Autonomous surface crack identification of concrete structures based on an improved one-stage object detection algorithm

Wu, Pei-Rong; Liu, Airong; Fu, Jiyang; Ye, Xijun; Zhao, Yinghao

doi:10.1016/j.engstruct.2022.114962

Cited by 58 publications

(18 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to evaluate the overall success of our research, it was essential to provide a comparison against similar studies conducted in recent years. Table 6 presents a comparison of our method with those in [ 14 , 29 ].…”

Section: Resultsmentioning

confidence: 99%

“…The authors of [ 14 ], based on a two-stage detector, identified no better balance between detection accuracy and speed, while [ 29 ] based on a one-stage detector, they reported better performance in detection accuracy and speed. Although our method had a lower mAP than [ 29 ], the difference was minimal and around 2.4%. In addition, our results presented a faster calculation speed.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Defect Detection of MEMS Based on Data Augmentation, WGAN-DIV-DC, and a YOLOv5 Model

Shi

Sang

Huang

et al. 2022

Sensors

View full text Add to dashboard Cite

Surface defect detection of micro-electromechanical system (MEMS) acoustic thin film plays a crucial role in MEMS device inspection and quality control. The performances of deep learning object detection models are significantly affected by the number of samples in the training dataset. However, it is difficult to collect enough defect samples during production. In this paper, an improved YOLOv5 model was used to detect MEMS defects in real time. Mosaic and one more prediction head were added into the YOLOv5 baseline model to improve the feature extraction capability. Moreover, Wasserstein divergence for generative adversarial networks with deep convolutional structure (WGAN-DIV-DC) was proposed to expand the number of defect samples and to make the training samples more diverse, which improved the detection accuracy of the YOLOv5 model. The optimal detection model achieved 0.901 mAP, 0.856 F1 score, and a real-time speed of 75.1 FPS. As compared with the baseline model trained using a non-augmented dataset, the mAP and F1 score of the optimal detection model increased by 8.16% and 6.73%, respectively. This defect detection model would provide significant convenience during MEMS production.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Defect Detection of MEMS Based on Data Augmentation, WGAN-DIV-DC, and a YOLOv5 Model

Shi

Sang

Huang

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…14,15 Machine learning (ML) approaches have been introduced in the field of structural property prediction and damage detection. [16][17][18][19][20][21][22][23] Damage detection using ML techniques often requires a procedure of feature extraction followed by damage classification. For feature extraction, structural modal parameters are used in ML-based parametric damage detection methods 24 ; in ML-based non-parametric damage detection methods, multiple features can be defined by users and extracted through different techniques, such as statistical analysis, 25 regression analysis, 26 principal component analysis, 27 and wavelet transform.…”

Section: Introductionmentioning

confidence: 99%

Parallel convolutional neural network toward high efficiency and robust structural damage identification

Cao²,

Liu

et al. 2023

Structural Health Monitoring

Self Cite

View full text Add to dashboard Cite

Artificial intelligence has been implemented recently for processing and analyzing monitored data for damage detection and identification in the field of structural health monitoring (SHM). Existing machine learning methods such as convolutional neural networks (CNNs) usually rely on inputs from a single domain (time or frequency), which may only provide partial information for damage identification. To address this issue, this work proposes a parallel convolutional neural network (P-CNN) that extracts multidimensional features assisted by a computer vision technique. The proposed network comprises a one-dimensional (1D) CNN branch, a two-dimensional (2D) CNN branch, and several fully connected layers. The efficiency and robustness of the proposed network were validated by a public experimental dataset. Our results show that (1) the features extracted by the P-CNN were separated more easily compared with those by 1D-CNN or 2D-CNN; (2) when detecting structural damages, the accuracy of the P-CNN is above 99.4%; (3) the P-CNN exhibits a robust performance when subjected to a high (5 dB) signal-to-noise ratio of the original data; and (4) when compared with traditionally used methods such as GoogLenet and Resnet, the P-CNN outperforms on many aspects of damage identification. We envision that the proposed P-CNN can be integrated into advanced SHM systems with high fidelity and intelligence.

show abstract

“…State-of-the-art DL models account for unforeseen scenarios by considering data for different environmental, optical, and structural variabilities. Holistic reviews of AI-based automated surface crack classifications are provided in [ 34 , 36 , 50 , 51 , 52 , 53 ]. Convolutional neural network (CNN)-based architectures used for crack detection are briefly overviewed in [ 33 ].…”

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

Autonomous surface crack identification of concrete structures based on an improved one-stage object detection algorithm

Cited by 58 publications

References 35 publications

Defect Detection of MEMS Based on Data Augmentation, WGAN-DIV-DC, and a YOLOv5 Model

Defect Detection of MEMS Based on Data Augmentation, WGAN-DIV-DC, and a YOLOv5 Model

Parallel convolutional neural network toward high efficiency and robust structural damage identification

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

Contact Info

Product

Resources

About