An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks
Water conservancy personnel usually need to know the water level by water gauge images in real-time and with an expected accuracy. However, accurately recognizing the water level from water gauge images is still a complex problem. This article proposes a composite method applied in the Wuyuan City,...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/14/23/6023 |
| _version_ | 1797462222529626112 |
|---|---|
| author | Chen Chen Rufei Fu Xiaojian Ai Chengbin Huang Li Cong Xiaohuan Li Jiange Jiang Qingqi Pei |
| author_facet | Chen Chen Rufei Fu Xiaojian Ai Chengbin Huang Li Cong Xiaohuan Li Jiange Jiang Qingqi Pei |
| author_sort | Chen Chen |
| collection | DOAJ |
| description | Water conservancy personnel usually need to know the water level by water gauge images in real-time and with an expected accuracy. However, accurately recognizing the water level from water gauge images is still a complex problem. This article proposes a composite method applied in the Wuyuan City, Jiangxi Province, in China. This method can detect water gauge areas and number areas from complex and changeable scenes, accurately detect the water level line from various water gauges, and finally, obtain the accurate water level value. Firstly, FCOS is improved by fusing a contextual adjustment module to meet the requirements of edge computing and ensure considerable detection accuracy. Secondly, to deal with scenes with indistinct water level features, we also apply the contextual adjustment module for Deeplabv3+ to segment the water gauge area above the water surface. Then, the area can be used to obtain the position of the water level line. Finally, the results of the previous two steps are combined to calculate the water level value. Detailed experiments prove that this method solves the problem of water level recognition in complex hydrological scenes. Furthermore, the recognition error of the water level by this method is less than 1 cm, proving it is capable of being applied in real river scenes. |
| first_indexed | 2024-03-09T17:33:26Z |
| format | Article |
| id | doaj.art-002ad1983b504a03871dbf94b722b664 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-09T17:33:26Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-002ad1983b504a03871dbf94b722b6642023-11-24T12:04:23ZengMDPI AGRemote Sensing2072-42922022-11-011423602310.3390/rs14236023An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural NetworksChen Chen0Rufei Fu1Xiaojian Ai2Chengbin Huang3Li Cong4Xiaohuan Li5Jiange Jiang6Qingqi Pei7School of Telecommunication Engineering, Xidian University, Xi’an 710071, ChinaSchool of Telecommunication Engineering, Xidian University, Xi’an 710071, ChinaSchool of Telecommunication Engineering, Xidian University, Xi’an 710071, ChinaState Grid Jilin Province Electric Power Company Limited Information Communication Company, Changchun 130021, ChinaState Grid Jilin Province Electric Power Company Limited Information Communication Company, Changchun 130021, ChinaSchool of Electronics and Information Engineering, Beihang University, Beijing 100191, ChinaSchool of Telecommunication Engineering, Xidian University, Xi’an 710071, ChinaSchool of Telecommunication Engineering, Xidian University, Xi’an 710071, ChinaWater conservancy personnel usually need to know the water level by water gauge images in real-time and with an expected accuracy. However, accurately recognizing the water level from water gauge images is still a complex problem. This article proposes a composite method applied in the Wuyuan City, Jiangxi Province, in China. This method can detect water gauge areas and number areas from complex and changeable scenes, accurately detect the water level line from various water gauges, and finally, obtain the accurate water level value. Firstly, FCOS is improved by fusing a contextual adjustment module to meet the requirements of edge computing and ensure considerable detection accuracy. Secondly, to deal with scenes with indistinct water level features, we also apply the contextual adjustment module for Deeplabv3+ to segment the water gauge area above the water surface. Then, the area can be used to obtain the position of the water level line. Finally, the results of the previous two steps are combined to calculate the water level value. Detailed experiments prove that this method solves the problem of water level recognition in complex hydrological scenes. Furthermore, the recognition error of the water level by this method is less than 1 cm, proving it is capable of being applied in real river scenes.https://www.mdpi.com/2072-4292/14/23/6023water level recognitionhydrological monitoringdeep learningcomputer vision |
| spellingShingle | Chen Chen Rufei Fu Xiaojian Ai Chengbin Huang Li Cong Xiaohuan Li Jiange Jiang Qingqi Pei An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks Remote Sensing water level recognition hydrological monitoring deep learning computer vision |
| title | An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks |
| title_full | An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks |
| title_fullStr | An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks |
| title_full_unstemmed | An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks |
| title_short | An Integrated Method for River Water Level Recognition from Surveillance Images Using Convolution Neural Networks |
| title_sort | integrated method for river water level recognition from surveillance images using convolution neural networks |
| topic | water level recognition hydrological monitoring deep learning computer vision |
| url | https://www.mdpi.com/2072-4292/14/23/6023 |
| work_keys_str_mv | AT chenchen anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT rufeifu anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT xiaojianai anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT chengbinhuang anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT licong anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT xiaohuanli anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT jiangejiang anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT qingqipei anintegratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT chenchen integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT rufeifu integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT xiaojianai integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT chengbinhuang integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT licong integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT xiaohuanli integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT jiangejiang integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks AT qingqipei integratedmethodforriverwaterlevelrecognitionfromsurveillanceimagesusingconvolutionneuralnetworks |