Deep learning-based visual inspection for the delayed brittle fracture of high-strength bolts in long-span steel bridges

2019 ◽  
Author(s):  
Jing Zhou ◽  
Linsheng Huo ◽  
Gangbing Song ◽  
Hongnan Li
Keyword(s):  
Deep Learning ◽  
Brittle Fracture ◽  
Visual Inspection ◽  
High Strength ◽  
Steel Bridges ◽  
Long Span

scholarly journals Computer Vision-Based Detection for Delayed Fracture of Bolts in Steel Bridges

2021 ◽  
Vol 2021 ◽  
pp. 1-12
Author(s):  
Jing Zhou ◽  
Linsheng Huo
Keyword(s):  
Computer Vision ◽  
Data Augmentation ◽  
Learning Algorithm ◽  
High Strength ◽  
Steel Bridges ◽  
Average Precision ◽  
Delayed Fracture ◽  
Long Span ◽  
Deep Learning Algorithm ◽  
Manual Inspection

The delayed fracture of high-strength bolts occurs frequently in the bolt connections of long-span steel bridges. This phenomenon can threaten the safety of structures and even lead to serious accidents in certain cases. However, the manual inspection commonly used in engineering to detect the fractured bolts is time-consuming and inconvenient. Therefore, a computer vision-based inspection approach is proposed in this paper to rapidly and automatically detect the fractured bolts. The proposed approach is realized by a convolutional neural network- (CNN-) based deep learning algorithm, the third version of You Only Look Once (YOLOv3). A challenge for the detector training using YOLOv3 is that only limited amounts of images of the fractured bolts are available in practice. To address this challenge, five data augmentation methods are introduced to produce more labeled images, including brightness transformation, Gaussian blur, flipping, perspective transformation, and scaling. Six YOLOv3 neural networks are trained using six different augmented training sets, and then, the performance of each detector is tested on the same testing set to compare the effectiveness of different augmentation methods. The highest average precision (AP) of the trained detectors is 89.14% when the intersection over union (IOU) threshold is set to 0.5. The practicality and robustness of the proposed method are further demonstrated on images that were never used in the training and testing of the detector. The results demonstrate that the proposed method can quickly and automatically detect the delayed fracture of high-strength bolts.

Image-Based Comprehensive Maintenance and Inspection Method for Bridges Using Deep Learning

2018 ◽  
Author(s):  
Xuefeng Zhao ◽  
Shengyuan Li ◽  
Hongguo Su ◽  
Lei Zhou ◽  
Kenneth J. Loh
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Visual Inspection ◽  
Health State ◽  
Bridge Management ◽  
Inspection Method ◽  
Maintenance Work ◽  
Maintenance And Inspection ◽  
Fine Turning

Bridge management and maintenance work is an important part for the assessment the health state of bridge. The conventional management and maintenance work mainly relied on experienced engineering staffs by visual inspection and filling in survey forms. However, the human-based visual inspection is a difficult and time-consuming task and its detection results significantly rely on subjective judgement of human inspectors. To address the drawbacks of human-based visual inspection method, this paper proposes an image-based comprehensive maintenance and inspection method for bridges using deep learning. To classify the types of bridges, a convolutional neural network (CNN) classifier established by fine-turning the AlexNet is trained, validated and tested using 3832 images with three types of bridges (arch, suspension and cable-stayed bridge). For the recognition of bridge components (tower and deck of bridges), a Faster Region-based Convolutional Neural Network (Faster R-CNN) based on modified ZF-net is trained, validated and tested by utilizing 600 bridge images. To implement the strategy of a sliding window technique for the crack detection, another CNN from fine-turning the GoogLeNet is trained, validated and tested by employing a databank with cropping 1455 raw concrete images into 60000 intact and cracked images. The performance of the trained CNNs and Faster R-CNN is tested on some new images which are not used for training and validation processes. The test results substantiate the proposed method can indeed recognize the types and components and detect cracks for a bridges.

scholarly journals Calculated research of aeroelastic stability of a bridge over the river Ob in Salekhard at the stage of assembling and operation

2019 ◽  
Vol 6 (3) ◽  
Author(s):  
Anastasiya Shustikova ◽  
Andrei Kozichev ◽  
Sergei Paryshev ◽  
Konstantin Strelkov
Keyword(s):  
Cfd Simulation ◽  
Load Capacity ◽  
Air Flow ◽  
Lift Coefficient ◽  
High Strength ◽  
Railway Bridge ◽  
Aeroelastic Stability ◽  
Ansys Fluent ◽  
Long Span ◽  
Long Span Bridge

Recently, long span bridge construction has been demanded for development of the regions of the Russian Federation. In terms of economy, it’s useful to build a combined road-railway bridge. Such bridges, generally, constitute a metal cross-cutting girder with carriageways on lower, upper or both zones of the girder. The major advantages of combined bridges are high strength and load capacity, plus cross-cutting to wind load. Focus of this research is a combined road-railway bridge over the Ob river at the stage of assembling and operation. The purpose of the study was to determine the limits of aeroelastic stability of combined road-railway bridge at the stage of assembling and operation using numerical simulation. To better understand the bridges behaviour in air flow, flow around a section model has been researched with CFD simulation in the ANSYS FLUENT. Then based on the given results of the calculations the dependence of the bridge vibrations on wind speed within a specified range is obtained, and also values of drag coefficient Сх, lift coefficient Су and torque coefficient Мz are received. These studies were carried out in the range of angles of attack α = ±3°. The possibility of divergence and galloping was also estimated. The results of the study made it possible to estimate the influence of air flow on combined bridge cross-cutting girder. Overall, the conducted research seems promising for further investigation and development in the field of bridge aeroelasticity.

scholarly journals Probability of Detection Study for Visual Inspection of Steel Bridges: Volume 2—Full Project Report

2019 ◽  
Author(s):  
Leslie E. Campbell ◽  
Luke R. Snyder ◽  
Julie M. Whitehead ◽  
Robert J. Connor
Keyword(s):  
Visual Inspection ◽  
Steel Bridges ◽  
Probability Of Detection ◽  
Project Report

scholarly journals Automated Brittle Fracture Rate Estimator for Steel Property Evaluation Using Deep Learning After Drop-Weight Tear Test

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 145095-145103 ◽  
Author(s):  
Gyogwon Koo ◽  
Crino Shin ◽  
Hyeyeon Choi ◽  
Jong-Hak Lee ◽  
Sang Woo Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Brittle Fracture ◽  
Fracture Rate ◽  
Drop Weight ◽  
Property Evaluation ◽  
Drop Weight Tear Test ◽  
Steel Property

scholarly journals Visual Inspection of the Aircraft Surface Using a Teleoperated Reconfigurable Climbing Robot and Enhanced Deep Learning Technique

2019 ◽  
Vol 2019 ◽  
pp. 1-14 ◽  
Author(s):  
Balakrishnan Ramalingam ◽  
Vega-Heredia Manuel ◽  
Mohan Rajesh Elara ◽  
Ayyalusami Vengadesh ◽  
Anirudh Krishna Lakshmanan ◽  
...  
Keyword(s):  
Deep Learning ◽  
Visual Inspection ◽  
Surface Defects ◽  
Learning Algorithm ◽  
Periodic Pattern ◽  
Surface Inspection ◽  
Detection Accuracy ◽  
Climbing Robot ◽  
Aircraft Skin ◽  
Surface Images

Aircraft surface inspection includes detecting surface defects caused by corrosion and cracks and stains from the oil spill, grease, dirt sediments, etc. In the conventional aircraft surface inspection process, human visual inspection is performed which is time-consuming and inefficient whereas robots with onboard vision systems can inspect the aircraft skin safely, quickly, and accurately. This work proposes an aircraft surface defect and stain detection model using a reconfigurable climbing robot and an enhanced deep learning algorithm. A reconfigurable, teleoperated robot, named as “Kiropter,” is designed to capture the aircraft surface images with an onboard RGB camera. An enhanced SSD MobileNet framework is proposed for stain and defect detection from these images. A Self-filtering-based periodic pattern detection filter has been included in the SSD MobileNet deep learning framework to achieve the enhanced detection of the stains and defects on the aircraft skin images. The model has been tested with real aircraft surface images acquired from a Boeing 737 and a compact aircraft’s surface using the teleoperated robot. The experimental results prove that the enhanced SSD MobileNet framework achieves improved detection accuracy of aircraft surface defects and stains as compared to the conventional models.

Compressive Testing of H-Shaped Steel Stub Columns Fabricated with Grade 800MPa High Performance Steel

2013 ◽  
Vol 671-674 ◽  
pp. 646-649
Author(s):  
Kang Min Lee ◽  
Myung Jae Lee ◽  
Young Suk Oh ◽  
T.S. Kim ◽  
Do Hwan Kim
Keyword(s):  
High Performance ◽  
High Strength ◽  
Structural Members ◽  
Basic Study ◽  
High Rise ◽  
Long Span ◽  
Compressive Testing ◽  
High Performance Steel ◽  
Stub Column ◽  
Structural Behaviors

With the increased demand for high-rise and long-span structures, high strength with high performance steels have been utilized for these kind of structures. For the grade 800MPa high performance steel, although it was included in Korean Standard as high strength steel(HSA 800), however the HSA 800 steel was excluded in Korean Building Code-Structures due to the rack of research results for the structural behaviors of members fabricated with HSA 800 steel. Therefore, this paper describes basic study for the design specification of structural members using HSA 800 high performance steel. For this purpose, welded H-shaped stub column specimens with various width-to-thickness ratios were designed and tested in order to investigate the buckling behaviors and ultimate compressive strength.

Sign in / Sign up

Export Citation Format

Share Document