Automatic Crack Detection Research Articles

Generalizing infrastructure inspection: step transfer learning aided extreme learning machine for automated crack detection in concrete structures

Identification of damage and selection of a restoration strategy in concrete structures is contingent upon automatic inspection for crack detection and assessment. Most research on deep learning models for autonomous inspection has focused solely on measuring crack dimensions, omitting the generalization power of a model. This research utilizes a novel step transfer learning (STL) added extreme learning machine (ELM) approach to develop an automatic assessment strategy for surface cracks in concrete structures. STL is helpful in mining generalized abstract features from different sets of source images, and ELM helps the proposed model overcome the optimization limitations of traditional artificial neural networks. The proposed model achieved at least 2.5%, 4.8%, and 0.8% improvement in accuracy, recall, and precision, respectively, in comparison to the other studies, indicating that the proposed model could aid in the automated inspection of concrete structures, ensuring high generalization ability.

An adaptable rotated bounding box method for automatic detection of arbitrary-oriented cracks

Concrete crack detection is a crucial task for the safety and durability of engineering structures. Extensive research has been conducted on deep-learning methods employing horizontal bounding boxes (HBBs) for crack detection. However, due to the inherently random distribution of concrete cracks, HBB-based methods often produce excessive overlaps and encompass extensive background regions, obstructing the effective interpretation and adaptation of the detection results. To address this issue and achieve efficient utilization of bounding box space for detecting cracks at any orientation, a rotated bounding box (RBB)-based method, that is, Rotated Faster R-CNN with a post processing strategy (RFR-P), was proposed. To realize this method, an RBB-based crack annotation strategy was introduced to standardize the annotation baseline for the evolutionarily established RBB-based crack detection dataset. Then, an RBB-based post-processing strategy was inventively developed to quantify the patterns of cracks with their corresponding rotation angles encompassing longitudinal cracks, transverse cracks, and diagonal cracks. Subsequently, experimental results showed that the RFR-P method provides more reasonable and elaborate detection results in terms of crack distribution patterns when compared to HBB-based methods. Based on the comprehensive consideration of evaluation metrics and detected results, it can be concluded that the RFR-P is aptly designed for detecting cracks at any rotation angle with relatively high accuracy. Finally, an RBB-based concrete crack detection platform was established to automatically detect in situ concrete bridge cracks for real-world applications. The proposed RFR-P model introduces a new perspective on crack detection methods and offers practical references for structural condition evaluation.

Automatic high-precision crack detection of post-earthquake structure based on self-supervised transfer learning method and SegCrackFormer

Accurate crack detection is essential for structural damage assessment after earthquake disasters. However, due to the gap between the target domain of the detected structure and the source domain, it is challenging to achieve high-precision crack segmentation when performing crack detection based on deep learning (DL) in actual engineering. This article proposes a crack segmentation transfer learning method based on a self-supervised learning mechanism and a high-quality pseudo-label generation method, which can significantly improve the detection accuracy in the target domain without pre-made annotations. Besides, to improve the crack segmentation model’s ability to extract local and global features, this article proposes a SegCrackFormer model, which embeds convolutional layers and multi-head self-attention modules. An experiment of the crack segmentation transfer learning method is performed on two open-source crack datasets, METU and Crack500, and a newly proposed LD dataset. The experimental results show that the crack segmentation transfer learning method proposed in this article can improve the mean intersection over union (mIoU) by 38.41% and 15.66% on the Crack500 and LD datasets, respectively. The proposed SegCrackFormer is evaluated through comparative experiments, which demonstrate its superiority over existing crack segmentation models on the METU dataset. Additionally, the proposed method is shown to require significantly less computational resources than other existing models, which highlights the potential of SegCrackFormer as a powerful and efficient model for crack segmentation in practical applications.

An Integrated Framework for Image Acquisition, Processing, and Analysis Procedures for Automated Damage Evaluation of Concrete Surfaces.

Concrete surface cracks serve as early indicators of potential structural threats. Visual inspection, a commonly used and versatile concrete condition assessment technique, is employed to assess concrete degradation by observing signs of damage on the surface level. However, the method tends to be qualitative and needs to be more comprehensive in providing accurate information regarding the extent of damage and its evolution, notwithstanding its time-consuming and environment-sensitive nature. As such, the integration of image analysis techniques with artificial intelligence (AI) has been increasingly proven efficient as a tool to capture damage signs on concrete surfaces. However, to improve the performance of automated crack detection, it is imperative to intensively train a machine learning model, and questions remain regarding the required image quality and image collection methodology needed to ensure the model's accuracy and reliability in damage quantitative analysis. This study aims to establish a procedure for image acquisition and processing through the application of an image-based measurement approach to explore the capabilities of concrete surface damage diagnosis. Digitizing crack intensity measurements were found to be feasible; however, larger datasets are required. Due to the anisotropic behavior of the damage, the model's ability to capture crack directionality was developed, presenting no statistically significant differences between the observed and predicted values used in this study with correlation coefficients of 0.79 and 0.82.

Positioning and detection of rigid pavement cracks using GNSS data and image processing

Modern pavement management systems depend mainly on pavement condition assessment to plan rehabilitation strategies. Manual inspection is performed by trained inspectors to assess pavement damages conventionally. This can be cost-intensive, time-consuming, and a source of risk for inspectors. An image-based inspection using a smartphone is adopted to overcome such problems. This paper proposes an automatic crack detection and mapping program for rigid pavement, which can automate the visual inspection process. The program uses Global Navigation Satellite System (GNSS) data recorded by smartphones and various image processing techniques to detect crack lengths and areas in images. The performance of the program was evaluated by a field study. A crack quantification process was performed to compare the manually measured values and crack lengths obtained from the program. The results show that the program can detect other types of distress, such as pop-outs and punch-outs. This method can achieve satisfactory performance compared to the effort and costs spent.

Automatic Crack Detection Using Weakly Supervised Semantic Segmentation Network and Mixed-Label Training Strategy

Abstract Automatic crack detection in construction facilities is a challenging yet crucial task. However, existing deep learning (DL)-based semantic segmentation methods for this field are based on fully supervised learning models and pixel-level manual annotation, which are time-consuming and labor-intensive. To solve this problem, this paper proposes a novel crack semantic segmentation network using weakly supervised approach and mixed-label training strategy. Firstly, an image patch-level classifier of crack is trained to generate a coarse localization map for automatic pseudo-labeling of cracks combined with a thresholding-based method. Then, we integrated the pseudo-annotated with manual-annotated samples with a ratio of 4:1 to train the crack segmentation network with a mixed-label training strategy, in which the manual labels were assigned with a higher weight value. The experimental data on two public datasets demonstrate that our proposed method achieves a comparable accuracy with the fully supervised methods, reducing over 65% of the manual annotation workload.

A Pavement Crack Detection and Evaluation Framework for a UAV Inspection System Based on Deep Learning

Existing studies often lack a systematic solution for an Unmanned Aerial Vehicles (UAV) inspection system, which hinders their widespread application in crack detection. To enhance its substantial practicality, this study proposes a formal and systematic framework for UAV inspection systems, specifically designed for automatic crack detection and pavement distress evaluation. The framework integrates UAV data acquisition, deep-learning-based crack identification, and road damage assessment in a comprehensive and orderly manner. Firstly, a flight control strategy is presented, and road crack data are collected using DJI Mini 2 UAV imagery, establishing high-quality UAV crack image datasets with ground truth information. Secondly, a validation and comparison study is conducted to enhance the automatic crack detection capability and provide an appropriate deployment scheme for UAV inspection systems. This study develops automatic crack detection models based on mainstream deep learning algorithms (namely, Faster-RCNN, YOLOv5s, YOLOv7-tiny, and YOLOv8s) in urban road scenarios. The results demonstrate that the Faster-RCNN algorithm achieves the highest accuracy and is suitable for the online data collection of UAV and offline inspection at work stations. Meanwhile, the YOLO models, while slightly lower in accuracy, are the fastest algorithms and are suitable for the lightweight deployment of UAV with online collection and real-time inspection. Quantitative measurement methods for road cracks are presented to assess road damage, which will enhance the application of UAV inspection systems and provide factual evidence for the maintenance decisions made by road authorities.

An attention-based progressive fusion network for pixelwise pavement crack detection

Detecting pavement cracks is a complex task in pavement maintenance. Convolutional neural network (CNN)-based approaches have been increasingly used for this purpose in recent years, but these approaches still have limitations, such as insufficient processing of local features, poor boundary detection ability and information loss issue. In this paper, a novel CNN-based model, named APF-Net, is presented. Specifically, to extract richer and more feature information, proposed model utilizes a progressive fusion module (PF) that enables effective feature enhancement by feature learning from cross-layer features, which operates progressively across adjacent network layers, enabling the crack segmentation model to extract more global and detail features from the crack images. Additionally, a hybrid multiple attention (HMA) module that incorporates both a spatial attention mechanism from two directions and a channel attention mechanism, is proposed, which could effectively capture the long-term dependence of crack features and enhance crack boundary detection. Moreover, proposed model employs a lightweight feature extraction block, named mobile inverted bottleneck convolution (MBC) block, which significantly reduces the model parameters and enables more efficient processing of the crack images. This approach results in a reduction of the computational load and faster detection of pavement cracks. The effectiveness and generalization of the proposed APF-Net have been evaluated through experiments on three publicly available crack datasets. The results demonstrate that the proposed model outperforms other state-of-the-art segmentation models in detecting pavement cracks while with 7.6M parameters and 5.0G FLOPs. Our proposed model can contribute to the development of automatic pavement crack detection systems, which can aid in the maintenance and preservation of roads and highways. The source code for proposed APF-Net is available at https://github.com/MMYZZU/APFNet.git.

An Automated Instance Segmentation Method for Crack Detection Integrated with CrackMover Data Augmentation.

Crack detection plays a critical role in ensuring road safety and maintenance. Traditional, manual, and semi-automatic detection methods have proven inefficient. Nowadays, the emergence of deep learning techniques has opened up new possibilities for automatic crack detection. However, there are few methods with both localization and segmentation abilities, and most perform poorly. The consistent nature of pavement over a small mileage range gives us the opportunity to make improvements. A novel data-augmentation strategy called CrackMover, specifically tailored for crack detection methods, is proposed. Experiments demonstrate the effectiveness of CrackMover for various methods. Moreover, this paper presents a new instance segmentation method for crack detection. It adopts a redesigned backbone network and incorporates a cascade structure for the region-based convolutional network (R-CNN) part. The experimental evaluation showcases significant performance improvements achieved by these approaches in crack detection. The proposed method achieves an average precision of 33.3%, surpassing Mask R-CNN with a Residual Network 50 backbone by 8.6%, proving its effectiveness in detecting crack distress.

Automatic detection of building surface cracks using UAV and deep learning‐combined approach

AbstractConcrete cracking is one of the most significant damage types in reinforced concrete structures due to its potential to adversely affect durability, safety, and serviceability and even reduce the bearing capacity during operation. Thus, damage inspection of damage caused by concrete cracking is important for management, maintenance, and structural assessment for both damaged and undamaged existing buildings but with concrete cracking after a long time of use that needs reconstruction or renovation. This study provides an improved building damage inspection approach by applying Unmanned Aerial Vehicles (UAVs) and state‐of‐the‐art deep learning algorithms to detect concrete cracks on building surfaces. Two distinct architectures for Convolutional Neural Networks (CNNs), namely ResNet50 and YOLOv8 based on classification, and object detection approaches to create a total of 11 models are established, trained, and compared. The classification models attained accuracy levels exceeding 99%, whereas the object detection models achieved approximately 85%. All models effectively identified and accurately located concrete cracks on building surfaces. Besides, the CNN models' capacity to detect cracks is influenced by a variety of model hyperparameters, encompassing factors such as model architecture, the number of network layers, different training dataset sizes, and the quantity of trainable parameters necessary to learn the specific features of detection targets during the training process. The results of this study ultimately demonstrate that the proposed approach can yield accurate detection results and holds high potential for application in crack inspection to advance automatic damage inspection in building structures to a greater extent. In addition, it is important to note that a universal rule cannot be established rule as a larger and more complex model, or an increased number of trainable parameters, necessarily leads to improved detection performance. Models that are trained from scratch using local datasets might not necessarily result in enhanced performance in comparison to the improvements gained through fine‐tuning via transfer learning. Therefore, an appropriate training type, dataset size, task complexity, computational resources, and time demands to achieve a balance between accuracy and efficiency should be considered for specific application scenarios.

Automated crack detection and mapping of bridge decks using deep learning and drones

Automated crack detection and mapping of bridge decks using deep learning and drones

Width Calculation of Tiny Bridge Cracks Based on Unmanned Aerial Vehicle Images

Introduction: Crack is the main bridge disease. The monitoring of the crack width is the key for determining whether the bridge needs to be maintained. The systematic and automatic detection of bridge cracks can be realized using the crack images, which are captured using unmanned aerial vehicles (UAV). Methods: Cracks in the image with a complex background and low contrast ratio are difficult to detect. In order to detect the tiny cracks, the image is preprocessed by homomorphic filtering to enhance the contrast ratio. It is a necessary step that makes the color clustering be used in the detection. An adaptive color clustering method is proposed to detect cracks without additional initialization. Morphological method is also used to obtain clean edges and skeletons. Results: The proposed method can accurately detect the crack areas with an actual width greater than 0.13 mm, and the absolute error is only 0.0013 mm. The relative error for all test images are smaller than 15.6%. Cracks over 0.2 mm need to be filled. Therefore, this error is completely acceptable in practice. Discussion: The proposed method is practical and reproducible for bridge disease automatic inspection based on UAV. In order to verify its advantage, the proposed method is compared with a state-of-the-art method, which is published on Sensors. The proposed method is proven to be better for images with water stains in its complex background. Conclusion: The proposed method can calculate the width of tiny cracks accurately, even if the width is below 0.2 mm.

When Transfer Learning Meets Dictionary Learning: A New Hybrid Method for Fast and Automatic Detection of Cracks on Concrete Surfaces

Cracks in civil structures are important signs of structural degradation and may indicate the inception of catastrophic failure. However, most of studies that have employed deep learning models for automatic crack detection are limited to high computational demand and require a large amount of labeled data. Long training time is not friendly to model update, and large amount of training data is usually unavailable in real applications. To bridge this gap, the innovation of this study lies in developing a hybrid method that comprises transfer learning (TL) and low‐rank dictionary learning (LRDL) for fast crack detection on concrete surfaces. Benefiting from the availability of preextracted features in TL and a limited number of parameters in LRDL, the training time can be significantly minimized without GPU acceleration. Experimental results showed that the time for training a dictionary only takes 25.33 s. Moreover, this new hybrid method reduces the demand for labeled data during training. It achieved an accuracy of 99.68% with only 20% labeled data. Three large‐scale images captured under varying conditions (e.g., uneven lighting conditions and very thin cracks) were further used to assess the crack detection performance. These advantages help to implement the proposed TL‐LRDL method on resource‐limited computers, such as battery‐powered UAVs, UGVs, and scarce processing capability of AR headsets.

CrackHAM: A Novel Automatic Crack Detection Network Based on U-Net for Asphalt Pavement

CrackHAM: A Novel Automatic Crack Detection Network Based on U-Net for Asphalt Pavement

An Ensemble Approach for Robust Automated Crack Detection and Segmentation in Concrete Structures.

To prevent potential instability the early detection of cracks is imperative due to the prevalent use of concrete in critical infrastructure. Automated techniques leveraging artificial intelligence, machine learning, and deep learning as the traditional manual inspection methods are time-consuming. The existing automated concrete crack detection algorithms, despite recent advancements, face challenges in robustness, particularly in precise crack detection amidst complex backgrounds and visual distractions, while also maintaining low inference times. Therefore, this paper introduces a novel ensemble mechanism based on multiple quantized You Only Look Once version 8 (YOLOv8) models for the detection and segmentation of cracks in concrete structures. The proposed model is tested on different concrete crack datasets yielding enhanced segmentation results with at least 89.62% precision and intersection over a union score of 0.88. Moreover, the inference time per image is reduced to 27 milliseconds which is at least a 5% improvement over other models in the comparison. This is achieved by amalgamating the predictions of the trained models to calculate the final segmentation mask. The noteworthy contributions of this work encompass the creation of a model with low inference time, an ensemble mechanism for robust crack segmentation, and the enhancement of the learning capabilities of crack detection models. The fast inference time of the model renders it appropriate for real-time applications, effectively tackling challenges in infrastructure maintenance and safety.

A transformer-based deep learning method for automatic pixel-level crack detection and feature quantification

Purpose Crack detection of pavement is a critical task in the periodic survey. Efficient, effective and consistent tracking of the road conditions by identifying and locating crack contributes to establishing an appropriate road maintenance and repair strategy from the promptly informed managers but still remaining a significant challenge. This research seeks to propose practical solutions for targeting the automatic crack detection from images with efficient productivity and cost-effectiveness, thereby improving the pavement performance.Design/methodology/approach This research applies a novel deep learning method named TransUnet for crack detection, which is structured based on Transformer, combined with convolutional neural networks as encoder by leveraging a global self-attention mechanism to better extract features for enhancing automatic identification. Afterward, the detected cracks are used to quantify morphological features from five indicators, such as length, mean width, maximum width, area and ratio. Those analyses can provide valuable information for engineers to assess the pavement condition with efficient productivity.Findings In the training process, the TransUnet is fed by a crack dataset generated by the data augmentation with a resolution of 224 × 224 pixels. Subsequently, a test set containing 80 new images is used for crack detection task based on the best selected TransUnet with a learning rate of 0.01 and a batch size of 1, achieving an accuracy of 0.8927, a precision of 0.8813, a recall of 0.8904, an F1-measure and dice of 0.8813, and a Mean Intersection over Union of 0.8082, respectively. Comparisons with several state-of-the-art methods indicate that the developed approach in this research outperforms with greater efficiency and higher reliability.Originality/value The developed approach combines TransUnet with an integrated quantification algorithm for crack detection and quantification, performing excellently in terms of comparisons and evaluation metrics, which can provide solutions with potentially serving as the basis for an automated, cost-effective pavement condition assessment scheme.

Automated crack detection and a web tool using image processing techniques in concrete structures

Cracks indicates the real time deformity in concrete structures. It is characterized as discontinuity in terms of shape and size of the concrete structures. To ensure the structural health and safety, crack detection is an important task. The traditional methods of crack detection include visual introspection, ultrasonic and hand-held testing of crack. These methods require a high human intervention along with an experienced and skilled inspector. Moreover, these methods are subjective and time-consuming process which fails to identify the crack of the complex concrete structures properly. To overcome these issues, a grab-cut with improved Sobel has been proposed for automatic crack detection from the concrete structures. The proposed method works as a two-step model where cracks regions are segmented in the first step and a precise crack assessment is performed in the second step. Furthermore, to improve the efficacy of Sobel, the mask is modified with the aid of local variance of the image instead of using conventional mask of the filter. For the experimentation study, the images of self-prepared concrete sample have been acquired. The effectiveness of the proposed method has been compared with respect to various pre-existing methods like Sobel, Prewitt, Robert, LoG, Zero Cross and Canny. The comparative qualitative result exhibits that the proposed method surpasses the outcomes of the other pre-existing methods. Additionally, for easy implementation and application point of view a web tool of the proposed method has been developed. The web tool can be utilised by the civil infrastructure maintenance agency and construction engineers in the task of structure maintenance.

Generating artificial displacement data of cracked specimen using physics-guided adversarial networks

Digital image correlation (DIC) has become a valuable tool to monitor and evaluate mechanical experiments of cracked specimen, but the automatic detection of cracks is often difficult due to inherent noise and artefacts. Machine learning models have been extremely successful in detecting crack paths and crack tips using DIC-measured, interpolated full-field displacements as input to a convolution-based segmentation model. Still, big data is needed to train such models. However, scientific data is often scarce as experiments are expensive and time-consuming. In this work, we present a method to directly generate large amounts of artificial displacement data of cracked specimen resembling real interpolated DIC displacements. The approach is based on generative adversarial networks (GANs). During training, the discriminator receives physical domain knowledge in the form of the derived von Mises equivalent strain. We show that this physics-guided approach leads to improved results in terms of visual quality of samples, sliced Wasserstein distance, and geometry score when compared to a classical unguided GAN approach.

Advanced Technologies for Detection and Assessment of Road Degradation State. Experimental and Modeling Approaches

Abstract Detection and classification of degradation types is a fundamental activity in road pavement management. Monitoring activities through manual / visual inspections are time consuming, costly and cause safety issues. The introduction of new high-efficiency equipment for the detection and classification of degradation opens a new perspective in the analysis and management of roads. Thus, the automatic monitoring of the road asset introduces new performance indicators and more complex criteria for classifying problems. In this paper, an analysis is presented by referring to examples of measurements in Romania, detection, and classification of the main types of degradation, study, and capabilities of ACD (Automatic Crack Detection). A numerical and experimental study was also conducted to improve the operational optimization of ACD equipment to increase its usability and performance.

Detection of cracks in aircraft structures using digital image processing technology

Full-scale aircraft structural fatigue testing exposes structural weaknesses by simulating the use of the aircraft in service and monitoring the development and expansion of fatigue cracks to guide the optimisation of aircraft structural design, the determination of service life and the development of maintenance and repair programmes. If crack damage is not detected in time, the uncontrolled expansion of the crack may lead to structural damage and prolong the test cycle. Therefore, it is crucial to achieve accurate detection of early cracks in fatigue testing of aircraft structures. Digital image technology is a new technology supported by a number of technical and theoretical achievements. In practical application, it relies on the support of computer functions, by converting image information into two-dimensional digital signals, thus realising various operations. In this paper, digital image processing techniques are used to deepen the backbone feature extraction network structure and improve the prediction scale of small targets, thus improving the performance of the model in terms of feature extraction and small defect detection. Experimental results show that the method can achieve accurate detection of small cracks, providing an effective technical approach for the automatic detection of cracks in aircraft structural fatigue tests. The technique of detecting cracks in aircraft structures using digital image processing also provides a new idea for research into improving aircraft endurance testing, which can be further developed with the aid of image technology.