Human factors affecting visual inspection of fatigue cracking in steel bridges

Campbell, Leslie E.; Connor, Robert; Whitehead, Julie; Washer, Glenn

doi:10.1080/15732479.2020.1813783

Cited by 17 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conventional scheduled maintenance identifies concrete cracks using visual inspection, which requires extensive manual observation and the training of inspectors. To a great extent, the quantification parameters of defects depend on the inspectors’ judgmental skills, which are subjective and prone to error [ 4 ]. Hence, there is a need for robust non-destructive inspection (NDI) techniques that result in consistent outcomes for different types of test structures.…”

Section: Introductionmentioning

confidence: 99%

Fast Detection of Missing Thin Propagating Cracks during Deep-Learning-Based Concrete Crack/Non-Crack Classification

Geetha

Yang

Sim

2023

Sensors

View full text Add to dashboard Cite

Existing deep learning (DL) models can detect wider or thicker segments of cracks that occupy multiple pixels in the width direction, but fail to distinguish the thin tail shallow segment or propagating crack occupying fewer pixels. Therefore, in this study, we proposed a scheme for tracking missing thin/propagating crack segments during DL-based crack identification on concrete surfaces in a computationally efficient manner. The proposed scheme employs image processing as a preprocessor and a postprocessor for a 1D DL model. Image-processing-assisted DL as a precursor to DL eliminates labor-intensive labeling and the plane structural background without any distinguishable features during DL training and testing; the model identifies potential crack candidate regions. Iterative differential sliding-window-based local image processing as a postprocessor to DL tracks missing thin cracks on segments classified as cracks. The capability of the proposed method is demonstrated on low-resolution images with cracks of single-pixel width, captured using unmanned aerial vehicles on concrete structures with different surface textures, different scenes with complicated disturbances, and optical variability. Due to the multi-threshold-based image processing, the overall approach is invariant to the choice of initial sensitivity parameters, hyperparameters, and the sequence of neuron arrangement. Further, this technique is a computationally efficient alternative to semantic segmentation that results in pixelated mapping/classification of thin crack regimes, which requires labor-intensive and skilled labeling.

show abstract

Section: Introductionmentioning

confidence: 99%

Fast Detection of Missing Thin Propagating Cracks during Deep-Learning-Based Concrete Crack/Non-Crack Classification

Geetha

Yang

Sim

2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Despite the fact that the importance of fatigue weldment cracking has been recognized and that numerous relevant research investigations have been conducted and reported since the 1920s [6], the detection of fatigue cracking in welds still mainly relies on visual inspections [7][8][9] and nondestructive evaluation (NDE) techniques (e.g. thermography [10], ultrasonic testing [11], x-ray imaging [12,13], magnetic particle testing [14], eddy-current testing [15], acoustic emission monitoring [16][17][18]).…”

Section: Introductionmentioning

confidence: 99%

Investigation of textured sensing skin for monitoring fatigue cracks on fillet welds

Liu¹,

Laflamme²,

Li³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

Load-induced fatigue cracking in welds is a critical safety concern for steel transportation infrastructure, and the automation of their detection using commercial sensing technologies remains challenging due to the randomness in crack initiation and propagation. The authors have previously proposed a corrugated soft elastomeric capacitor (cSEC), which is a flexible and ultra-compliant thin-film strain gauge that transduces strain into a measurable change in capacitance. The cSEC technology has been successfully demonstrated for measuring bending strain as well as angular rotation in a folded configuration. This study builds on prior discoveries to characterize the sensor’s capability at monitoring fatigue cracks in corner welds, for which the sensor needs to be installed in a folded configuration. A crack monitoring algorithm is developed to fuse the cSEC data into actionable information. Experimental work is conducted on an orthogonal welded connection, mimicking a plate-to-web joint in steel bridges, with cSECs folded over the fillet welds. The sensor’s electromechanical behavior is characterized, and results confirm that the cSEC is capable of fatigue crack detection and quantification. In particular, results show that the cSEC can detect a minimum crack length of 0.48 mm and that its overall sensing performance, including signal linearity, resolution, and accuracy, is adequate under no damage, yet decreases with increasing crack size, likely attributable to the simplification of the electromechanical model and higher noise produced by the loading equipment under smaller applied displacement.

show abstract

“…Visual inspection is currently the most popular approach used in detecting fatigue cracks, but the process is timeconsuming, labor-intensive, and relies on the inspector's judgment [4]. A recent study revealed that early-stage cracks have a low probability of detection through visual inspection [5].…”

Section: Introductionmentioning

confidence: 99%

Investigation of surface textured sensing skin for fatigue crack localization and quantification

Liu

Laflamme

et al. 2021

Smart Mater. Struct.

View full text Add to dashboard Cite

The timely discovery and monitoring of fatigue cracks in steel structures is an important task in order to ensure structural integrity. However, off-the-shelf strain sensors are small and their deployment is too spatially localized to successfully locate new crack formation or growth within acceptable confidence. A solution is the use of large-area electronics capable of covering large surfaces. The authors have previously developed a sensing skin technology based on a soft elastomeric capacitor (SEC) that consists of a highly compliant, low-cost, and scalable strain gauge that transduces surface strain into a measurable change in capacitance. In prior work, the SEC was fabricated using three flat layers that formed a parallel plate capacitor. An improvement to that design has recently been proposed, in which the dielectric layer is textured using a corrugated pattern, and preliminary work showed a substantial improvement in sensing performance due to the added in-plane stiffness and decrease in the sensor's transverse Poisson's ratio. This paper extends preliminary work by studying the textured sensor's capability to quantify and discover a fatigue crack, and to study the general sensing performance in terms of linearity, sensitivity, resolution, and accuracy. Four specific corrugation patterns are investigated: a symmetric grid, a diagonal diagrid, a reinforced diagrid, and a non-symmetric re-entrant hexagonal honeycomb (auxetic) pattern. Experimental results show that the use of a texture results in a significant increase in sensing performance, with auxetic and reinforced diagrid patterns outperforming the other patterns. In particular, the auxetic and reinforced diagrid patterns allowed the discovery of a 0.28 mm and 0.31 mm fatigue crack, compared to a 0.53 mm crack for the untextured SEC, and resulted in up to 106% increase in sensitivity, 113% in linearity, 319% in resolution, and 582% in accuracy, compared to untextured SECs.

show abstract

Human factors affecting visual inspection of fatigue cracking in steel bridges

Cited by 17 publications

References 17 publications

Fast Detection of Missing Thin Propagating Cracks during Deep-Learning-Based Concrete Crack/Non-Crack Classification

Fast Detection of Missing Thin Propagating Cracks during Deep-Learning-Based Concrete Crack/Non-Crack Classification

Investigation of textured sensing skin for monitoring fatigue cracks on fillet welds

Investigation of surface textured sensing skin for fatigue crack localization and quantification

Contact Info

Product

Resources

About