A robust real‐time method for identifying hydraulic tunnel structural defects using deep learning and computer vision

Abstract

Robots with cameras provide a non-contact information acquisition solution for hydraulic tunnels, while manual damage-related information extraction is timeconsuming and costs labor. This study proposes a robust real-time framework for identifying hydraulic tunnel underwater structural damage using deep learning and computer vision. First, a high-performance detector is built via the You Only Look Once v5s and adaptively spatial feature fusion module. A series of comparative experiments are used to explore the setting of the sparsification and pruning ratios to trade off a balance between accuracy and efficiency. Model sparsity ratio of 0.01 with a pruning ratio of 0.3 can be combined to change the weight distribution in the batch normalization layer and reduce network redundant parameters for slimming. Then, model fine-tuning with knowledge distillation is utilized to recover the accuracy degradation caused by pruning. A hydraulic tunnel is utilized as the case study, and three defects including rust, exfoliation, and calcification precipitate are utilized as research items. The performance of the models was evaluated based on detection accuracy, robustness, and efficiency. Five extreme attributes of underwater scenes, including oblique angles, high brightness, uneven illumination, low visibility, and obstacle interference, were considered to test model generalization and efficacy. Experimental results show it achieves good detection performance in complicated underwater scenes, achieving 0.814 precision, 0.980 recall, 0.889 F1_score, and 0.894 Mean Average Precision (mAP)@0.5 in the test set. Moreover, the proposed method achieved a 50 Frames Per Second (FPS) detection speed when detecting video with 1080p, indicating its real-time detection capability. INTRODUCTIONThe linings of hydraulic tunnels inevitably suffer from defects, like cracks, rust, exfoliation, and calcification precipitates, under the coupling effect of external pressure and high-speed water flow erosion (Zhou et al., 2022). Timely and reliable identification of these structural defects is critical for characterizing the health of © 2022 Computer-Aided Civil and Infrastructure Engineering.hydraulic tunnels. The periodic visual inspection by professionals has been a practical solution in civil infrastructures (Wang et al., 2021). However, it suffers from problems like high risk, labor cost, and low efficiency, especially for hydraulic tunnels in water-filled states. Moreover, emptying the water in tunnels may cause stress redistribution, even resulting in the local collapse of linings.

Enhanced YOLO v3 for precise detection of apparent damage on bridges amidst complex backgrounds

“…This makes it more appropriate for use in complex backgrounds where apparent bridge damage needs to be detected. Additionally, the detection speed fps of the enhanced algorithm is only 2 fps slower than the original YOLO v3, meaning that it can still identify bridge damage with greater accuracy and at a high speed, a speed that can enable it to detect defects in real time 28 .…”

Heterogeneity‐oriented ensemble learning for rail monitoring based on vehicle‐body vibration

“…Since the publication of the first journal article on civil engineering applications of neural networks (Adeli & Yeh, 1989), this topic has gained increasing attention with computer hardware F I G U R E 2 As a brief demonstration of the technique route and relevant challenges, only common waveform features were extracted and fitted in three-dimensional feature spaces to provide rapid visualization. A systematic approach is discussed step-by-step, starting from Section 3. improvements in recent years (Li, Bao, et al, 2023). Lu et al (2020) applied a modified CNN combined with a significant extraction method to rail detection tasks on a high-speed railway line and achieved good performance.…”

“…Machine learning applications in industries, particularly in steel structure research, have emerged as early as the last century (Adeli & Park, 1996). Since the publication of the first journal article on civil engineering applications of neural networks (Adeli & Yeh, 1989), this topic has gained increasing attention with computer hardware improvements in recent years (Li, Bao, et al., 2023). Lu et al.…”

Image-based automatic multiple-defect detection of urban utility tunnel using UUTNet

“…Object detection makes use of bounding boxes to classify and locate defects in images (Redmon et al, 2016; Ren et al, 2017). Li (Li et al, 2022) performed the sparsification and pruning based on You Only Look Once v5s (YOLOv5s), which reduced network redundancy parameters. Then, knowledge distillation was used to restore the accuracy reduction brought by pruning, and real-time identification of defects such as rust, exfoliation, and calcification precipitate in hydraulic structures was finally realized.…”