Deep learning techniques have recently been the most popular method for automatically detecting bridge damage captured by unmanned aerial vehicles(UAVs).However,their wider application to real-world scenarios is hinde...Deep learning techniques have recently been the most popular method for automatically detecting bridge damage captured by unmanned aerial vehicles(UAVs).However,their wider application to real-world scenarios is hindered by three challenges:①defect scale variance,motion blur,and strong illumination significantly affect the accuracy and reliability of damage detectors;②existing commonly used anchor-based damage detectors struggle to effectively generalize to harsh real-world scenarios;and③convolutional neural networks(CNNs)lack the capability to model long-range dependencies across the entire image.This paper presents an efficient Vision Transformer-enhanced anchor-free YOLO(you only look once)method to address these challenges.First,a concrete bridge damage dataset was established,augmented by motion blur and varying brightness.Four key enhancements were then applied to an anchor-based YOLO method:①Four detection heads were introduced to alleviate the multi-scale damage detection issue;②decoupled heads were employed to address the conflict between classification and bounding box regression tasks inherent in the original coupled head design;③an anchor-free mechanism was incorporated to reduce the computational complexity and improve generalization to real-world scenarios;and④a novel Vision Transformer block,C3MaxViT,was added to enable CNNs to model long-range dependencies.These enhancements were integrated into an advanced anchor-based YOLOv5l algorithm,and the proposed Vision Transformer-enhanced anchor-free YOLO method was then compared against cutting-edge damage detection methods.The experimental results demonstrated the effectiveness of the proposed method,with an increase of 8.1%in mean average precision at intersection over union threshold of 0.5(mAP_(50))and an improvement of 8.4%in mAP@[0.5:.05:.95]respectively.Furthermore,extensive ablation studies revealed that the four detection heads,decoupled head design,anchor-free mechanism,and C3MaxViT contributed improvements of 2.4%,1.2%,2.6%,and 1.9%in mAP50,respectively.展开更多
This paper presents a highly efficient method for recognizing the existence of bridge coating rust defects by using color image processing. The detection of defects on steel bridge surfaces during the operation and ma...This paper presents a highly efficient method for recognizing the existence of bridge coating rust defects by using color image processing. The detection of defects on steel bridge surfaces during the operation and maintenance of bridge structures is important to ensure the safety and reliability of them. More advanced techniques such as digital image processing have been studied for better monitoring and detection as existing infrastructure systems are aged and deteriorated rapidly. Recently, image-based defect recognition and assessment methods have gained considerable attention in the civil engineering domain due to their accuracy, speed, and lower cost. The proposed method in this paper is a fast decision-making system by utilizing color image processing. It was developed by processing original bridge coating images to generate color values and calculating eigenvalues from each digitized image. The values from two different groups, a defective group and a nondefective group, are compared each other to figure out the feasibility of this approach. Finally, an automated defect recognition method is presented and tested with more images. This method can be used to make a decision whether a given digitized image contains defects.展开更多
In order to achieve an access to strain sensor data with wireless transmission in bridge engineering structure testing, a wireless strain test system is presented based on the resistance strain sensor of networks. The...In order to achieve an access to strain sensor data with wireless transmission in bridge engineering structure testing, a wireless strain test system is presented based on the resistance strain sensor of networks. The wireless bridge strain test system composed of master station and substation adopts the wireless method to realize the high accuracy data acquisition between the master station and the substation under a reliable communication protocol. The system has been tested in contrast with the present strain apparatus. Results show that the wireless system is high-reliable, and has many characteristics such as high efficiency, good precision, high stability with low cost, and good flexibility, without using the present communication network.展开更多
This paper presents an air-coupled impact echo(IE)technique that relies on the phase spectrum of the collected data to find the frequencies corresponding to the reflections from delaminations.The proposed technique ta...This paper presents an air-coupled impact echo(IE)technique that relies on the phase spectrum of the collected data to find the frequencies corresponding to the reflections from delaminations.The proposed technique takes advantage of the fact that the IE compression wave is not a propagating wave,but it is the 1st order symmetrical(S1)mode Lamb wave at zero group velocity(S1-ZGV).Therefore,it searches the phase spectra of the data collected by multiple sensors to locate the frequency corresponding to the lowest phase difference.As a result,the technique reduces the effect of propagating waves,including the direct acoustic wave and ambient noise.It is named the Constant Phase IE(CPIE).The performance of the CPIE is experimentally compared with the regular amplitude spectrum-based IE technique and two other multisensor IE techniques.The CPIE shows a performance advantage,especially in a noisy environment.展开更多
基金support by University of Auckland Faculty Research Development Fund(3716476).
文摘Deep learning techniques have recently been the most popular method for automatically detecting bridge damage captured by unmanned aerial vehicles(UAVs).However,their wider application to real-world scenarios is hindered by three challenges:①defect scale variance,motion blur,and strong illumination significantly affect the accuracy and reliability of damage detectors;②existing commonly used anchor-based damage detectors struggle to effectively generalize to harsh real-world scenarios;and③convolutional neural networks(CNNs)lack the capability to model long-range dependencies across the entire image.This paper presents an efficient Vision Transformer-enhanced anchor-free YOLO(you only look once)method to address these challenges.First,a concrete bridge damage dataset was established,augmented by motion blur and varying brightness.Four key enhancements were then applied to an anchor-based YOLO method:①Four detection heads were introduced to alleviate the multi-scale damage detection issue;②decoupled heads were employed to address the conflict between classification and bounding box regression tasks inherent in the original coupled head design;③an anchor-free mechanism was incorporated to reduce the computational complexity and improve generalization to real-world scenarios;and④a novel Vision Transformer block,C3MaxViT,was added to enable CNNs to model long-range dependencies.These enhancements were integrated into an advanced anchor-based YOLOv5l algorithm,and the proposed Vision Transformer-enhanced anchor-free YOLO method was then compared against cutting-edge damage detection methods.The experimental results demonstrated the effectiveness of the proposed method,with an increase of 8.1%in mean average precision at intersection over union threshold of 0.5(mAP_(50))and an improvement of 8.4%in mAP@[0.5:.05:.95]respectively.Furthermore,extensive ablation studies revealed that the four detection heads,decoupled head design,anchor-free mechanism,and C3MaxViT contributed improvements of 2.4%,1.2%,2.6%,and 1.9%in mAP50,respectively.
文摘This paper presents a highly efficient method for recognizing the existence of bridge coating rust defects by using color image processing. The detection of defects on steel bridge surfaces during the operation and maintenance of bridge structures is important to ensure the safety and reliability of them. More advanced techniques such as digital image processing have been studied for better monitoring and detection as existing infrastructure systems are aged and deteriorated rapidly. Recently, image-based defect recognition and assessment methods have gained considerable attention in the civil engineering domain due to their accuracy, speed, and lower cost. The proposed method in this paper is a fast decision-making system by utilizing color image processing. It was developed by processing original bridge coating images to generate color values and calculating eigenvalues from each digitized image. The values from two different groups, a defective group and a nondefective group, are compared each other to figure out the feasibility of this approach. Finally, an automated defect recognition method is presented and tested with more images. This method can be used to make a decision whether a given digitized image contains defects.
基金Sponsored by the Multidisciline Scientific Research Foundation of Harbin Institute of Technology(Grant No.HIT.MD2003.14)the Scientific Research Foundation of Liaoning Provincial Communication Department(Grant No.200516)
文摘In order to achieve an access to strain sensor data with wireless transmission in bridge engineering structure testing, a wireless strain test system is presented based on the resistance strain sensor of networks. The wireless bridge strain test system composed of master station and substation adopts the wireless method to realize the high accuracy data acquisition between the master station and the substation under a reliable communication protocol. The system has been tested in contrast with the present strain apparatus. Results show that the wireless system is high-reliable, and has many characteristics such as high efficiency, good precision, high stability with low cost, and good flexibility, without using the present communication network.
文摘This paper presents an air-coupled impact echo(IE)technique that relies on the phase spectrum of the collected data to find the frequencies corresponding to the reflections from delaminations.The proposed technique takes advantage of the fact that the IE compression wave is not a propagating wave,but it is the 1st order symmetrical(S1)mode Lamb wave at zero group velocity(S1-ZGV).Therefore,it searches the phase spectra of the data collected by multiple sensors to locate the frequency corresponding to the lowest phase difference.As a result,the technique reduces the effect of propagating waves,including the direct acoustic wave and ambient noise.It is named the Constant Phase IE(CPIE).The performance of the CPIE is experimentally compared with the regular amplitude spectrum-based IE technique and two other multisensor IE techniques.The CPIE shows a performance advantage,especially in a noisy environment.
