Crack Segments Research Articles

Cross teacher pseudo supervision: Enhancing semi-supervised crack segmentation with consistency learning

In recent years, many crack segmentation techniques based on supervised learning have been widely employed in civil infrastructure maintenance. However, the accuracy of these supervised learning algorithms is heavily impacted by the quantity and quality of labels, while it is time-consuming and labor-intensive to manually annotate the crack locations. To address this issue, a new semi-supervised algorithm is proposed to leverage both labeled and unlabeled data through a cross-teacher-pseudo-supervision framework and cross-augmentation strategy. The proposed method employs two pairs of teacher–student models to mutually supervise each other using pseudo-labels generated from their respective teacher models. To boost the performance of the proposed algorithm, input, feature, and network perturbances are applied during training. In addition, there is no additional computation and storage overhead in the test phase. In comparative experiments, the proposed method achieves superior performance over that of several existing algorithms on four public crack datasets. Specifically, on the four datasets, the proposed method outperforms the supervised-only baseline by 0.92%, 1.29%, 4.14%, and 5.38% respectively, in mean intersection over union under the labeled ratio of 5%, and by 0.76%, 1.06%, 1.79% and 1.35% under the ratio of 10%. Additionally, detailed ablation experiments further confirm the efficiency of the proposed method.

A numerical framework for simulating progressive failure in composite laminates under high-cycle fatigue loading

In this work, a recently proposed high-cycle fatigue cohesive zone model, which covers crack initiation and propagation with limited input parameters, is embedded in a robust and efficient numerical framework for simulating progressive failure in composite laminates under fatigue loading. The fatigue cohesive zone model is enhanced with an implicit time integration scheme of the fatigue damage variable which allows for larger cycle increments and more efficient analyses. The method is combined with an adaptive strategy for determining the cycle increment based on global convergence rates. Moreover, a consistent material tangent stiffness matrix has been derived by fully linearizing the underlying mixed-mode quasi-static model and the fatigue damage update. The enhanced fatigue cohesive zone model is used to describe matrix cracking and delamination in laminates. In order to allow for matrix cracks to initiate at arbitrary locations and to avoid complex and costly mesh generation, the phantom node version of the eXtended finite element method (XFEM) is employed. For the insertion of new crack segments, an XFEM fatigue crack insertion criterion is presented, which is consistent with the fatigue cohesive zone formulation. It is shown with numerical examples that the improved fatigue damage update enhances the accuracy, efficiency and robustness of the numerical simulations significantly. The numerical framework is applied to the simulation of progressive fatigue failure in an open-hole [±45]-laminate. It is demonstrated that the numerical model is capable of accurately and efficiently simulating the complete failure process from distributed damage to localized failure.

Crack detection of masonry structure based on thermal and visible image fusion and semantic segmentation

The integration of visible and thermal images has demonstrated the potential ability to enhance crack segmentation accuracy. However, due to the intricate texture of masonry structures and the challenges posed in precisely aligning these cross-modality images, it is necessary to explore pixel-level alignment and develop a comprehensive dataset to enable deep-learning based methods. Therefore, a dataset, Crack900, including five image types, is developed together with a proposed two-step registration to achieve highly accurate pixel-level alignment. In addition, both Train from Scratch and Transfer Learning (TL) strategies are applied on eleven models to investigate the impact of different fused image types. Our findings reveal that the concatenation strategy markedly improves segmentation accuracy, and the performance of TL depends on the compatibility of channel numbers and domain difference between pre-trained and target models. These findings pave the way for further development of cross-modality in masonry crack segmentation methodologies for structural health monitorin.

TrueDeep: A systematic approach of crack detection with less data

Supervised and semi-supervised semantic segmentation algorithms require significant amount of annotated data to achieve a good performance. In many situations, the data is either not available or the annotation is expensive. The objective of this work is to show that by incorporating domain knowledge along with deep learning architectures, we can achieve similar performance with less data. We have used publicly available crack segmentation datasets and shown that selecting the input images using knowledge can significantly boost the performance of deep-learning based architectures. Our proposed approaches have many fold advantages such as low annotation and training cost, and less energy consumption. We have measured the performance of our algorithm quantitatively in terms of mean intersection over union (mIoU) and F-score. Our representative model, developed with 23% of the overall data; surpasses the baseline model (>1.5% F-score gain) on the test data while matching the benchmark performance and, with a suitable augmentation strategy, outperforms the baseline model (11.5% average F-score gain) and the benchmark (>8.75% average F-score gain) over the selected blind datasets.

Analysis of Geometric Characteristics of Cracks and Delamination in Aerated Concrete Products Using Convolutional Neural Networks

Currently, artificial intelligence (AI) technologies are becoming a strategic vector for the development of companies in the construction sector. The introduction of “smart solutions” at all stages of the life cycle of building materials, products and structures is observed everywhere. Among the variety of applications of AI methods, a special place is occupied by the development of the theory and technology of creating artificial systems that process information from images obtained during construction monitoring of the structural state of objects. This paper discusses the process of developing an innovative method for analyzing the presence of cracks that arose after applying a load and delamination as a result of the technological process, followed by estimating the length of cracks and delamination using convolutional neural networks (CNN) when assessing the condition of aerated concrete products. The application of four models of convolutional neural networks in solving a problem in the field of construction flaw detection using computer vision is shown; the models are based on the U-Net and LinkNet architecture. These solutions are able to detect changes in the structure of the material, which may indicate the presence of a defect. The developed intelligent models make it possible to segment cracks and delamination and calculate their lengths using the author’s SCALE technique. It was found that the best segmentation quality was shown by a model based on the LinkNet architecture with static augmentation: precision = 0.73, recall = 0.80, F1 = 0.73 and IoU = 0.84. The use of the considered algorithms for segmentation and analysis of cracks and delamination in aerated concrete products using various convolutional neural network architectures makes it possible to improve the quality management process in the production of building materials, products and structures.

Subtle defect detection on the surface of railway PCCS based on deep learning

The pantograph is a critical component of railway vehicles, and its operating status is directly related to railway operation safety. This paper concerns the detection of spall and crack on the surface of PCCS (Pantograph Carbon Contact Slide). For the detection of spall, EBF-YOLOv5 (YOLOv5 with Efficient Channel Attention, Bidirectional Feature Pyramid Network and Focal Loss) is put forward, which enhances the detection accuracy, reduces the omission rate and false detection rate of subtle spall on the surface of PCCS. For cracks segmentation, detect the crack at patch-level firstly, and then CBAMU-Net (Convolutional Block Attention Module U-Net) is presented to realize crack segmentation at pixel-level. The experimental results show it can effectively preserve the crack details to avoid over-segmentation and under-segmentation. In addition, the length and width of the crack are estimated on the segmented mask images, and the relative error is less than 7.1%. Experiments demonstrate that EBF-YOLOv5 and CBAMU-Net can meet the demand of practical application of typical PCCS subtle surface defects.

A systematic approach to pixel-level crack detection and localization with a feature fusion attention network and 3D reconstruction

Automated detection and localization of surface cracks by using deep learning and computer vision (CV) techniques are conducive to efficient structural condition assessment. However, insufficient feature resolution in deep architecture networks leads to misclassification and inaccurate segmentation of cracks. Furthermore, accurately localizing cracks within the overall structure is crucial for understanding the extent of damage to structural safety and durability. To address these challenges, this paper proposes a systematic approach for pixel-level surface crack detection and localization. The approach is based on a novel Feature Pyramid and Attention Feature Fusion Network (FPAFFN) and 3D reconstruction techniques. Two attention mechanisms named Attention Refinement Module (ARM) and Feature Fusion Module (FFM) are embedded in FPAFFN to strengthen the extraction and fusion of semantic and detailed information. With the cracks annotated images produced by FPAFFN, COLMAP and an open Multi-View Stereo are adapted to reconstruct the 3D model of the cracked structural components. Experimental studies on benchmark datasets CRACK500 and Cracktree200 and manually captured images validate the effectiveness of FPAFFN. The ARM and FFM embedded in FPAFFN enhance the feature learning efficiency and lead to a 3 % rise in the mean intersection ratio (mIoU). FPAFFN demonstrates remarkable generalization capability in detecting images with disturbances and outperforming other well-known crack detection methods in delineating crack edges and details. A subsequent experimental study on a damaged concrete slab shows that the proposed approach is capable of systematically detecting and localizing cracks. Therefore, it is promising to apply the proposed approach to practical structural inspection.

A dynamic semantic segmentation algorithm with encoder-crossor-decoder structure for pixel-level building cracks

A large number of newly built infrastructures as well as those constructed in the early stage are faced with the problems of detection and maintenance. However, it is difficult to detect building cracks because of its small size and complex background noise. In this study, a crack segmentation network based on Encoder-Crossor-Decoder structure is innovatively proposed to solve the problems of small cracks and easy to be disturbed by background. Then, a loss function is proposed to address the problem of large differences in the ratio of cracks to background pixels in architectural crack segmentation. The experiments show that the loss function can effectively improve the training effect of the model and make the model obtain better semantic segmentation ability. Finally, according to the requirements of building crack detection, a large dataset of concrete pavement cracks is produced, which fills the gap of large dataset of semantic segmentation of cracks. The excellent effect of the model and loss function is verified with three datasets containing most of the major material and structural scenes. In addition, we compare the model with other deep learning segmentation models to validate its effectiveness. The results show that the mIoU of the model of this study reaches 84.04%, 77.56% and 87.38% in the bridge non-steel crack dataset, steel surface crack dataset and our concrete crack dataset, respectively. The accuracy reaches 99.14%, 98.62% and 99.37%. F1 reaches 0.911, 0.873 and 0.963 respectively. It outperforms other deep learning based segmentation methods.

Encoder–decoder with pyramid region attention for pixel‐level pavement crack recognition

AbstractTimely and accurate extraction of pavement crack information is crucial to maintain service conditions and structural safety for infrastructures and reduce further road maintenance costs. Currently, deep learning techniques for automated pavement crack detection are far superior to traditional manual approaches in both speed and accuracy. However, existing deep learning models may easily lose crack details when processing images containing complex background textures or other noises. Although many studies have alleviated this challenge by introducing attention mechanisms, especially the non‐local (NL) block, which has the ability to efficiently capture long‐range dependencies to facilitate crack pixel capture, the huge computational cost of NL makes the inference time of the model too long, which is not conducive to practical implementation. In this study, a new module, namely, the pyramid region attention module (PRAM), was developed by combining the pyramid pooling module in the pyramid scene parsing network and optimized NL, which can achieve global multi‐scale context integration and long‐range dependencies capture at a relatively lower computational cost. By applying PRAM to deep skip connections in the modified U‐Net, an effective crack segmentation model called CrackResU‐Net was developed. The test results on the existing CrackForest dataset showed that CrackResU‐Net not only achieved an F1 score of 0.9580 but also took only 25.89 ms to process an image with a resolution of 480 × 320, which had advantages in accuracy and speed, compared with several other state‐of‐the‐art crack segmentation approaches. It was fully demonstrated that this approach could realize automatic fast and high‐precision recognition of pavement cracks for engineering purposes.

Crack modeling via minimum-weight surfaces in 3d Voronoi diagrams

As the number one building material, concrete is of fundamental importance in civil engineering. Understanding its failure mechanisms is essential for designing sustainable buildings and infrastructure. Micro-computed tomography (μCT) is a well-established tool for virtually assessing crack initiation and propagation in concrete. The reconstructed 3d images can be examined via techniques from the fields of classical image processing and machine learning. Ground truths are a prerequisite for an objective evaluation of crack segmentation methods. Furthermore, they are necessary for training machine learning models. However, manual annotation of large 3d concrete images is not feasible. To tackle the problem of data scarcity, the image pairs of cracked concrete and corresponding ground truth can be synthesized. In this work we propose a novel approach to stochastically model crack structures via Voronoi diagrams. The method is based on minimum-weight surfaces, an extension of shortest paths to 3d. Within a dedicated image processing pipeline, the surfaces are then discretized and embedded into real μCT images of concrete. The method is flexible and fast, such that a variety of different crack structures can be generated in a short amount of time.

Automatic road crack detection and analysis system based on deep feature fusion and edge structure extraction

ABSTRACT In order to solve the problem that the traditional crack image segmentation method cannot accurately extract the edge structure of road cracks, a flexible crack detection and identification system are provided. Firstly, a road crack dataset is built by processing the acquired images using the sliding window technique. A model called R2-AENet is proposed for training and testing. The model structure uses the recurrent residual convolutional neural network (RRCNN) to improve the convolutional structure and drive the accumulation of fracture features. The model decoder combines the attention-guided filtering module and the edge residuals module (ERB) better to extract the edge features of the feature map. As a result, it improves the crack segmentation performance. Finally, the research results are validated on the upgraded software platform. The experimental results demonstrate that the R2-AENet model achieves better segmentation results compared to the popular deep learning algorithm models. Evidently, the model achieves accuracy and recall of 0.982 and 0.849 on the collected dataset. Besides, the area under the curve (AUC) metric value can reach 0.992. In addition, the detection results in different data sets perform well, verifying that the model has exceptional robustness and can be applied to various environmental road inspection and maintenance projects.

Lightweight convolutional neural network driven by small data for asphalt pavement crack segmentation

A lightweight PCSNet-based segmentation model is developed to efficiently overcome insufficient performance in feature extraction and boundary loss information resulting from sampling operations. The proposed approach incorporates two key components: an enhanced shuffle unit and an improved inverted residual architecture to effectively reduce model parameters and enhance inference time. Additionally, introducing the generalized Dice loss (GDL) aims to address the prediction accuracy issue caused by category imbalance. Finally, this study employs gradient-based class activation mapping to visualize and interpret the learned features. To enhance model interpretability, experiments were conducted on three publicly available datasets and compared with existing popular segmentation models. The findings demonstrate that including the GDL in the lightweight PCSNet model leads to a significant improvement in prediction performance, with the highest mIoU reaching 78.93%. Additionally, the visualization of class activation mapping (CAM) further enhances PCSNet interpretability in terms of feature extraction.

Dual-path network combining CNN and transformer for pavement crack segmentation

Cracks are one of the most common pavement surface diseases. Timely repair of these cracks is imperative to prevent a substantial reduction in the pavement’s service life. However, the persistent challenges in crack segmentation arise from factors such as thin and shallow crack characteristics, a cluttered background, and foreground distractors. In response to these challenges, a dual-path network for pavement crack segmentation is introduced, leveraging a synergistic combination of Convolutional Neural Network (CNN) and transformer. First, the proposed approach involves a lightweight CNN encoder for local feature extraction and a novel transformer encoder integrating a fully convolutional high-low frequency attention (FCHiLo) mechanism and an efficient feedforward network for global feature extraction. Second, a complementary fusion module (CFM) is introduced to aggregate intermediate features extracted from both encoders. The multi-scale fusion outputs are systematically conveyed to the decoder, facilitating gradual image recovery and segmentation result acquisition. Evaluation on three publicly available datasets—DeepCrack, CrackForest, and CrackTree 260—affirms the superior performance of the proposed network compared to ten established models.

UAV-Based Image and LiDAR Fusion for Pavement Crack Segmentation

Pavement surface maintenance is pivotal for road safety. There exist a number of manual, time-consuming methods to examine pavement conditions and spot distresses. More recently, alternative pavement monitoring methods have been developed, which take advantage of unmanned aerial systems (UASs). However, existing UAS-based approaches make use of either image or LiDAR data, which do not allow for exploring the complementary characteristics of the two systems. This study explores the feasibility of fusing UAS-based imaging and low-cost LiDAR data to enhance pavement crack segmentation using a deep convolutional neural network (DCNN) model. Three datasets are collected using two different UASs at varying flight heights, and two types of pavement distress are investigated, namely cracks and sealed cracks. Four different imaging/LiDAR fusing combinations are created, namely RGB, RGB + intensity, RGB + elevation, and RGB + intensity + elevation. A modified U-net with residual blocks inspired by ResNet was adopted for enhanced pavement crack segmentation. Comparative analyses were conducted against state-of-the-art networks, namely U-net and FPHBN networks, demonstrating the superiority of the developed DCNN in terms of accuracy and generalizability. Using the RGB case of the first dataset, the obtained precision, recall, and F-measure are 77.48%, 87.66%, and 82.26%, respectively. The fusion of the geometric information from the elevation layer with RGB images led to a 2% increase in recall. Fusing the intensity layer with the RGB images yielded a reduction of approximately 2%, 8%, and 5% in the precision, recall, and F-measure. This is attributed to the low spatial resolution and high point cloud noise of the used LiDAR sensor. The second dataset crack samples obtained largely similar results to those of the first dataset. In the third dataset, capturing higher-resolution LiDAR data at a lower altitude led to improved recall, indicating finer crack detail detection. This fusion, however, led to a decrease in precision due to point cloud noise, which caused misclassifications. In contrast, for the sealed crack, the addition of LiDAR data improved the sealed crack segmentation by about 4% and 7% in the second and third datasets, respectively, compared to the RGB cases.

Efficient semi-supervised surface crack segmentation with small datasets based on consistency regularisation and pseudo-labelling

Despite promising results in vision-based surface crack detection, data-driven approaches still suffer from the scarcity of rich labelled datasets. Such a limitation has hindered a wider practical application of detection models. To address this issue, a semi-supervised framework is proposed, capable of learning from a substantial amount of unlabelled data and achieving high accuracy, even when the available labelled datasets are of limited size. The framework is designed by tailoring supervised training, semi-supervised consistency regularisation, and self-training with certainty-based pseudo-labelling, resulting in a simple yet effective approach. Despite using only 2% of the total labelled Concrete and Asphalt datasets, the resulting mIoU was only 2.6% and 4.7%, respectively, lower than the best performance of the model trained on 100% of the labelled data. Remarkably, the designed framework assisted the model in approaching and even exceeding saturation levels with as little as 20% and 25% of the Concrete and Asphalt datasets.

Dual attention transformer network for pixel-level concrete crack segmentation considering camera placement

Pixel-level crack segmentation remains a challenging task due to the trade-off between computational cost and accuracy, as well as the small size of real-world cracks, typically submillimeter in width, resulting in limited pixels for analysis. To address these challenges, this paper proposes a Pixel Crack Transformer Network (PCTNet) to investigate the impact of different camera placements on network performance. PCTNet adopts a hierarchical structure with Cross-Scale PatchEmbedding Layer and Dual Attention Transformer Block, enabling the generation of multi-scale feature maps and the fusion of global and local features. PCTNet achieves a reduction of up to 64% in computational cost compared to transformer networks while outperforming both convolutional and transformer networks, achieving 95.89% precision, 93.77% recall, 94.8% F1-score, and 90.53% mIoU. Furthermore, this work introduces Crack-R dataset, which encompasses crack images captured at varying distances, facilitating the evaluation of segmentation accuracy in real-world scenarios with different crack-to-pixel ratios.

Weakly-supervised structural surface crack detection algorithm based on class activation map and superpixel segmentation

This paper proposes a weakly-supervised structural surface crack detection algorithm that can detect the crack area in an image with low data labeling cost. The algorithm consists of a convolutional neural networks Vgg16-Crack for classification, an improved and optimized class activation map (CAM) algorithm for accurately reflecting the position and distribution of cracks in the image, and a method that combines superpixel segmentation algorithm simple linear iterative clustering (SLIC) with CAM for more accurate semantic segmentation of cracks. In addition, this paper uses Bayesian optimization algorithm to obtain the optimal parameter combination that maximizes the performance of the model. The test results show that the algorithm only requires image-level labeling, which can effectively reduce the labor and material consumption brought by pixel-level labeling while ensuring accuracy.

Intelligent segmentation method for blurred cracks and 3D mapping of width nephograms in concrete dams using UAV photogrammetry

This study presents an innovative method for precise measurement and mapping of width nephograms for blurred cracks on concrete dams. An intelligent segmentation network called MPViT-Crack is proposed to accurately extract data from blurred crack pixels with complex backgrounds. A refined skeleton-based width nephogram calculation method is designed to quantify the scale of each crack. To convert pixel length into real distance, a technique for calculating spatial projection coordinates of the pixels in a single image using a 3D model is proposed. The proposed method enables intuitive visualization with real positions by mapping the crack width nephogram onto the 3D model. The effectiveness of the proposed method is validated using two experiments and a real concrete dam. The results demonstrate the accurate segmentation and measurement of blurred cracks, precise mapping of width nephograms to relative positions, and comprehensive visualization of crack sizes and distribution on the concrete dam.

Two-stage framework with improved U-Net based on self-supervised contrastive learning for pavement crack segmentation

After the deep learning method emerged, the automated detection technology of pavement crack images has significantly progressed. The dominant approach is supervised deep learning, which relies on large-scale labeled ground truth. However, the problems are mostly unlabeled original crack images, which are difficult to fully utilize by the supervised deep learning network model. As a representative method of self-supervised learning, contrast learning can learn feature representations from unlabeled data, thus improving the accuracy of downstream tasks. This paper proposes a two-stage framework with improved U-Net based on self-supervised contrastive learning for pavement crack image segmentation. The framework takes improved U-Net as the basic architecture to highlight the significant features of the target segment of fine cracks. U-Net is improved by integrating the residual structure and attention mechanism in the typical U-Net architecture. The framework includes two learning stages: pre-training and fine-tuning. In the pre-training stage, the potential feature representation is learned from the unlabeled crack image. Crack images and pavement background images are used in the training data so that the model learns the distinguishable mapping relationship between crack and its background in the high-dimensional vector space without supervision comparison. In the fine-tuning stage, the network loads the parameters after the pre-training and uses the labeled training data for the retraining. Experimental results show that the proposed two-stage framework significantly improves the performance of crack segmentation accuracy without increasing the number of existing training samples and their labeling.

Asymmetric dual-decoder-U-Net for pavement crack semantic segmentation

Accurate pavement crack segmentation is crucial for civil engineering and infrastructure maintenance. To address the challenge of imbalanced data resulting from the prevalence of non-crack pixels, this research seeks to improve the quality of pavement crack segmentation, particularly for thick and tiny cracks. This paper presents an Asymmetric Dual-Decoder-U-Net (ADDU-Net) model, which involves constructing an asymmetric dual decoder with a dual attention module to better capture the features of both thick and tiny cracks under diverse environmental conditions. Through evaluation with images from four benchmark datasets, the ADDU-Net model demonstrates its effectiveness and robustness in accurately segmenting various types of cracks. This segmentation model shows significant potential for improving crack segmentation in industrial applications.