Crack Images Research Articles

Data-driven strength-based seismic damage index measurement for RC columns using crack image-derived parameters

The identification of the updated damage state of the structural components is crucial following an earthquake for safety assessment and seismic performance evaluation. The lateral strength loss in RC columns following an earthquake is measured in this paper through surface crack image analysis. The generalized fractal indices of the crack textures for damaged RC columns are considered as the quantitative representatives of the complexity and irregularity of the images. The proposed methodology is generated and verified using an extensive collected research databank comprising crack maps from experimental results on cyclic-loaded RC column specimens with varying residual strengths at pre-cap and post-cap regimes. A strength-based damage index is proposed for the measurement of the extent of strength deterioration in RC columns via five regression-based closed-form equations. The sensitivity of the proposed empirical equations to the visual damage features, structural properties, and geometric parameters is investigated. The findings of this study support the efficiency of fractal analysis for post-earthquake strength loss measurement in the RC columns with different failure modes. Finally, the efficiency of the proposed procedure in the measurement of the residual strength is assessed for a real RC column damaged in 2017, MW 7.3, Kermanshah earthquake. The updated strength values derived from the offered procedure can be incorporated in subsequent structural analysis aimed at collapse safety evaluation or potential weak story identification.

Automated Bridge Crack Detection Based on Improving Encoder–Decoder Network and Strip Pooling

The detection of bridge cracks is an important task in bridge maintenance. It can also reflect the health of the bridge. However, cracks are usually in the form of strips, which are different from the concrete surface. Most crack detection algorithms cannot adapt to this situation well. In this paper, the original image of bridge cracks is collected and the data set is obtained through image processing. A bridge crack detection method based on improving encoder-decoder and mixed pooling module is proposed in this article. The basic features of the crack images are extracted by an encoder with dilated convolution. In this way, the resolution of the feature image can be guaranteed, and large receptive field can be obtained. Then the feature picture through the mix pooling module, which helps to capture remote context information and establish a remote dependency. Finally, the decoder restores the picture to its original size and integrates the original features. In the comparison experiment with the same experimental conditions, we compared with the classic image segmentation methods such as PSPNet, U-Net, FCN, and DeepLabv3+. The results show that our method achieves 98.3%, 97.3%, 97.6%, and 84.5% in precision, recall, F1-score, and MIoU. The results show that our method does have certain advantages in the field of crack detection and segmentation.

Intelligent Detection Method for Concrete Dam Surface Cracks Based on Two-Stage Transfer Learning

The timely identification and detection of surface cracks in concrete dams, an important public safety infrastructure, is of great significance in predicting engineering hazards and ensuring dam safety. Due to their low efficiency and accuracy, manual detection methods are gradually being replaced by computer vision techniques, and deep learning semantic segmentation methods have higher accuracy and robustness than traditional image methods. However, the lack of data images and insufficient detection performance remain challenges in concrete dam surface crack detection scenarios. Therefore, this paper proposes an intelligent detection method for concrete dam surface cracks based on two-stage transfer learning. First, relevant domain knowledge is transferred to the target domain using two-stage transfer learning, cross-domain and intradomain learning, allowing the model to be fully trained with a small dataset. Second, the segmentation capability is enhanced by using residual network 50 (ResNet50) as a UNet model feature extraction network to enhance crack feature information extraction. Finally, multilayer parallel residual attention (MPR) is integrated into its jump connection path to improve the focus on critical information for clearer fracture edge segmentation. The results show that the proposed method achieves optimal mIoU and mPA of 88.3% and 92.7%, respectively, among many advanced semantic segmentation models. Compared with the benchmark UNet model, the proposed method improves mIoU and mPA by 4.6% and 3.2%, respectively, reduces FLOPs by 36.7%, improves inference speed by 48.9%, verifies its better segmentation performance on dam face crack images with a low fine crack miss detection rate and clear crack edge segmentation, and achieves an accuracy of over 85.7% in crack area prediction. In summary, the proposed method has higher efficiency and accuracy in concrete dam face crack detection, with greater robustness, and can provide a better alternative or complementary approach to dam safety inspections than the benchmark UNet model.

Unsupervised domain adaptation for crack detection

The reliable and fast detection of cracks is crucial for assessing the condition and maintaining civil infrastructure. However, due to diverse construction materials, imaging conditions, and environmental interference, there exists a domain shift between crack images collected from civil infrastructure. This shift results in significant performance drops of crack detection models trained on one dataset when applied to another, limiting their cross-dataset applicability. To address this issue, this paper proposes DACrack, an unsupervised domain adaptation framework for crack detection of civil infrastructure. The proposed method performs domain adaptation at the input, feature, and output levels using contrastive mechanisms, adversarial learning, and variational autoencoders. Extensive experiments demonstrate the effectiveness and robustness of the proposed method for cross-dataset crack detection. By mitigating the impact of domain shift, DACrack offers a more reliable and accurate solution for assessing the condition of civil infrastructure.

Study on the Drying Process and the Influencing Factors of Desiccation Cracking of Cohesive Soda Saline-Alkali Soil in the Songnen Plain, China

The surface of cohesive soda saline-alkali soil in the Songnen Plain of China exhibits obvious desiccation cracking phenomenon during water evaporation. Quantitative research on the process and influencing factors of desiccation cracking are significant for determining the mechanical properties of salt-affected soil, improving the salinization status and the local ecological environment. This study aims to conduct laboratory-controlled cracking experiments on soda saline-alkali soils, and to regularly measure the soil moisture and the cracking status of 40 soil samples. After processing the crack images uniformly, crack characteristics including crack length, crack area, contrast texture feature, and box-counting fractal dimension are extracted. The results indicate that the decrease in soil moisture can be divided into three stages such as fast evaporation stage, slow evaporation stage and stable evaporation stage, and that the evaporation rate shows a linear negative correlation with soil salinity. The measurements also indicate that crack length rapidly increases and reaches stability with a decrease in water content compared with the crack area. The high correlation between different crack characteristics and the soil total salinity (correlation coefficient from 0.44 to 0.95) shows that salt content is the dominant factor affecting the desiccation cracking of soda saline-alkali soil. A multiple linear regression prediction model was finally established based on crack characteristics for different salinity parameters with high prediction accuracy for Na+, EC and total salinity (R2 > 0.91 and RPD > 2.5) and also certain prediction accuracy for Cl− and CO32− (R2 > 0.66 and RPD > 1.5).

Automatic concrete infrastructure crack semantic segmentation using deep learning

To solve the privation of crack information during feature transmission, we propose an automatic concrete infrastructure crack semantic segmentation method using deep learning. Initially, based on the statistical analysis results of crack images, a multi-stage feature extraction network is created with multi-resolution parallel transmission to extract the crack features. Then, the extracted features are aggregated according to the correlation between segmentation classes and pixels to enhance the localization performance of the model for cracks. Moreover, using statistical analysis results as constraints construct the loss function to optimize the model and overcome the data imbalance issue. Experiments are conducted on a self-made manual annotation dataset, which contains 2000 images from the dam, the bridge, and the spillway tunnel, and our method reach 94.51% Precision, 86.39% Recall, 82.26% Intersection-over-Unions, and 90.27% F1_measure on the dataset. The experimental results show that the proposed method is optimal for the semantic segmentation of crack images.

Implicit function‐based continuous representation for meticulous segmentation of cracks from high‐resolution images

AbstractHigh‐resolution (HR) crack images offer more detailed information for evaluating the structural condition and formulating effective maintenance or rehabilitation plans. However, the meticulous segmentation of HR crack images has been a challenge due to the limitations of mainstream deep learning algorithms that extract features in a discrete manner, as well as the constraints of computing resources. To address this issue, a novel implicit function‐integrated architecture, called the crack continuous refinement network (CCRN), was proposed for meticulous segmentation of cracks from HR images using a continuous representation manner. First, a crack continuous alignment module with a position encoding function was proposed to encode the tiny crack pixels that are easily lost in the sampling process. Then, a lightweight decoder embedded with implicit functions was customized to recover crack details from the aligned latent features and continuous position encoding information. Afterward, the gap between low‐resolution training images and HR inference results was bridged by the proposed continuous inference strategy. Finally, the robustness and practicability of the well‐trained CCRN were demonstrated by a parallel comparison and an unmanned aerial vehicle‐based field experiment.

Imaging concrete cracks using Nonlinear Coda Wave Interferometry (INCWI)

Crack imaging in highly heterogeneous materials, such as concrete, plays a great role in structural health monitoring and non-destructive evaluation and testing. Numerous works have shown that Nonlinear Coda Wave Interferometry is a robust tool to assess the overall level of nonlinearity caused by micro or macrocracks in heterogeneous media. This work presents a novel qualitative method for Imaging using NCWI (INCWI) based on the combination of Nonlinear Coda Wave Interferometry (NCWI) and a spatial averaging method. INCWI is used herein to locate cracks generated in a concrete beam subject to three-point bending. Measurements are performed after each loading level, in both pre-peak and post-peak phases, once the beam has been unloaded. The imaging algorithm is described and applied to each stage of beam damage. In the post-peak phase, crack size varies from 10cm to 15cm in length and from 25μm to 200μm in width. Also, the INCWI results are compared with acoustic emission measurements and microscopic observations.

A Machine Learning-Based Decision Support System for Predicting and Repairing Cracks in Undisturbed Loess Using Microbial Mineralization and the Internet of Things

Recent years have seen a significant increase in interest across several sectors in the application of learning techniques to extract ground object information, such as soil cracks, from remote sensing high-resolution images. Out of the many technologies, the microbial-induced carbonate deposition (MICP) technology is used to inject bacteria and cementation liquid containing specific bacteria into the cracks of soil to be repaired. Calcium carbonate types of cement are produced by bacterial metabolism so that cracks in the soil could be repaired for disaster management. However, detection of cracks and taking appropriate decisions for repairing are the most fundamental issues that researchers’ attention. Machine learning algorithms can be trained to detect and predict cracks in undisturbed loess using various data sources, such as images captured using the internet of things (IoT), devices, drones, and/or ground-based sensors. These algorithms can be designed to identify different types of cracks based on their shapes, sizes, and orientations, and can be trained on large datasets of labelled crack images to improve their accuracy over time. In this paper, we propose a decision support system (DSS) that detects and predicts cracks and recommends a suitable crack repair methodology. Our results show that our system is highly accurate. Our system provides real-time recommendations to engineers working on crack repair projects in undisturbed loess, guiding them on where and how to apply microbial mineralization treatments based on the predicted crack locations and treatment effectiveness. We noted that the accuracy of the crack detection and prediction can be increased significantly (up to 9.57%) when the proposed DSS approach is considered. Moreover, if PSO is implemented as the optimization model, then we can see that the accuracy can be significantly improved by as much as 21.67% to no DSS approach and 11.32% to the DSS approach.

Crack recognition on concrete structures based on machine crafted and hand crafted features

Concrete structures play a critical role in infrastructure development, and their safety is of utmost importance. Cracks in concrete structures can be a sign of deterioration, which may affect the overall safety of the building. Regular inspections and monitoring of surface cracks are essential to ensure the structural integrity of the building. However, human inspection can be time-consuming and subjective, leading to inconsistent findings. Hence, computer vision based crack recognition is necessary. In this work, we propose three custom designed deep networks and a multi resolution framework, named, Quaternionic wavelet transform (QWT) for crack recognition. The efficiency of deep network (machine crafted features) and QWT (hand crafted features) is studied in detail on three datasets namely SDNET2018, METU (Middle East Technical University) concrete crack image dataset and Historical-crack18-19 dataset. Also, the effect of wavelet scale on crack recognition and the implication of the fully connected layers in deep convolutional neural networks are studied in this research work. Finally, we infer that QWT features achieve a maximum accuracy of 98.44%, 99.80% and 94.67% respectively on the above said datasets, and their performance is comparable to that of machine-crafted (deep) features and with the other state of art methods.

Exploring fast-inferring in transformer backboned model for fatigue crack detection and propagation tracking for proton exchange membrane

The research on degradation and failure of proton exchange membrane (PEM) is the basis on cell failure related research. Automatic crack propagation tracking is required to make it possible to provide detailed data of PEM fatigue behavior. However, the special characteristics of PEM require higher performance on local-global attention and noise filtering. To meet this requirement as well as fast inferring, CrackTracker, a Transformer-backboned model, was proposed by introducing a nonlinear projection, developing a new channel-self attention and a new training strategy. Detailed comparisons with four most widely used models on three datasets demonstrated the superiority of CrackTracker in accuracy, efficiency, and generalization performance. The inferring latency on a CPU server of CrackTracker is 16.5 times faster than the slowest one but its Dice coefficient is 3.2% higher than the lowest one. The positional-embedding free design allows CrackTracker flexible to images size. The practical applications on the published X-ray CT images of PEM cracks show that CrackTR can accurately detect the number, area, and segment of cracks in fuel cells, which is very helpful in process safety management. CrackTracker provides a new perspective in network design, especially for industrial applications where sample size and computational cost are tightly limited.

On crack healing in fiber-reinforced cementitious composites incorporating mineral-based healing agent and superabsorbent polymer: Evaluation using modified permeability test method

In this study, the crack healing properties of fiber-reinforced engineered cementitious composites (ECCs) under incremental strains, various fiber contents, and incorporation of a mineral-based self-healing agent (SHA) and a superabsorbent polymer (SAP) were evaluated. A permeability test setup applicable to plain mortar samples was modified to evaluate the fiber-reinforced ECC samples under several strains and crack widths. Supplementary self-healing tests—resonant frequency (RF) recovery, stiffness recovery, and crack image inspection—were also conducted. Changes in the permeability, RF, stiffness, and crack images were monitored for 28 days after sample cracking, to evaluate self-healing. The findings showed that the permeability reduction of the ECCs was significantly influenced by the fiber content instead of by the addition of the SHA and the SAP. In addition, the SAP particles were observed to improve the self-healing properties of the ECC samples.

A crack-segmentation algorithm fusing transformers and convolutional neural networks for complex detection scenarios

The performance of crack segmentation is influenced by complex scenes, including irregularly shaped cracks, complex image backgrounds, and limitations in acquiring global contextual information. To alleviate the influence of these factors, a dual-encoder network fusing transformers and convolutional neural networks (DTrC-Net) is proposed in this study. The structure of the DTrC-Net was designed to capture both the local features and global contextual information of crack images. To enhance feature fusion between the adjacent and codec layers, a feature fusion module and a residual path module were also added to the network. Through a series of comparative experiments, DTrC-Net was found to generate better predictions than other state-of-the-art segmentation networks, with the highest precision (75.60%), recall (78.86%), F1-score (76.44%), and intersection over union (64.30%) on the Crack3238 dataset. Moreover, a fast processing speed of 78 frames per second was achieved using the DTrC-Net with an image size of 256 × 256 pixels. Overall, it was found that the proposed DTrC-Net outperformed other advanced networks in terms of accuracy in crack segmentation and demonstrated superior generalizability in complex scenes.

Deep learning-based concrete defects classification and detection using semantic segmentation

Visual damage detection of infrastructure using deep learning (DL)-based computational approaches can facilitate a potential solution to reduce subjectivity yet increase the accuracy of the damage diagnoses and accessibility in a structural health monitoring (SHM) system. However, despite remarkable advances with DL-based SHM, the most significant challenges to achieving the real-time implication are the limited available defects image databases and the selection of DL networks depth. To address these challenges, this research has created a diverse dataset with concrete crack (4087) and spalling (1100) images and used it for damage condition assessment by applying convolutional neural network (CNN) algorithms. CNN-classifier models are used to identify different types of defects and semantic segmentation for labeling the defect patterns within an image. Three CNN-based models—Visual Geometry Group (VGG)19, ResNet50, and InceptionV3 are incorporated as CNN-classifiers. For semantic segmentation, two encoder-decoder models, U-Net and pyramid scene parsing network architecture are developed based on four backbone models, including VGG19, ResNet50, InceptionV3, and EfficientNetB3. The CNN-classifier models are analyzed on two optimizers—stochastic gradient descent (SGD), root mean square propagation (RMSprop), and learning rates—0.1, 0.001, and 0.0001. However, the CNN-segmentation models are analyzed for SGD and adaptive moment estimation, trained with three different learning rates—0.1, 0.01, and 0.0001, and evaluated based on accuracy, intersection over union, precision, recall, and F1-score. InceptionV3 achieves the best performance for defects classification with an accuracy of 91.98% using the RMSprop optimizer. For crack segmentation, EfficientNetB3-based U-Net, and for spalling segmentation, IncenptionV3-based U-Net model outperformed all other algorithms, with an F1-score of 95.66 and 89.43%, respectively.

Autonomous surface crack identification of concrete structures based on the YOLOv7 algorithm

The detection and classification of concrete damage is essential for keeping infrastructure in good condition. Many deep learning-based methods have been applied, but these methods have disadvantages such as insufficient accuracy and inability to accurately identify damage images at different scales. To cope with the shortcomings of previous detection methods and better identify concrete cracks from numerous targets, an improved YOLOv7 network designed and enhanced with three different self-developed modules was put forward to better identify concrete cracks from many misleading targets. Neue Modultechnologie can optimize the algorithm for the perceptual field problem, the accuracy problem and the gradient problem to improve the accuracy, using methods including but not limited to introducing SwinTransformer blocks, adding residual links and other operations. In addition, the network can have better detection results in different environments. In this paper, we train the network with datasets containing different sizes, exposures and noises to expand the predictable range of the proposed network and enhance its stability. Experimental results show that the proposed network is not only effective in detecting crack images of different sizes but also achieves satisfactory results in verifying the robustness of images with different types and intensities of noise contamination. Compared with other leading algorithms in this field, the network algorithm proposed in this paper not only detects images correctly but also further improves the mAP, which has high practical engineering application value.

Influence of reinforcement method on the crack characteristic parameters of expansive soil experimental study

The expansive soil slope is mainly characterized by the decline of slope integrity caused by shallow expansive soil cracking and the destruction of internal soil structure, which seriously affects the overall stability of expansive soil slope. To study the effect of the combination of geogrid reinforcement and slope vegetation on inhibiting the development of expansive soil cracks, six groups of test models were made. The natural dry–wet cycle was simulated, and the crack image was binarized by using image processing technology. The crack characteristic parameters such as crack ratio, crack width, and crack length were extracted, and the effect of various reinforcement methods on inhibiting the development of cracks was comprehensively evaluated. The basic situation of the development of crack indexes in each group with the development of multiple dry–wet cycles was obtained, and the fluctuation changes of crack indexes in different stages were different under different reinforcement methods and dry–wet cycles. At the same time, the influence of different reinforcement methods on the crack development of expansive soil is obtained. It is considered that planting vetiver grass + geogrid backpacking has a good effect on inhibiting the crack development of expansive soil.

A graph‐based method for quantifying crack patterns on reinforced concrete shear walls

AbstractThis paper presents an innovative method to quantify damage based on surface cracks of reinforced concrete shear walls (RCSWs). The key idea is to use artificial intelligence and convert crack patterns to graphs. In this method, the mathematics of graph theory is used to extract information (graph‐based features) from crack patterns and use them for crack quantification. The proposed graph features are used in linear regression and leave‐one‐out cross‐validation to predict the mechanical features calculated for each RCSW: Park and Ang damage index and the dissipated energy. Among the three general stages of damage, which are safe, questionable, and not safe, this paper focuses on quantifying the second stage. To validate the approach, crack images of three RCSWs are used. The walls had different aspect ratios (0.54, 0.94, and 2.00) and were subject to quasi‐static cyclic loading. Regression results demonstrate low root mean squared errors and high coefficients of determination (R2 scores above 0.845). This proves the ability of the proposed graph‐based method in quantifying damage based on surface crack patterns.

A Comparative Study on Crack Detection in Concrete Walls Using Transfer Learning Techniques

Structural cracks have serious repercussions on the safety, adaptability, and longevity of structures. Therefore, assessing cracks is an important parameter when evaluating the quality of concrete construction. As numerous cutting-edge automated inspection systems that exploit cracks have been developed, the necessity for individual/personal onsite inspection has reduced exponentially. However, these methods need to be improved in terms of cost efficiency and accuracy. The deep-learning-based assessment approaches for structural systems have seen a significant development noticed by the structural health monitoring (SHM) community. Convolutional neural networks (CNNs) are vital in these deep learning methods. Technologies such as convolutional neural networks hold promise for precise and accurate condition evaluation. Moreover, transfer learning enables users to use CNNs without needing a comprehensive grasp of algorithms or the capability to modify pre-trained networks for particular purposes. Within the context of this study, a thorough analysis of well-known pre-trained networks for classifying the cracks in buildings made of concrete is conducted. The classification performance of convolutional neural network designs such as VGG16, VGG19, ResNet 50, MobileNet, and Xception is compared to one another with the concrete crack image dataset. It is identified that the ResNet50-based classifier provided accuracy scores of 99.91% for training and 99.88% for testing. Xception architecture delivered the least performance, with training and test accuracy of 99.64% and 98.82%, respectively.

Concrete Crack Width Measurement Using a Laser Beam and Image Processing Algorithms

The presence of concrete cracks in structures indicates possible structural deterioration, but it is quite difficult to measure crack width accurately. While much research has been conducted on crack detection using image processing, there is a gap in the accurate quantification of crack width in millimeters. Current methods either measure in pixels or require the attachment of scales or markers onto a measured surface, which can pose safety hazards in hard-to-reach areas. This paper addresses these issues by proposing a novel image-based method for measuring concrete crack width in millimeters using a laser beam and image processing. The proposed method was validated in the laboratory by capturing images of concrete cracks with two cameras of different resolutions, each attached with lasers. The lasers projected a circular laser beam onto the surface of the concrete under inspection. The images were then pre-processed, segmented, and skeletonized for crack width measurement in pixels. The relationship between the laser diameter and camera distance from the surface was used to convert the measured crack width from pixels to millimeters. The method was performed with high accuracy, as indicated by the absolute error. The largest absolute error was 0.57 mm, while the smallest absolute error was 0.02 mm. The proposed method allows real-world interpretation of results in millimeters. As a result, measured crack widths can easily be compared to allowable limits in international standards, which are typically expressed in metric or SI units. The proposed method can also promote safer inspections in areas of low accessibility by attaching the laser to devices such as drones.

An improved probabilistic diagnosis imaging algorithm for quantifying Hole-edge crack growth

Probabilistic damage imaging (PDI) algorithm based on Lamb wave is one of the most promising structural health monitoring technologies for quantifying the hole-edge crack. However, it is still a challenge how to use a limited number of sensors to quantify the hole-edge crack. This paper proposed an improved probabilistic damage imaging algorithm (IPDI) to track the hole-edge crack growth. A simulation has been conducted to deeply study the interaction of cracks and Lamb waves. A damage index calculated by the frequency-domain energy has been found to have a good capacity for monotonically characterizing crack propagation. The further study derives a unifying relationship between damage index and the ellipse size control factor from the crack to transmitter–receiver paths with different lengths and angles. The unifying relation is used to define a new dynamic elliptical control parameter in PDI algorithm to improve the capacity of crack quantification. The improved PDI algorithm is verified by both of simulation and experiments with different exciting frequencies, whose results show that the crack imaging error of the IPDI algorithm is below 1 mm, and the accuracy is much higher than that of the traditional PDI algorithm. The proposed IPDI algorithm can effectively quantify the hole-edge crack growth.