• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.365-381
• /
• 2023

The existing vision-based techniques for inspection and condition assessment of civil infrastructure are mostly manual and consequently time-consuming, expensive, subjective, and risky. As a viable alternative, researchers in the past resorted to deep learning-based autonomous damage detection algorithms for expedited post-disaster reconnaissance of structures. Although a number of automatic damage detection algorithms have been proposed, the scarcity of labeled training data remains a major concern. To address this issue, this study proposed a semi-supervised learning (SSL) framework based on consistency regularization and cross-supervision. Image data from post-earthquake reconnaissance, that contains cracks, spalling, and exposed rebars are used to evaluate the proposed solution. Experiments are carried out under different data partition protocols, and it is shown that the proposed SSL method can make use of unlabeled images to enhance the segmentation performance when limited amount of ground truth labels are provided. This study also proposes DeepLab-AASPP and modified versions of U-Net++ based on channel-wise attention mechanism to better segment the components and damage areas from images of reinforced concrete buildings. The channel-wise attention mechanism can effectively improve the performance of the network by dynamically scaling the feature maps so that the networks can focus on more informative feature maps in the concatenation layer. The proposed DeepLab-AASPP achieves the best performance on component segmentation and damage state segmentation tasks with mIoU scores of 0.9850 and 0.7032, respectively. For crack, spalling, and rebar segmentation tasks, modified U-Net++ obtains the best performance with Igou scores (excluding the background pixels) of 0.5449, 0.9375, and 0.5018, respectively. The proposed architectures win the second place in IC-SHM2021 competition in all five tasks of Project 2.

• Structural Monitoring and Maintenance
• /
• v.2 no.1
• /
• pp.19-34
• /
• 2015

One of the main causes of a limited use of nondestructive evaluation (NDE) technologies in bridge deck assessment is the speed of data collection and analysis. The paper describes development and implementation of the RABIT (Robotics Assisted Bridge Inspection Tool) for data collection using multiple NDE technologies. The system is designed to characterize three most common deterioration types in concrete bridge decks: rebar corrosion, delamination, and concrete degradation. It implements four NDE technologies: electrical resistivity (ER), impact echo (IE), ground-penetrating radar (GPR), and ultrasonic surface waves (USW) method. The technologies are used in a complementary way to enhance the interpretation. In addition, the system utilizes advanced vision to complement traditional visual inspection. Finally, the RABIT collects data at a significantly higher speed than it is done using traditional NDE equipment. The robotic system is complemented by an advanced data interpretation. The associated platform for the enhanced interpretation of condition assessment in concrete bridge decks utilizes data integration, fusion, and deterioration and defect visualization. This paper concentrates on the validation and field implementation of two NDE technologies. The first one is IE used in the delamination detection and characterization, while the second one is the USW method used in the assessment of concrete quality. The validation of performance of the two methods was conducted on a 9 m long and 3.6 m wide fabricated bridge structure with numerous artificial defects embedded in the deck.

• Journal of the Korea Concrete Institute
• /
• v.25 no.6
• /
• pp.621-630
• /
• 2013

Several techniques for steel corrosion detection are proposed and HCP (half cell potential) technique is widely adopted for field investigation. If concrete has cracks on surface, steel corrosion is rapidly accelerated due to additional intrusion of chloride and carbon dioxide ions. This study is for an evaluation of HCP in cracked concrete exposed chloride attack. For this work, RC (reinforced concrete) beams are prepared considering 3 w/c ratios (0.35, 0.55, and 0.70) and several cover depths (10~60 mm) and various crack widths of 0.0~1.0 mm are induced. For 35 days, SST (salt spraying test) is performed for corrosion acceleration, and HCP and corrosion length of rebar are evaluated. With increasing crack width, w/c ratios, and decreasing cover depth, HCP measurements increase. HCP evaluation technique is proposed considering the effects of w/c ratios, crack width, and cover depth. Furthermore anti-corrosive cover depths are obtained through Life365 program and the results are compared with those from this study. The results shows relatively big difference in cracked concrete, however provide similar anti-corrosive conditions in sound concrete.