Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images

Concrete dams are prone to various hidden dangers after long-term operation and may lead to significant risk if failed to be detected in time. However, the existing hollowing detection techniques are few as well as inefficient when facing the demands of comprehensive coverage and intelligent managem...

Full description

Saved in:
Bibliographic Details
Main Authors: Lili Zhang, Zihan Jin, Yibo Wang, Ziyi Wang, Zeyu Duan, Taoran Qi, Rui Shi
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/10/3199
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327119512895488
author Lili Zhang
Zihan Jin
Yibo Wang
Ziyi Wang
Zeyu Duan
Taoran Qi
Rui Shi
author_facet Lili Zhang
Zihan Jin
Yibo Wang
Ziyi Wang
Zeyu Duan
Taoran Qi
Rui Shi
author_sort Lili Zhang
collection DOAJ
description Concrete dams are prone to various hidden dangers after long-term operation and may lead to significant risk if failed to be detected in time. However, the existing hollowing detection techniques are few as well as inefficient when facing the demands of comprehensive coverage and intelligent management for regular inspections. Hence, we proposed an innovative, non-destructive infrared inspection method via constructed dataset and proposed deep learning algorithms. We first modeled the surface temperature field variation of concrete dams as a one-dimensional, non-stationary partial differential equation with Robin boundary. We also designed physics-informed neural networks (PINNs) with multi-subnets to compute the temperature value automatically. Secondly, we obtained the time-domain features in one-dimensional space and used the diffusion techniques to obtain the synthetic infrared images with dam hollowing by converting the one-dimensional temperatures into two-dimensional ones. Finally, we employed adaptive joint learning to obtain the spatio-temporal features. We designed the experiments on the dataset we constructed, and we demonstrated that the method proposed in this paper can handle the low-data (few shots real images) issue. Our method achieved 94.7% of recognition accuracy based on few shots real images, which is 17.9% and 5.8% higher than maximum entropy and classical OTSU methods, respectively. Furthermore, it attained a sub-10% cross-sectional calculation error for hollowing dimensions, outperforming maximum entropy (70.5% error reduction) and OTSU (7.4% error reduction) methods, which shows our method being one novel method for automated intelligent hollowing detection.
format Article
id doaj-art-a8e828cc014448c9bb2c8d1f314db829
institution Kabale University
issn 1424-8220
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-a8e828cc014448c9bb2c8d1f314db8292025-08-20T03:47:58ZengMDPI AGSensors1424-82202025-05-012510319910.3390/s25103199Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared ImagesLili Zhang0Zihan Jin1Yibo Wang2Ziyi Wang3Zeyu Duan4Taoran Qi5Rui Shi6College of Computer Science and Software Engineering, Hohai University, Nanjing 211100, ChinaCollege of Information Science and Engineering, Hohai University, Changzhou 213200, ChinaCollege of Information Science and Engineering, Hohai University, Changzhou 213200, ChinaCollege of Information Science and Engineering, Hohai University, Changzhou 213200, ChinaCollege of Information Science and Engineering, Hohai University, Changzhou 213200, ChinaCollege of Information Science and Engineering, Hohai University, Changzhou 213200, ChinaThe National Key Laboratory of Water Disaster Prevention, Nanjing Hydraulic Research Institute, Nanjing 210029, ChinaConcrete dams are prone to various hidden dangers after long-term operation and may lead to significant risk if failed to be detected in time. However, the existing hollowing detection techniques are few as well as inefficient when facing the demands of comprehensive coverage and intelligent management for regular inspections. Hence, we proposed an innovative, non-destructive infrared inspection method via constructed dataset and proposed deep learning algorithms. We first modeled the surface temperature field variation of concrete dams as a one-dimensional, non-stationary partial differential equation with Robin boundary. We also designed physics-informed neural networks (PINNs) with multi-subnets to compute the temperature value automatically. Secondly, we obtained the time-domain features in one-dimensional space and used the diffusion techniques to obtain the synthetic infrared images with dam hollowing by converting the one-dimensional temperatures into two-dimensional ones. Finally, we employed adaptive joint learning to obtain the spatio-temporal features. We designed the experiments on the dataset we constructed, and we demonstrated that the method proposed in this paper can handle the low-data (few shots real images) issue. Our method achieved 94.7% of recognition accuracy based on few shots real images, which is 17.9% and 5.8% higher than maximum entropy and classical OTSU methods, respectively. Furthermore, it attained a sub-10% cross-sectional calculation error for hollowing dimensions, outperforming maximum entropy (70.5% error reduction) and OTSU (7.4% error reduction) methods, which shows our method being one novel method for automated intelligent hollowing detection.https://www.mdpi.com/1424-8220/25/10/3199spatio-temporal infrared featuresnon-destructive inspectionhollowing detectioninfrared imageslow-data learningadaptive joint learning
spellingShingle Lili Zhang
Zihan Jin
Yibo Wang
Ziyi Wang
Zeyu Duan
Taoran Qi
Rui Shi
Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images
Sensors
spatio-temporal infrared features
non-destructive inspection
hollowing detection
infrared images
low-data learning
adaptive joint learning
title Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images
title_full Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images
title_fullStr Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images
title_full_unstemmed Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images
title_short Novel Spatio-Temporal Joint Learning-Based Intelligent Hollowing Detection in Dams for Low-Data Infrared Images
title_sort novel spatio temporal joint learning based intelligent hollowing detection in dams for low data infrared images
topic spatio-temporal infrared features
non-destructive inspection
hollowing detection
infrared images
low-data learning
adaptive joint learning
url https://www.mdpi.com/1424-8220/25/10/3199
work_keys_str_mv AT lilizhang novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages
AT zihanjin novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages
AT yibowang novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages
AT ziyiwang novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages
AT zeyuduan novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages
AT taoranqi novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages
AT ruishi novelspatiotemporaljointlearningbasedintelligenthollowingdetectionindamsforlowdatainfraredimages