Benchmarking Unmanned Aerial Systems-Assisted Inspection of Steel Bridges for Fatigue Cracks

Dorafshan, Sattar; Campbell, Leslie E.; Maguire, Marc; Connor, Robert

doi:10.1177/03611981211001073

Cited by 13 publications

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Commonly, RGB and IRT cameras are utilized to identify fatigue cracks. The effectiveness of UAV-based fatigue crack detection is greatly influenced by the platform that is used, the environment, and the lighting [13].…”

Section: Bmentioning

confidence: 99%

Untitled

2024

CTCSE

View full text Add to dashboard Cite

show abstract

Section: Bmentioning

confidence: 99%

Untitled

2024

CTCSE

View full text Add to dashboard Cite

show abstract

“…Many departments of transportation (DOTs) use unmanned aerial systems (UASs) for bridge inspections, allowing them to mitigate the problems related to manual inspections [ 10 ], which might be followed by the use of conventional image processing methods to detect steel corrosion [ 11 , 12 , 13 , 14 ]. The problems related to conventional image processing techniques are inaccurate user-defined parameters and adverse lighting condition effects [ 11 ].…”

Section: Introductionmentioning

confidence: 99%

Autonomous Image-Based Corrosion Detection in Steel Structures Using Deep Learning

Das,

Dorafshan,

Kaabouch

2024

Sensors

Self Cite

View full text Add to dashboard Cite

Steel structures are susceptible to corrosion due to their exposure to the environment. Currently used non-destructive techniques require inspector involvement. Inaccessibility of the defective part may lead to unnoticed corrosion, allowing the corrosion to propagate and cause catastrophic structural failure over time. Autonomous corrosion detection is essential for mitigating these problems. This study investigated the effect of the type of encoder–decoder neural network and the training strategy that works the best to automate the segmentation of corroded pixels in visual images. Models using pre-trained DesnseNet121 and EfficientNetB7 backbones yielded 96.78% and 98.5% average pixel-level accuracy, respectively. Deeper EffiecientNetB7 performed the worst, with only 33% true-positive values, which was 58% less than ResNet34 and the original UNet. ResNet 34 successfully classified the corroded pixels, with 2.98% false positives, whereas the original UNet predicted 8.24% of the non-corroded pixels as corroded when tested on a specific set of images exclusive to the investigated training dataset. Deep networks were found to be better for transfer learning than full training, and a smaller dataset could be one of the reasons for performance degradation. Both fully trained conventional UNet and ResNet34 models were tested on some external images of different steel structures with different colors and types of corrosion, with the ResNet 34 backbone outperforming conventional UNet.

show abstract

“…To detect fatigue cracks, RGB and IRT cameras are usually used [30]. Careful selection of a UAV platform, environmental conditions, and lighting conditions are important factors that affect UAV-based fatigue crack detection [75].…”

mentioning

confidence: 99%

Impact of UAV Hardware Options on Bridge Inspection Mission Capabilities

Ameli

Aremanda

Friess

et al. 2022

Drones

View full text Add to dashboard Cite

Uncrewed Aerial Vehicles (UAV) constitute a rapidly evolving technology field that is becoming more accessible and capable of supplementing, expanding, and even replacing some traditionally manual bridge inspections. Given the classification of the bridge inspection types as initial, routine, in-depth, damage, special, and fracture critical members, specific UAV mission requirements can be developed, and their suitability for UAV application examined. Results of a review of 23 applications of UAVs in bridge inspections indicate that mission sensor and payload needs dictate the UAV configuration and size, resulting in quadcopter configurations being most suitable for visual camera inspections (43% of visual inspections use quadcopters), and hexa- and octocopter configurations being more suitable for higher payload hyperspectral, multispectral, and Light Detection and Ranging (LiDAR) inspections (13%). In addition, the number of motors and size of the aircraft are the primary drivers in the cost of the vehicle. 75% of vehicles rely on GPS for navigation, and none of them are capable of contact inspections. Factors that limit the use of UAVs in bridge inspections include the UAV endurance, the capability of navigation in GPS deprived environments, the stability in confined spaces in close proximity to structural elements, and the cost. Current research trends in UAV technologies address some of these limitations, such as obstacle detection and avoidance methods, autonomous flight path planning and optimization, and UAV hardware optimization for specific mission requirements.

show abstract

Benchmarking Unmanned Aerial Systems-Assisted Inspection of Steel Bridges for Fatigue Cracks

Cited by 13 publications

References 20 publications

Untitled

Untitled

Autonomous Image-Based Corrosion Detection in Steel Structures Using Deep Learning

Impact of UAV Hardware Options on Bridge Inspection Mission Capabilities

Contact Info

Product

Resources

About