Accurate Crack Research Articles

SnakeConv and SFC boosting precise segmentation on the crack of tunnel lining surface: based on DeepLabV3+ with improved Swin transformer V2

Abstract Tunnel cracks pose a significant threat to structural integrity, potentially leading to localized collapse of the infrastructure. Traditional manual crack detection methods are prohibitively expensive, highlighting the need for an efficient and accurate automatic crack segmentation model. To address this challenge, we propose a novel crack segmentation model for subway tunnel lining surface based on the DeepLabV3+ architecture. In this model, we design an improved Swin transformer V2 Base (SwinV2*) as the backbone to enhance crack segmentation performance. Considering the tubular morphology of tunnel cracks, we introduce a snake convolution module to better capture their unique features. To prevent performance degradation when fusing shallow and deep features, we incorporate a spatial feature calibration module that facilitates feature alignment and grouping along the channel dimension. We assess our model’s effectiveness using thousands of crack images captured by the image acquisition system designed for subway tunnel surfaces. Experimental results show that our model achieves strong performance metrics: 68.96% IoU, 84.33% mIoU, 87.57% PA. Compared to the original DeepLabV3+, our approach demonstrates superior performance, with a 2.89% improvement in IoU, a 1.45% increase in mIoU and notably, a significant 10.39% improvement in PA.

Detection and Classification of Surface Cracks Using Deep Learning Based Autoencoders in Eddy Current Testing

AbstractIndustrial equipment subjected to rigorous conditions of high speed and pressure leads to the development of cracks on metal surfaces. These cracks reduce the service life and threaten the safety of parts, and the deeper the crack, the greater the resulting damage. Crack detection and crack depth evaluation continue to take center stage in quantitative non‐destructive testing and evaluation (NDT&E 4.0). The accuracy of the rotating uniform eddy current (RUEC) probe in achieving fast and efficient detection of surface cracks is corroborated by a comparison with previous experimental results. Next, accurate crack depth classification is achieved by building deep learning model based on a sparse autoencoder (SAE) and a multi‐layer perceptron (MLP) model. These classifiers are combined with eddy current testing (ECT) data, including the normal magnetic component Bz. As a result, evaluation metrics such as accuracy increased by up to 100% with both precision and recall scores of 1 for the deep sparse autoencoder classifier compared to MLP performance. The originality of our approach is evident in the application of deep SAE, which achieves high classification accuracy. Furthermore, the integration of our high‐resolution NDT&E RUEC probe with advanced machine learning models for depth classification is both novel and impactful. This unique combination offers a comprehensive framework for crack analysis, from precise detection to detailed characterization. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.

Unsupervised PG-DDPM-augmented mixed dataset for training an accurate concrete bridge crack detection model under small samples

Unsupervised PG-DDPM-augmented mixed dataset for training an accurate concrete bridge crack detection model under small samples

Thermal effects on mode I fracture of sandstone: Accurate crack identification in thermal-mechanical coupled peridynamic simulations

Thermal effects on mode I fracture of sandstone: Accurate crack identification in thermal-mechanical coupled peridynamic simulations

Numerical prediction and experiments for 3D crack propagation in brittle materials based on 3D-generalized maximum tangential strain criterion

Accurately predicting and tracing the three-dimensional (3D) propagation and fracture trajectory of a crack inside brittle materials is challenging. One difficulty is that the 3D crack propagation exhibits complex I/II/III mixed-mode expansion, and there is a lack of accurate crack initiation criteria and effective simulation methods. In our previous studies, the 3D-generalized maximum tangential strain (3D-GMTSN) criterion was proposed to determine the direction and onset of 3D fracture initiation. In this study, a Python program based on the 3D-GMTSN criterion is developed and integrated into FRANC3D to predict the 3D propagation trajectory of an arbitrary crack inside brittle solids. To verify the reliability of the new method, three different types of 3D crack modes, namely Internal Inclined Cracks Cuboid (IICC), Edge Notched Disc Bend (ENDB), and Three-Point Bending (TPB), are used for fracture experiments. The 3D crack propagation morphology is identified using high-resolution CT imaging techniques. The IICC, ENDB, and TPB models are simulated using the new method and the conventional numerical method based on the maximum shear stress (MSS), maximumtensile stress (MTS), and maximum energy release rate (MERR) criteria. Comparisons indicate that the proposed method based on the 3D-GMTSN criterion can predict the 3D crack propagation trajectory more accurately than the conventional methods.

DESIGN OF A CAMERA-ENABLED MOBILE ROBOT FOR IN-PIPE INSPECTION

The role of pipeline infrastructure is crucial in modern society, as it transports necessary resources such as water, gas, and oil. It is essential to prioritize the maintenance of these pipelines to prevent leaks, environmental harm, and economic losses. Traditional inspection techniques, which involve manual inspection, can be inefficient, laborious, and susceptible to mistakes. To overcome these obstacles, this article introduces a camera-equipped mobile robot specifically created for automated crack detection in pipelines. The robot, which is equipped with a high-resolution camera and advanced image processing algorithms, navigates inside pipelines on its own to capture detailed images of the pipe walls. These images are then processed using computer vision techniques to identify and categorize potential cracks. The algorithms utilize machine learning models that have been trained on a dataset of labelled crack images, enabling accurate and efficient crack detection. This paper provides a comprehensive overview of the robot system's design, development, and validation, detailing the critical hardware components such as the mobility platform, camera system, and sensor integration, as well as the software architecture, which encompasses image processing and machine learning frameworks. Field studies demonstrate the robot's ability to effectively detect cracks across various pipe materials and under diverse environmental conditions. Results indicate that the automated system not only improves inspection efficiency and accuracy but also significantly reduces the need for human intervention, thereby enhancing overall pipeline safety Keywords: pipeline inspection, crack detection, mobile robot, image processing, computer vision.

Nonlinear vibration characteristics and damage detection method of blade with breathing fatigue crack

In aero-engine, blades operate under harsh conditions for extended periods, which makes them highly susceptible to fatigue cracks. The localized and nonlinear structural changes caused by blade crack lead to complex vibration characteristics that are not well understood, posing challenges to the identification of these cracks. This paper proposes an approach for identifying breathing fatigue crack in compressor blade, leveraging the nonlinear features induced by the cracks. Crucial sensitive characteristics are extracted and a crack indicator for accurate crack detection is defined. Initially addressing the typical morphology of fatigue breathing cracks, a dynamic model of blade with cracks is developed to analyze their impact on the natural frequencies. Nonlinear contact forces are introduced to characterize the "breathing" effect of the fatigue crack, allowing for the determination of the harmonic distribution patterns in the blade's vibration response under various excitation conditions. A finite element model is then established considering the crack morphology, by adopting the contact element. Based on that, the nonlinear response characteristics associated with the blade's primary, sub-harmonic, and super-harmonic resonances are explored. A crack-sensitive parameter is defined and extracted from the nonlinear responses, under different crack propagation stages. At last, effectiveness and reliability of the defined parameter for identifying cracks is validated through a vibration fatigue test.

Lightweight concrete crack detection based on spiking neural networks

PurposeMost existing methods for concrete crack detection are based on deep learning techniques such as convolutional neural networks. However, these models, due to their large memory footprint, high power consumption and insufficient feature extraction capabilities, face challenges in mobile applications. To address these issues, this paper proposes a lightweight spiking neural network detection model.Design/methodology/approachThis model achieves fast and accurate crack detection. Firstly, the Gabor-Spiking (GS) module preprocesses input images, extracting texture features and edge features of crack images through Gabor filter convolution modules and spiking convolution modules, respectively. Next, the multiscale residual (MR) module is designed, composed of convolutional layers and residual modules of various scales, to process the fused features and perform crack detection.FindingsExperimental results demonstrate that the model’s size can be reduced to 4.6 MB, achieving accuracy improvements to 87.3 and 96.4% on the SDNET and OCD datasets, respectively.Originality/valueThis paper proposes a lightweight spiking neural network detection model based on the GS module for edge texture feature fusion and the MR module for crack detection.

An anisotropic eigenfracture approach accounting for mixed fracture modes in wooden structures by the Representative Crack Element framework

Finite Element analysis of anisotropic fracture phenomena in wood is a challenging task, particularly when dealing with intricate loading scenarios and mode-specific behavior. The appeal of energetically motivated approaches, such as the eigenfracture method, is that they enable simulation of fracture without prior knowledge of the crack path. The promising eigenfracture method has shown good numerical performance for isotropic materials, and this contribution showcases its application to anisotropic materials. Wood is one such anisotropic material and in this manuscript, the directional dependence of both elasticity and fracture evolution are incorporated into the eigenfracture approach. Further, the eigenfracture approach is used in conjunction with Representative Crack Elements (RCE), which permit accurate modeling of physical crack deformations. The governing equations are systematically derived and implemented into the Finite Element framework. By representative numerical examples, some advantages over the alternative phase-field method are demonstrated. Another highlight of this work is that it is possible to provide a realistic ratio of the energy release rates parallel to and perpendicular to the fiber direction in order to achieve physically accurate crack patterns. Additionally, the calculation effort is reduced, because the unknowns required to determine the crack kinematics can be solved analytically at the material level, a feature that also enables parallelization.

Topology-aware mamba for crack segmentation in structures

CrackMamba, a Mamba-based model, is designed for efficient and accurate crack segmentation for monitoring the structural health of infrastructure. Traditional Convolutional Neural Network (CNN) models struggle with limited receptive fields, and while Vision Transformers (ViT) improve segmentation accuracy, they are computationally intensive. CrackMamba addresses these challenges by utilizing the VMambaV2 with pre-trained ImageNet-1 k weights as the encoder and a newly designed decoder for better performance. To handle the random and complex nature of crack development, a Snake Scan module is proposed to reshape crack feature sequences, enhancing feature extraction. Additionally, the three-branch Snake Conv VSS (SCVSS) block is proposed to target cracks more effectively. Experiments show that CrackMamba achieves state-of-the-art (SOTA) performance on the CrackSeg9k and SewerCrack datasets, and demonstrates competitive performance on the retinal vessel segmentation dataset CHASE_DB1, highlighting its generalization capability.

A multi-scale deep residual network-based guided wave imaging evaluation method for fatigue crack quantification

Abstract As a promising structural health monitoring technology, guided wave (GW) imaging is gaining increasing attention for crack monitoring of aircraft structures. However, actual fatigue crack propagation is a complex dynamically evolving process affected by various variabilities. It is still challenging to accurately track and quantify the dynamic fatigue crack propagation with GW imaging methods. Therefore, in order to achieve more accurate fatigue crack quantification, this paper proposes a multi-scale deep residual network-based GW imaging evaluation method. A convolutional neural network (CNN) is utilized to evaluate the entire pixel distribution of GW imaging maps to fuse damage-related information from multiple GW monitoring paths. By designing multi-scale convolutional kernels and deep residual learning, a robust quantitative image feature extraction is ensured with the dynamic evolution process of fatigue crack growth and the performance degradation is avoided as the CNN goes deeper, thereby improving the quantification accuracy. The method is validated on a fatigue test of landing gear beams, which are important load-carrying aircraft structural components. The results demonstrate that the proposed method can extract multi-scale crack length-related features and accurately track fatigue crack propagations. For batch specimens, the maximum quantification error is reduced from the original 6.1 mm to 1.6 mm, marking a significant improvement.

Small fatigue crack behavior of CP-Ti in thin-walled cruciform specimens under biaxial loading

This study investigates the small fatigue crack propagation behavior of commercially pure titanium (CP-Ti) using thin-walled cruciform specimens under in-plane biaxial loading, considering the effects of biaxial ratio and phase angle. Increasing phase angle results in more secondary cracks merging with main cracks perpendicular to the rolling direction (RD) and transverse direction (TD), a phenomenon attributed to the rise in shear stress that accelerates main crack growth. Higher loading biaxiality or a lower phase angle leads to decreased crack propagation rates and increased biaxial fatigue life. Electron backscatter diffraction (EBSD) analysis reveals that when the maximum normal stress aligns with the RD, prismatic slip primarily governs crack propagation, thereby accelerating crack propagation rates. Conversely, alignment with the TD reduces prismatic slip activity and crack propagation rates. Under equi-biaxial loading, prismatic slip activity decreases further, and crack propagation is dominated by multiple slip and twinning, consequently resulting in the slowest propagation rates. Additionally, a higher proportion of prismatic slip under high phase angle also accelerates crack propagation. Finally, incorporating Findley equivalent stress into the Chapetti model, which considers the crack length-dependent threshold effect, a highly accurate biaxial small fatigue crack propagation rate model is proposed.

Crack SAM: enhancing crack detection utilizing foundation models and Detectron2 architecture

Accurate crack detection is crucial for maintaining pavement integrity, yet manual inspections remain labor-intensive and prone to errors, underscoring the need for automated solutions. This study proposes a novel crack segmentation approach utilizing advanced visual models, specifically Detectron2 and the Segment Anything Model (SAM), applied to the CFD and Crack500 datasets, which exhibit intricate and diverse crack patterns. Detectron2 was tested with four configurations—mask_rcnn_R_50_FPN_3x, mask_rcnn_R_101_FPN_3x, faster_rcnn_R_50_FPN_3x, and faster_rcnn_R_101_FPN_3x—while SAM was compared using Focal Loss, DiceCELoss, and DiceFocalLoss. SAM with DiceFocalLoss outperformed Detectron2, achieving mean IoU scores of 0.69 and 0.59 on the CFD and Crack500 datasets, respectively. The integration of Detectron2 with faster_rcnn_R_101_FPN_3x and SAM using DiceFocalLoss involves generating bounding boxes with Detectron2, which serve as prompts for SAM to produce segmentation masks. This approach achieves mIoU scores of 0.83 for CFD dataset and 0.75 for Crack500 dataset. These results highlight the potential of combining foundation models with Detectron2 for advancing crack detection technologies, offering valuable insights for enhancing highway maintenance systems.

A Crack Detection Method for Civil Engineering Bridges Based on Feature Extraction and Parametric Modeling of Point Cloud Data

Accurate detection and analysis of cracks is critical for ensuring the safety and reliability of concrete bridges. Point cloud data (PCD) obtained from 3D scanning provides a promising avenue for automated crack assessment. However, processing the massive and unstructured PCD poses significant challenges in feature extraction and crack modeling. This paper proposes a novel method for bridge crack analysis by combining PCD feature extraction with a hierarchical neural network and Rodriguez rotation. The method first extracts crack features from PCD using outlier removal, denoising, and 3D coordinate conversion. A crack analysis model is then constructed by integrating multi-scale feature extraction and Rodriguez rotation into a hierarchical neural network, enabling the capture of both local and global crack patterns. Experiments on a benchmark data set demonstrate the effectiveness of the proposed approach, achieving 92.83% feature extraction accuracy, 95.73% parameter analysis accuracy, 93.51% recognition accuracy, and 0.91 F1 score. The method also shows improved efficiency compared to existing techniques. These results highlight the potential of the proposed PCD-based approach for accurate and efficient crack analysis in concrete bridges.

Effect of random variation in input parameter on cracked orthotropic plate using extended isogeometric analysis (XIGA) under thermomechanical loading

Abstract This research introduces the Stochastic Extended Isogeometric Analysis (XIGA) method to investigate fracture behavior of isotropic and orthotropic materials under mechanical, thermal, and thermomechanical loads. Employing knot spans from Isogeometric Analysis (IGA) for domain discretization, the study utilizes identical basis functions for geometry construction and solution discretization. Utilizing Extended Finite Element Method (XFEM) enrichment functions, accurate crack face displacement discontinuity and tip singularity within the stress field are characterized. Additionally, employing a second-order perturbation technique within XIGA framework, the research derives mean and coefficient of variance values for mixed-mode Stress Intensity Factors (SIF). Stochastic variations in material elastic properties, crack length, and crack angle are considered in this computation. Credibility and robustness of the study are confirmed through comparative analyses against available literatures and Monte Carlo Simulations (MCS). The observed exceptional agreement validates the precision and reliability of the proposed stochastic XIGA method for fracture analysis in orthotropic material systems under thermomechanical loading conditions.

An iterative up-sampling convolutional neural network for glass curtain crack detection using unmanned aerial vehicles

It is inefficient, subjective, and potentially life-threatening for both inspectors and pedestrians to manually inspect glass curtain cracks of high-rise buildings. Due to the development of image instance segmentation techniques for deep learning and the operation of compact unmanned aerial vehicles (UAVs), automatic, accurate, and safe glass curtain crack inspection can be achieved. A novel approach proposes to segment glass curtain cracks efficiently and accurately, combining a convolutional neural network (CNN), the region of interest alignment (RoIAlign), and an iterative up-sampling method. The approach proposed achieves excellent performance with an accuracy = 99.89 %, precision = 91.1 %, recall = 72.1 %, F1-score = 79.7 %, and mIoU = 78.7 %. Comparative studies have been conducted to examine the performance of the proposed approach relative to Solo, Swin, Mask RCNN, and ConvNext V2. The results show that the proposed network outperforms the networks used for comparison and that it can more accurately detect and segment a variety of crack types in complex reflective environments of glass curtains.

A strain-based J-integral formulation for an internal circumferential surface crack of pipeline under inner pressure and large axial deformation

The presence of cracks on pipelines poses a potential threat to their operational status, and it is critical to assess the permissibility of pipelines containing cracks. The dimensional analysis combined with the finite element method is applied to investigate the fracture behavior of circumferential crack on the internal surface of pipe under internal pressure and large axial deformation. Dimensionless parameters are determined to represent the effects of crack size, pipe geometry, pipe material, and external load on the crack front driving force, and a strain-based J-integral formulation is obtained by a stepwise coefficient fitting approach rather than a polynomial fitting method. This J-integral formula can be used to quickly assess the crack front driving force of a pipe in service condition and subjected to axial displacement. The diameter-to-thickness ratio of the pipe and the dimensionless pressure of the pipe are found to act together in a combined form on the crack front driving force. Increases in dimensionless crack depth, dimensionless crack length, the ratio of circumferential stress to yield strength of the pipe, and strain hardening exponent cause an increase in the crack front driving force. The effect of dimensionless crack depth on crack front driving force is more significant than other dimensionless parameters. Changes in the other dimensionless parameters do not significantly change the crack front driving force when the dimensionless crack depth is small. Other dimensionless parameters have a progressively greater influence on the crack front driving force as the crack dimensionless crack depth increases. Large deformations in the ligament zone and increasing axial stress are the main reasons for the high crack front driving force. The J-integral formula has a similar form to that of the J-integral in the Electric Power Research Institute (EPRI) method when the effect of internal pressure is not considered. It can be reduced to predict the crack front driving force of a surface cracked plate subjected to uniaxial tensile loading. For the interaction between an internal surface crack and an embedded crack, re-characterizing the crack size using BS 7910 will overestimate the equivalent crack depth, and a more accurate equivalent crack size can be obtained using the J-integral formula proposed.

Crack analysis of tall concrete wind towers using an ad-hoc deep multiscale encoder–decoder with depth separable convolutions under severely imbalanced data

An accurate and timely cracking assessment, including the presence, location and crack geometric feature measurement, is crucial for evaluating concrete wind towers. Therefore, the early identification of cracks is a critical procedure in promptly evaluating structural integrity. This study proposed an ad-hoc encoder–decoder network based on DeepLabv3+ with depth separable convolutions to automatically segment cracks from real-world images captured from various concrete wind towers. The combined advantages of the improved DeepLabv3+ and the lightweight MobileNet v2 are suitable as a benchmark due to their high performance and universality. Four experiments were conducted to determine the model design choice and crack feature measurement capability: (1) six parametric tests using various pre-trained base networks and algorithm optimisers, (2) the influence of complex background noise (i.e., handwriting script) on crack segmentation performance, (3) comparative studies with cutting-edge pixel-wise segmentation models and (4) crack feature measurement (i.e., length and width). The research outcome demonstrated that DeepLabv3+ with MobileNet v2 can potentially be applied for efficient and accurate crack segmentation in concrete wind towers with complex backgrounds.

Enhanced pavement crack segmentation with minimal labeled data: a triplet attention teacher-student framework

ABSTRACT Effective crack detection is critical for pavement maintenance, yet existing methods face significant challenges. Deep learning has become a popular solution due to its superior accuracy and automation capabilities, but supervised methods require extensive labelled data, making it costly and time-intensive. Unsupervised methods, while less data-reliant, struggle with feature extraction, especially for indistinct crack edges. Semi-supervised methods attempt to bridge this gap by using minimal labelled data, enhancing accuracy while reducing manual labelling efforts. However, accurate segmentation remains difficult, particularly with sparse labelled data or complex crack patterns. These limitations highlight the need for a more effective approach. To address these challenges, we propose a novel framework based on a teacher-student model enhanced with triplet attention. Leveraging the Unet architecture, this approach enables efficient feature extraction with minimal labelled data. The teacher model refines predictions using an exponential moving average of the student model’s weights, while the student model is trained with supervised and uncertainty losses. Evaluations on three benchmark datasets demonstrate our framework outperforms fully supervised models, achieving higher intersection-over-union and dice scores with just 2% labelled data. This approach overcomes the limitations of existing methods, enabling accurate crack segmentation with minimal labelled data.

Probabilistic Deep Learning Approach for Fatigue Crack Width Estimation and Prognosis in Lap Joint Using Acoustic Waves

Abstract Accurate fatigue crack width estimation is crucial for aircraft safety, however, limited research exists on (i) the direct relationship between fatigue crack width and Lamb wave signatures and (ii) probabilistic artificial intelligence approach for automated analysis using acoustic emission waveforms. This paper presents a probabilistic deep learning approach for fatigue crack width estimation, employing an automated wavelet feature extractor and probabilistic Bayesian neural network. A dataset constituting the fatigue experiment on aluminum lap joint specimens is considered, in which Lamb wave signals were recorded at several time instants for each specimen. Signals acquired from the piezo actuator–receiver sensor pairs are related to the optically measured surface crack length. The sensitive features are automatically extracted from the signals using decomposition techniques called maximal overlap discrete wavelet transform (MODWT). The extracted features are then mapped through the deep learning model, which incorporates Bayesian inference to account for both aleatoric as well as epistemic uncertainty, that provides outcomes in the form of providing probabilistic estimates of crack width with uncertainty quantification. Thus, employing an automated wavelet feature extractor (MODWT) on a dataset of fatigue experiments, the framework learns the relationship between Lamb wave signals and crack width. Validation on unseen in situ data demonstrates the efficacy of the approach for practical implementation, paving the way for more reliable fatigue life prognosis.