Li3Chao Wang4Hao Wang5School of Civil Engineering, Central South University, Changsha 410075, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaState Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100038, ChinaDue to the frequent and sudden occurrence of urban waterlogging, targeted and rapid risk monitoring is extremely important for urban management. To improve the efficiency and accuracy of urban waterlogging monitoring, a real-time determination method of urban waterlogging based on computer vision technology was proposed in this study. First, city images were collected and then identified using the ResNet algorithm to determine whether a waterlogging risk existed in the images. Subsequently, the recognition accuracy was improved by image augmentation and the introduction of an attention mechanism (SE-ResNet). The experimental results showed that the waterlogging recognition rate reached 99.50%. In addition, according to the actual water accumulation process, real-time images of the waterlogging area were obtained, and a threshold method using the inverse weight of the time interval (T-IWT) was proposed to determine the times of the waterlogging occurrences from the continuous images. The results showed that the time error of the waterlogging identification was within 30 s. This study provides an effective method for identifying urban waterlogging risks in real-time.https://www.mdpi.com/2072-4292/15/6/1696urban waterloggingreal-time monitoringcomputer visionResNet |
| spellingShingle | Haocheng Huang Xiaohui Lei Weihong Liao Haichen Li Chao Wang Hao Wang A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale Remote Sensing urban waterlogging real-time monitoring computer vision ResNet |
| title | A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale |
| title_full | A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale |
| title_fullStr | A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale |
| title_full_unstemmed | A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale |
| title_short | A Real-Time Detecting Method for Continuous Urban Flood Scenarios Based on Computer Vision on Block Scale |
| title_sort | real time detecting method for continuous urban flood scenarios based on computer vision on block scale |
| topic | urban waterlogging real-time monitoring computer vision ResNet |
| url | https://www.mdpi.com/2072-4292/15/6/1696 |
| work_keys_str_mv | AT haochenghuang arealtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT xiaohuilei arealtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT weihongliao arealtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT haichenli arealtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT chaowang arealtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT haowang arealtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT haochenghuang realtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT xiaohuilei realtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT weihongliao realtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT haichenli realtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT chaowang realtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale AT haowang realtimedetectingmethodforcontinuousurbanfloodscenariosbasedoncomputervisiononblockscale |