Semantic segmentation of concrete bridge defect images frequently encounters challenges due to insufficient precision and the limited computational capabilities ofmobile devices,thereby considerably affecting the reli...Semantic segmentation of concrete bridge defect images frequently encounters challenges due to insufficient precision and the limited computational capabilities ofmobile devices,thereby considerably affecting the reliability of bridge defect monitoring and health assessment.To tackle these issues,a concrete defects dataset(including spalling,crack,and exposed steel rebar)was curated and multiple semantic segmentation models were developed.In these models,a deep convolutional network or a lightweight convolutional network were employed as the backbone feature extraction networks,with different loss functions configured and various attention mechanism modules introduced for conducting multi-angle comparative research.The comparison of results indicates that utilizing VGG16 as the backbone network of U-Net for semantic segmentation of multi-class concrete defects images resulted in the highest recognition accuracy,achieving a Mean Intersection over Union(MIoU)of 80.37%and a Mean Pixel Accuracy(MPA)of 90.03%.The optimal combination of loss functions was found to be Focal loss and Dice loss.The lightweight convolutional network Mobile NetV2-DeeplabV3 slightly reduced recognition accuracy but significantly decreased the number of parameters,resulting in a faster detection speed of 71.87 frames/s,making it suitable for real-time defect detection.After integrating the SE(Squeeze-and-Excitation),CBAM(Convolutional Block Attention Module),and Coordinate Attention(CA)modules,both VGG16-U-Net and MobileNetV2-DeeplabV3 achieved improved recognition accuracy.Among them,the CAmodule(Coordinate Attention)effectively guides the model to accurately identify subtle concrete defects.The improved VGG16-U-Net can identify previously the new untrained concrete defect images in the concrete structural health monitoring(SHM)system,and the recognition accuracy can meet the demand for intelligent defect image recognition for structural health monitoring of concrete structures.展开更多
When the synthetic aperture focusing technology （SAFT） is used for the detection of the concrete, the signal-to-noise ratio （SNR） and detection depth are not satisfactory. Therefore, the application of SAFT is usu...When the synthetic aperture focusing technology （SAFT） is used for the detection of the concrete, the signal-to-noise ratio （SNR） and detection depth are not satisfactory. Therefore, the application of SAFT is usually limited. In this paper, we propose an improved SAFT technique for the detection of concrete based on the pulse compression technique used in the Radar domain. The proposed method first transmits a linear frequency modulation （LFM） signal, and then compresses the echo signal using the matched filtering method, after which a compressed signal with a narrower main lobe and higher SNR is obtained. With our improved SAFT, the compressed signals are manipulated in the imaging process and the image contrast is improved. Results show that the SNR is improved and the imaging resolution is guaranteed compared with the conventional short-pulse method. From theoretical and experimental results, we show that the proposed method can suppress noise and improve imaging contrast, and can also be used to detect multiple defects in concrete.展开更多
The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect...The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect detection.In contrast to the subjective and inefficient manual visual inspection,deep learning-based algorithms for concrete defect detection exhibit remarkable advantages,emerging as a focal point in recent research.This paper comprehensively analyzes the research progress of deep learning algorithms in the field of surface defect detection in concrete bridges in recent years.It introduces the early detection methods for surface defects in concrete bridges and the development of deep learning.Subsequently,it provides an overview of deep learning-based concrete bridge surface defect detection research from three aspects:image classification,object detection,and semantic segmentation.The paper summarizes the strengths and weaknesses of existing methods and the challenges they face.Additionally,it analyzes and prospects the development trends of surface defect detection in concrete bridges.展开更多
This study aims to develop a damage-detection algorithm based on the electromagnetic wave properties inside a reinforced concrete structure.The proposed method involves employing two algorithms based on data measured ...This study aims to develop a damage-detection algorithm based on the electromagnetic wave properties inside a reinforced concrete structure.The proposed method involves employing two algorithms based on data measured using ground-penetrating radar—a common electromagnetic wave method in civil engineering.The possible defect area was identified based on the energy dissipated by the damage in the frequency-wavenumber domain,with the damage localized using the calculated relative permittivity of the measurements.The proposed method was verified through a finite difference time-domain-based numerical analysis and a testing slab with artificial damage.As a result of verification,the proposed method quickly identified the presence of damage inside the concrete,especially for honeycomb-like defects located at the top of the rebar.This study has practical significance in scanning structures over a large area more quickly than other non-destructive testing methods,such as ultrasonic methods.展开更多
An experimental and numerical investigation into the structural performance of reinforced concrete box sewers with typical corrosion-related extreme defects localized at the ceiling was conducted.Firstly,during the la...An experimental and numerical investigation into the structural performance of reinforced concrete box sewers with typical corrosion-related extreme defects localized at the ceiling was conducted.Firstly,during the large-scale laboratory test,some key struc-tural responses were captured and evaluated,including the crack width development process(via digital image correlation measurement),ceiling deflection,and material strains of both complete and typical defective boxes.The failure modes and load-carrying mechanism throughout the specimen loading phases were analyzed.Furthermore,the specimen failure process was simulated using a damage-basedfinite element method,and a related parameter sensitivity analysis was performed.The results indicate that the defective ceiling cracked at mid-span under a low load value,but the bending capacity loss can be substituted by two shoulders and carry three tofive times more load before completely collapsing.The simulation matched the lab test qualitatively,and with the suggested set strategy of material parameters,the load-deflection feature curve could provide a practical prediction of the ultimate bearing capacity of the defec-tive sewers,with a 10–15%error on the safe side.展开更多
基金supported by the National Natural Science Foundation of China(51708188).
文摘Semantic segmentation of concrete bridge defect images frequently encounters challenges due to insufficient precision and the limited computational capabilities ofmobile devices,thereby considerably affecting the reliability of bridge defect monitoring and health assessment.To tackle these issues,a concrete defects dataset(including spalling,crack,and exposed steel rebar)was curated and multiple semantic segmentation models were developed.In these models,a deep convolutional network or a lightweight convolutional network were employed as the backbone feature extraction networks,with different loss functions configured and various attention mechanism modules introduced for conducting multi-angle comparative research.The comparison of results indicates that utilizing VGG16 as the backbone network of U-Net for semantic segmentation of multi-class concrete defects images resulted in the highest recognition accuracy,achieving a Mean Intersection over Union(MIoU)of 80.37%and a Mean Pixel Accuracy(MPA)of 90.03%.The optimal combination of loss functions was found to be Focal loss and Dice loss.The lightweight convolutional network Mobile NetV2-DeeplabV3 slightly reduced recognition accuracy but significantly decreased the number of parameters,resulting in a faster detection speed of 71.87 frames/s,making it suitable for real-time defect detection.After integrating the SE(Squeeze-and-Excitation),CBAM(Convolutional Block Attention Module),and Coordinate Attention(CA)modules,both VGG16-U-Net and MobileNetV2-DeeplabV3 achieved improved recognition accuracy.Among them,the CAmodule(Coordinate Attention)effectively guides the model to accurately identify subtle concrete defects.The improved VGG16-U-Net can identify previously the new untrained concrete defect images in the concrete structural health monitoring(SHM)system,and the recognition accuracy can meet the demand for intelligent defect image recognition for structural health monitoring of concrete structures.
基金supported by the National Natural Science Foundation of China(No.11074273)the ministry of water resources'special funds for scientific research on public causes(No.201301061)
文摘When the synthetic aperture focusing technology （SAFT） is used for the detection of the concrete, the signal-to-noise ratio （SNR） and detection depth are not satisfactory. Therefore, the application of SAFT is usually limited. In this paper, we propose an improved SAFT technique for the detection of concrete based on the pulse compression technique used in the Radar domain. The proposed method first transmits a linear frequency modulation （LFM） signal, and then compresses the echo signal using the matched filtering method, after which a compressed signal with a narrower main lobe and higher SNR is obtained. With our improved SAFT, the compressed signals are manipulated in the imaging process and the image contrast is improved. Results show that the SNR is improved and the imaging resolution is guaranteed compared with the conventional short-pulse method. From theoretical and experimental results, we show that the proposed method can suppress noise and improve imaging contrast, and can also be used to detect multiple defects in concrete.
基金supported by the Key Research and Development Program of Shaanxi Province-International Science and Technology Cooperation Program Project (No.2020KW-001)the Contract for Xi'an Municipal Science and Technology Plan Project-Xi'an City Strong Foundation Innovation Plan (No.21XJZZ0074)the Key Project of Graduate Student Innovation Fund at Xi'an University of Posts and Telecommunications (No.CXJJZL2023013)。
文摘The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect detection.In contrast to the subjective and inefficient manual visual inspection,deep learning-based algorithms for concrete defect detection exhibit remarkable advantages,emerging as a focal point in recent research.This paper comprehensively analyzes the research progress of deep learning algorithms in the field of surface defect detection in concrete bridges in recent years.It introduces the early detection methods for surface defects in concrete bridges and the development of deep learning.Subsequently,it provides an overview of deep learning-based concrete bridge surface defect detection research from three aspects:image classification,object detection,and semantic segmentation.The paper summarizes the strengths and weaknesses of existing methods and the challenges they face.Additionally,it analyzes and prospects the development trends of surface defect detection in concrete bridges.
基金National Research Foundation of Korea(NRF)Funded by the Korean Government(MSIT)under Grant Nos.RS-2023-00210317 and 2021R1A4A3030117the Digital-Based Building Construction and Safety Supervision Technology Research Program Funded by the Ministry of Land,Infrastructure,and Transport of the Korean Government under Grant No.RS-2022-00143493the Korea Institute of Civil Engineering and Building Technology(KICT)of the Republic of Korea,Project under Grant No.2023-0097。
文摘This study aims to develop a damage-detection algorithm based on the electromagnetic wave properties inside a reinforced concrete structure.The proposed method involves employing two algorithms based on data measured using ground-penetrating radar—a common electromagnetic wave method in civil engineering.The possible defect area was identified based on the energy dissipated by the damage in the frequency-wavenumber domain,with the damage localized using the calculated relative permittivity of the measurements.The proposed method was verified through a finite difference time-domain-based numerical analysis and a testing slab with artificial damage.As a result of verification,the proposed method quickly identified the presence of damage inside the concrete,especially for honeycomb-like defects located at the top of the rebar.This study has practical significance in scanning structures over a large area more quickly than other non-destructive testing methods,such as ultrasonic methods.
基金support received from the Science and Technology Commission of Shanghai Municipality(Contract Number:16DZ1200500).
文摘An experimental and numerical investigation into the structural performance of reinforced concrete box sewers with typical corrosion-related extreme defects localized at the ceiling was conducted.Firstly,during the large-scale laboratory test,some key struc-tural responses were captured and evaluated,including the crack width development process(via digital image correlation measurement),ceiling deflection,and material strains of both complete and typical defective boxes.The failure modes and load-carrying mechanism throughout the specimen loading phases were analyzed.Furthermore,the specimen failure process was simulated using a damage-basedfinite element method,and a related parameter sensitivity analysis was performed.The results indicate that the defective ceiling cracked at mid-span under a low load value,but the bending capacity loss can be substituted by two shoulders and carry three tofive times more load before completely collapsing.The simulation matched the lab test qualitatively,and with the suggested set strategy of material parameters,the load-deflection feature curve could provide a practical prediction of the ultimate bearing capacity of the defec-tive sewers,with a 10–15%error on the safe side.
