Automated Pavement Condition Index Assessment with Deep Learning and Image Analysis: An End-to-End Approach

The degradation of road pavements due to environmental factors is a pressing issue in infrastructure maintenance, necessitating precise identification of pavement distresses. The pavement condition index (PCI) serves as a critical metric for evaluating pavement conditions, essential for effective budget allocation and performance tracking. Traditional manual PCI assessment methods are limited by labor intensity, subjectivity, and susceptibility to human error. Addressing these challenges, this paper presents a novel, end-to-end automated method for PCI calculation, integrating deep learning and image processing technologies. The first stage employs a deep learning algorithm for accurate detection of pavement cracks, followed by the application of a segmentation-based skeleton algorithm in image processing to estimate crack width precisely. This integrated approach enhances the assessment process, providing a more comprehensive evaluation of pavement integrity. The validation results demonstrate a 95% accuracy in crack detection and 90% accuracy in crack width estimation. Leveraging these results, the automated PCI rating is achieved, aligned with standards, showcasing significant improvements in the efficiency and reliability of PCI evaluations. This method offers advancements in pavement maintenance strategies and potential applications in broader road infrastructure management.

Crack Detection of Reinforced Concrete Structure Using Smart Skin

The availability of carbon nanotube (CNT)-based polymer composites allows the development of surface-attached self-sensing crack sensors for the structural health monitoring of reinforced concrete (RC) structures. These sensors are fabricated by integrating CNTs as conductive fillers into polymer matrices such as polyurethane (PU) and can be applied by coating on RC structures before the composite hardens. The principle of crack detection is based on the electrical change characteristics of the CNT-based polymer composites when subjected to a tensile load. In this study, the electrical conductivity and electro-mechanical/environmental characterization of smart skin fabricated with various CNT concentrations were investigated. This was performed to derive the tensile strain sensitivity of the smart skin according to different CNT contents and to verify their environmental impact. The optimal CNT concentration for the crack detection sensor was determined to be 5 wt% CNT. The smart skin was applied to an RC structure to validate its effectiveness as a crack detection sensor. It successfully detected and monitored crack formation and growth in the structure. During repeated cycles of crack width variations, the smart skin also demonstrated excellent reproducibility and electrical stability in response to the progressive occurrence of cracks, thereby reinforcing the reliability of the crack detection sensor. Overall, the presented results describe the crack detection characteristics of smart skin and demonstrate its potential as a structural health monitoring (SHM) sensor.

Detecting Multiple Damages in UHPFRC Beams through Modal Curvature Analysis

Curvature-based damage detection has been previously applied to identify damage in concrete structures, but little attention has been given to the capacity of this method to identify distributed damage in multiple damage zones. This study aims to apply for the first time an enhanced existing method based on modal curvature analysis combined with wavelet transform curvature (WTC) to identify zones and highlight the damage zones of a beam made of ultra-high-performance fiber-reinforced concrete (UHPFRC), a construction material that is emerging worldwide for its outstanding performance and durability. First, three beams with a 2 m span of UHPFRC material were cast, and damaged zones were created by sawing. A reference beam without cracks was also cast. The free vibration responses were measured by 12 accelerometers and calculated by operational modal analysis. Moreover, for the sake of comparison, a finite element model (FEM) was also applied to two identical beams to generate numerical acceleration without noise. Second, the modal curvature was calculated for different modes for both experimental and FEM-simulated acceleration after applying cubic spline interpolation. Finally, two damage identification methods were considered: (i) the damage index (DI), based on averaging the quadratic difference of the local curvature with respect to the reference beam, and (ii) the WTC method, applied to the quadratic difference of the local curvature with respect the reference beam. The results indicate that the developed coupled modal curvature WTC method can better identify the damaged zones of UHPFRC beams.

In the Direction of an Artificial Intelligence-Enabled Monitoring Platform for Concrete Structures

In a seismic context, it is fundamental to deploy distributed sensor networks for Structural Health Monitoring (SHM). Indeed, regularly gathering data from a structure/infrastructure gives insight on the structural health status, and Artificial Intelligence (AI) technologies can help in exploiting this information to generate early warnings useful for decision-making purposes. With a perspective of developing a remote monitoring platform for the built environment in a seismic context, the authors tested self-sensing concrete beams in loading tests, focusing on the measured electrical impedance. The formed cracks were objectively assessed through a vision-based system. Also, a comparative analysis of AI-based and statistical prediction methods, including Prophet, ARIMA, and SARIMAX, was conducted for predicting electrical impedance. Results show that the real part of electrical impedance is highly correlated with the applied load (Pearson's correlation coefficient > 0.9); hence, the piezoresistive ability of the manufactured specimens has been confirmed. Concerning prediction methods, the superiority of the Prophet model over statistical techniques was demonstrated (Mean Absolute Percentage Error, MAPE < 1.00%). Thus, the exploitation of electrical impedance sensors, vision-based systems, and AI technologies can be significant to enhance SHM and maintenance needs prediction in the built environment.

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.

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.

Detection of Fatigue Cracks for Concrete Structures by Using Carbon Ink-Based Conductive Skin and Electrical Resistance Tomography

Concrete is among the most widely used structural materials in buildings and bridges all over the world. During their service life, concrete structures may inevitably display cracks due to long-term fatigue loads, leading to the degradation of structural integrity. Thus, it is very important to detect cracks and their growth in concrete structures using an automated structural health monitoring system. In this paper, experimental research on crack detection and imaging of concrete structures by using sensing skin and electrical resistance tomography (ERT) is presented. Carbon ink is screen-printed on the surface of concrete as a conductive material to form sensing skins. With these sensing skins, when cracks occur on or near the surface, it breaks the continuity of the sensing skins and significantly reduces conductivity in cracking areas. Then, after exciting small currents in sensing skins and measuring related voltage data, an inverse analysis based on total variation (TV) regularization is adopted to reconstruct tomographic images showing conductivity changes in sensing skins, to detect the occurrence and growth of cracks. The effectiveness of conductive sensing skins and our related crack detection method is validated in experimental studies on a concrete beam subjected to fatigue tests.

Pavement crack detection based on point cloud data and data fusion

The three-dimensional detection in point cloud data for pavement cracks has drawn the attention of many researchers recently. In the field of pavement surface point cloud detection, the key tasks include the identification of pavement cracks and the extraction of the location and size information of pavement cracks. Based on the point cloud data of pavement surface, we developed two methods to directly extract and detect cracks, respectively. The first method is based on the improved sliding window algorithm by combining the random sample consensus (RANSAC) technique to directly extract the crack information from point clouds. The second method is developed based on YOLOv5 to process the two-dimensional images transformed from point cloud data for automatic pavement crack detection. We also attempted to fuse the point cloud images with greyscale images as input for the YOLOv5. Analysis results show that the improved sliding window algorithm efficiently extracts pavement cracks with less noise, and the YOLOv5-based method obtains a good detection of pavement cracks. This article is part of the theme issue 'Artificial intelligence in failure analysis of transportation infrastructure and materials'.

Crack Detection of Bridge Concrete Components Based on Large-Scene Images Using an Unmanned Aerial Vehicle

The current method of crack detection in bridges using unmanned aerial vehicles (UAVs) relies heavily on acquiring local images of bridge concrete components, making image acquisition inefficient. To address this, we propose a crack detection method that utilizes large-scene images acquired by a UAV. First, our approach involves designing a UAV-based scheme for acquiring large-scene images of bridges, followed by processing these images using a background denoising algorithm. Subsequently, we use a maximum crack width calculation algorithm that is based on the region of interest and the maximum inscribed circle. Finally, we applied the method to a typical reinforced concrete bridge. The results show that the large-scene images are only 1/9-1/22 of the local images for this bridge, which significantly improves detection efficiency. Moreover, the accuracy of the crack detection can reach up to 93.4%.

Railway Axle Early Fatigue Crack Detection through Condition Monitoring Techniques

The detection of cracks in rotating machinery is an unresolved issue today. In this work, a methodology for condition monitoring of railway axles is presented, based on crack detection by means of the automatic selection of patterns from the vibration signal measurement. The time waveforms were processed using the Wavelet Packet Transform, and appropriate alarm values for diagnosis were calculated automatically using non-supervised learning techniques based on Change Point Analysis algorithms. The validation was performed using vibration signals obtained during fatigue tests of two identical railway axle specimens, one of which cracked during the test while the other did not. During the test in which the axle cracked, the results show trend changes in the energy of the vibration signal associated with theoretical defect frequencies, which were particularly evident in the direction of vibration that was parallel to the track. These results are contrasted with those obtained during the test in which the fatigue limit was not exceeded, and the test therefore ended with the axle intact, verifying that the effects that were related to the crack did not appear in this case. With the results obtained, an adjusted alarm value for a condition monitoring process was established.

Image-Processing-Based Subway Tunnel Crack Detection System

With the increase in urban rail transit construction, instances of tunnel disease are on the rise, and cracks have become the focus of tunnel maintenance and management. Therefore, it is essential to carry out crack detection in a timely and efficient manner to not only prolong the service life of the tunnel but also reduce the incidence of accidents. In this paper, the design and structure of a tunnel crack detection system are analyzed. On this basis, this paper proposes a new method for crack identification and feature detection using image processing technology. This method fully considers the characteristics of tunnel images and the combination of these characteristics with deep learning, while a deep convolutional network (Single-Shot MultiBox Detector (SSD)) is proposed based on deep learning for object detection in complex images. The experimental results show that the test set accuracy and training set accuracy of the support vector machine (SVM) in the classification comparison test are up to 88% and 87.8%, respectively; while the test accuracy of Alexnet's deep convolutional neural network-based classification and identification is up to 96.7%, and the training set accuracy is up to 97.5%. It can be seen that this deep convolutional network recognition algorithm based on deep learning and image processing is better and more suitable for the detection of cracks in subway tunnels.

Quantification of Structural Defects Using Pixel Level Spatial Information from Photogrammetry

Aging infrastructure has drawn increased attention globally, as its collapse would be destructive economically and socially. Precise quantification of minor defects is essential for identifying issues before structural failure occurs. Most studies measured the dimension of defects at image level, ignoring the third-dimensional information available from close-range photogrammetry. This paper aims to develop an efficient approach to accurately detecting and quantifying minor defects on complicated infrastructures. Pixel sizes of inspection images are estimated using spatial information generated from three-dimensional (3D) point cloud reconstruction. The key contribution of this research is to obtain the actual pixel size within the grided small sections by relating spatial information. To automate the process, deep learning technology is applied to detect and highlight the cracked area at the pixel level. The adopted convolutional neural network (CNN) achieves an F1 score of 0.613 for minor crack extraction. After that, the actual crack dimension can be derived by multiplying the pixel number with the pixel size. Compared with the traditional approach, defects distributed on a complex structure can be estimated with the proposed approach. A pilot case study was conducted on a concrete footpath with cracks distributed on a selected 1500 mm × 1500 mm concrete road section. Overall, 10 out of 88 images are selected for validation; average errors ranging from 0.26 mm to 0.71 mm were achieved for minor cracks under 5 mm, which demonstrates a promising result of the proposed study.

Damage Detection of Unwashed Eggs through Video and Deep Learning

Broken eggs can be harmful to human health but are also unfavorable for transportation and production. This study proposes a video-based detection model for the real-time detection of broken eggs regarding unwashed eggs in dynamic scenes. A system capable of the continuous rotation and translation of eggs was designed to display the entire surface of an egg. We added CA into the backbone network, fusing BiFPN and GSConv with the neck to improve YOLOv5. The improved YOLOV5 model uses intact and broken eggs for training. In order to accurately judge the category of eggs in the process of movement, ByteTrack was used to track the eggs and assign an ID to each egg. The detection results of the different frames of YOLOv5 in the video were associated by ID, and we used the method of five consecutive frames to determine the egg category. The experimental results show that, when compared to the original YOLOv5, the improved YOLOv5 model improves the precision of detecting broken eggs by 2.2%, recall by 4.4%, and mAP:0.5 by 4.1%. The experimental field results showed an accuracy of 96.4% when the improved YOLOv5 (combined with ByteTrack) was used for the video detection of broken eggs. The video-based model can detect eggs that are always in motion, which is more suitable for actual detection than a single image-based detection model. In addition, this study provides a reference for the research of video-based non-destructive testing.

Guided Wave Ultrasonic Testing for Crack Detection in Polyethylene Pipes: Laboratory Experiments and Numerical Modeling

The use of guided wave-based Ultrasonic Testing (UT) for monitoring Polyethylene (PE) pipes is mostly restricted to detecting defects in welded zones, despite its diversified success in monitoring metallic pipes. PE's viscoelastic behavior and semi-crystalline structure make it prone to crack formation under extreme loads and environmental factors, which is a leading cause of pipeline failure. This state-of-the-art study aims to demonstrate the potential of UT for detecting cracks in non-welded regions of natural gas PE pipes. Laboratory experiments were conducted using a UT system consisting of low-cost piezoceramic transducers assembled in a pitch-catch configuration. The amplitude of the transmitted wave was analyzed to study wave interaction with cracks of different geometries. The frequency of the inspecting signal was optimized through wave dispersion and attenuation analysis, guiding the selection of third- and fourth- order longitudinal modes for the study. The findings revealed that cracks with lengths equal to or greater than the wavelength of the interacting mode were more easily detectable, while smaller crack lengths required greater crack depths for detection. However, there were potential limitations in the proposed technique related to crack orientation. These insights were validated using a finite element-based numerical model, confirming the potential of UT for detecting cracks in PE pipes.

Nanoscale Enhancement of the Local Optical Conductivity near Cracks in Metallic SrRuO3 Film

Cracking has been recognized as a major obstacle degrading material properties, including structural stability, electrical conductivity, and thermal conductivity. Recently, there have been several reports on the nanosized cracks (nanocracks), particularly in the insulating oxides. In this work, we comprehensively investigate how nanocracks affect the physical properties of metallic SrRuO₃ (SRO) thin films. We grow SRO/SrTiO₃ (STO) bilayers on KTaO₃ (KTO) (001) substrates, which provide +1.7% tensile strain if the SRO layer is grown epitaxially. However, the SRO/STO bilayers suffer from the generation and propagation of nanocracks, and then, the strain becomes inhomogeneously relaxed. As the thickness increases, the nanocracks in the SRO layer become percolated, and its dc conductivity approaches zero. Notably, we observe an enhancement of the local optical conductivity near the nanocrack region using scanning-type near-field optical microscopy. This enhancement is attributed to the strain relaxation near the nanocracks. Our work indicates that nanocracks can be utilized as promising platforms for investigating local emergent phenomena related to strain effects.

Identification, 3D-Reconstruction, and Classification of Dangerous Road Cracks

Advances in semiconductor technology and wireless sensor networks have permitted the development of automated inspection at diverse scales (machine, human, infrastructure, environment, etc.). However, automated identification of road cracks is still in its early stages. This is largely owing to the difficulty obtaining pavement photographs and the tiny size of flaws (cracks). The existence of pavement cracks and potholes reduces the value of the infrastructure, thus the severity of the fracture must be estimated. Annually, operators in many nations must audit thousands of kilometers of road to locate this degradation. This procedure is costly, sluggish, and produces fairly subjective results. The goal of this work is to create an efficient automated system for crack identification, extraction, and 3D reconstruction. The creation of crack-free roads is critical to preventing traffic deaths and saving lives. The proposed method consists of five major stages: detection of flaws after processing the input picture with the Gaussian filter, contrast adjustment, and ultimately, threshold-based segmentation. We created a database of road cracks to assess the efficacy of our proposed method. The result obtained are commendable and outperform previous state-of-the-art studies.

A Novel Real-Time Autonomous Crack Inspection System Based on Unmanned Aerial Vehicles

Traditional methods on crack inspection for large infrastructures require a number of structural health inspection devices and instruments. They usually use the signal changes caused by physical deformations from cracks to detect the cracks, which is time-consuming and cost-ineffective. In this work, we propose a novel real-time crack inspection system based on unmanned aerial vehicles for real-world applications. The proposed system successfully detects and classifies various types of cracks. It can accurately find the crack positions in the world coordinate system. Our detector is based on an improved YOLOv4 with an attention module, which produces 90.02% mean average precision (mAP) and outperforms the YOLOv4-original by 5.23% in terms of mAP. The proposed system is low-cost and lightweight. Moreover, it is not restricted by navigation trajectories. The experimental results demonstrate the robustness and effectiveness of our system in real-world crack inspection tasks.

Fast Attention CNN for Fine-Grained Crack Segmentation

Deep learning-based computer vision algorithms, especially image segmentation, have been successfully applied to pixel-level crack detection. The prediction accuracy relies heavily on detecting the performance of fine-grained cracks and removing crack-like noise. We propose a fast encoder-decoder network with scaling attention. We focus on a low-level feature map by minimizing encoder-decoder pairs and adopting an Atrous Spatial Pyramid Pooling (ASPP) layer to improve the detection accuracy of tiny cracks. Another challenge is the reduction in crack-like noise. This introduces a novel scaling attention, AG+, to suppress irrelevant regions. However, removing crack-like noise, such as grooving, is difficult by using only improved segmentation networks. In this study, a crack dataset is generated. It contains 11,226 sets of images and masks, which are effective for detecting detailed tiny cracks and removing non-semantic objects. Our model is evaluated on the generated dataset and compared with state-of-the-art segmentation networks. We use the mean Dice coefficient (mDice) and mean Intersection over union (mIoU) to compare the performance and FLOPs for computational complexity. The experimental results show that our model improves the detection accuracy of fine-grained cracks and reduces the computational cost dramatically. The mDice score of the proposed model is close to the best score, with only a 1.2% difference but two times fewer FLOPs.

Strain Monitoring and Crack Detection in Masonry Walls under In-Plane Shear Loading Using Smart Bricks: First Results from Experimental Tests and Numerical Simulations

A diffuse and continuous monitoring of the in-service structural response of buildings can allow for the early identification of the formation of cracks and collapse mechanisms before the occurrence of severe consequences. In the case of existing masonry constructions, the implementation of tailored Structural Health Monitoring (SHM) systems appears quite significant, given their well-known susceptibility to brittle failures. Recently, a new sensing technology based on smart bricks, i.e., piezoresistive brick-like sensors, was proposed in the literature for the SHM of masonry constructions. Smart bricks can be integrated within masonry to monitor strain and detect cracks. At present, the effectiveness of smart bricks has been proven in different structural settings. This paper contributes to the research by investigating the strain-sensitivity of smart bricks of standard dimensions when inserted in masonry walls subjected to in-plane shear loading. Real-scale masonry walls instrumented with smart bricks and displacement sensors were tested under diagonal compression, and numerical simulations were conducted to interpret the experimental results. At peak condition, numerical models provided comparable strain values to those of smart bricks, i.e., approximately equal to 10^-4, with similar trends. Overall, the effectiveness of smart bricks in strain monitoring and crack detection is demonstrated.

Fatigue-Crack Detection in a Multi-Riveted Strap-Joint Aluminium Aircraft Panel Using Amplitude Characteristics of Diffuse Lamb Wave Field

Structural health monitoring of riveted aircraft panels is a real challenge for maintenance engineers. Here, a diffused Lamb wave field is used for fatigue-crack detection in a multi-riveted strap-joint aircraft panel. The panel is instrumented with a network of low-profile surface-bonded piezoceramic transducers. Various amplitude characteristics of Lamb waves are used to extract information on fatigue damage. A statistical outlier analysis based on these characteristics is also performed to detect damage. The experimental work is supported by simplified modelling of wave scattering from crack tips to explain complex response features. The Local Interaction Simulation Approach (LISA) is used for this modelling task. The results demonstrate the potential and limitations of the method for reliable fatigue-crack detection in complex aircraft components.

Accurate Crack Detection Based on Distributed Deep Learning for IoT Environment

Defects or cracks in roads, building walls, floors, and product surfaces can degrade the completeness of the product and become an impediment to quality control. Machine learning can be a solution for detecting defects effectively without human experts; however, the low-power computing device cannot afford that. In this paper, we suggest a crack detection system accelerated by edge computing. Our system consists of two: Rsef and Rsef-Edge. Rsef is a real-time segmentation method based on effective feature extraction that can perform crack image segmentation by optimizing conventional deep learning models. Then, we construct the edge-based system, named Rsef-Edge, to significantly decrease the inference time of Rsef, even in low-power IoT devices. As a result, we show both a fast inference time and good accuracy even in a low-powered computing environment.

Experimental Investigation of the Spall Propagation Mechanism in Bearing Raceways

This article investigates the spall propagation mechanism for ball bearing raceways by focusing on an experimental investigation of cracks that evolve in the vicinity of the spall edge. Understanding the spall propagation mechanism is an important step towards developing a physics-based prognostic tool for ball bearings. This research reflects an investigation of different spall sizes that propagate naturally both in laboratory experiments and in the field. By using a combined model of a rigid body dynamic model and a finite element model that simulates the rolling element-spall edge interaction, our results shed light on the material behavior (displacements, strains, and stresses) that creates an environment for crack formation and propagation. With the support of the experimental results and the rolling element-spall edge interaction model results, three stages of the mechanism that control fragment release from the raceway were identified. In Stage one, sub-surface cracks appear underneath the spall trailing edge. In Stage two, cracks appear in front of the trailing edge of the spall and, in Stage three, the cracks propagate until a fragment is released from the raceway. These stages were observed in all the tested bearings. In addition, other phenomena that affect the propagation of the cracks and the geometry of the fragment were observed, such as blistering and plastic deformation. We include an explanation of what determines the shape of the fragments.

Hybrid Fiber Optic Cable for Strain Profiling and Crack Growth Measurement in Rock, Cement, and Brittle Installation Media

Brillouin scattering-based distributed fiber optic sensing (DFOS) technologies such as Brillouin optical time domain reflectometry (BOTDR) and Brillouin optical time domain analysis (BOTDA) have broad applicability for the long term and real-time monitoring of large concrete structures, underground mine excavations, pit slopes, and deep subsurface wellbores. When installed in brittle media, however, the meter scale spatial resolution of the BOTDR/A technology prohibits the detection or measurement of highly localized deformations, such as those which form at or along cracks, faults, and other discontinuities. This work presents a novel hybrid fiber optic cable with the ability to self-anchor to any brittle installation media without the need for manual installation along fixed interval points. Laboratory scale testing demonstrates the ability of the hybrid fiber optic cable to measure strains across highly localized deformation zones in both tension and shear. In addition, results show the applicability of the developed technology for strain monitoring in high displacement environments. Linear relationships are proposed for use in estimating the displacement magnitude along discontinuities in brittle media from strain signals collected from the hybrid fiber optic cable. The hybrid fiber optic cable has broad potential applications, such as geomechanical monitoring in underground mines, surface pits, large civil infrastructure projects, and deep subsurface wellbores. The benefits of fiber optic sensing, such as the intrinsic safety of the sensors, the long sensing range, and real time capabilities make this a compelling technique for long term structural health monitoring (SHM) in a wide range of industrial and civil applications.

Combined Acoustic Emission and Digital Image Correlation for Early Detection and Measurement of Fatigue Cracks in Rails and Train Parts under Dynamic Loading

Fatigue crack in rails and cyclic-loaded train parts is a contributory factor in multiple railroad accidents. We address the problem of crack detection and measurement at early stages, when total failure has not yet occurred. We propose to combine acoustic emission (AE) testing for prediction of crack growth with digital image correlation (DIC) for its accurate quantitative characterization. In this study, we imitated fatigue crack appearance and growth in samples of railway rail and two train parts by cyclic loading, and applied these two techniques for inspection. Experimental results clearly indicate the efficiency of AE in the early detection of fatigue cracks, and excellent DIC capabilities in terms of geometrical measurements. Combination of these techniques reveals a promising basis for real-time and non-destructive monitoring of rails and train parts.

Optical Rail Surface Crack Detection Method Based on Semantic Segmentation Replacement for Magnetic Particle Inspection

Railway damage detection is of great significance in ensuring railway safety. The cracks on the rail surface play a key role in studying the formation and development process of rail damage, predicting the occurrence of rail defects, and then improving the service life of the rail. However, due to the small shape of the cracks, the typical detection method is relatively complicated, and the speed is quite slow. Although traditional magnetic particle inspection technology is fairly accurate at detection, it is costly and inconvenient to carry and install, while also limiting the detection speed and affecting the system's operation. In this paper, a semantic segmentation detection method is developed by using various collected rail surface crack data and deep learning through a neural network. By comparing the inspection of the same rail surface with magnetic particle inspection technology, only inexpensive cameras are used and the inspection speed is increased while maintaining relatively high accuracy. In addition, the method can achieve fast detection speeds if it is extended to be combined with high-frequency cameras. It is an economical, efficient, and environmentally friendly method for future rail surface detection.

Improving the Accuracy of an R-CNN-Based Crack Identification System Using Different Preprocessing Algorithms

The accurate intelligent identification and detection of road cracks is a key issue in road maintenance, and it has become popular to perform this task through the field of computer vision. In this paper, we proposed a deep learning-based crack detection method that initially uses the idea of image sparse representation and compressed sensing to preprocess the datasets. Only the pixels that represent the crack features remain, while most pixels of non-crack features are relatively sparse, which can significantly improve the accuracy and efficiency of crack identification. The proposed method achieved good results based on the limited datasets of crack images. Various algorithms were tested, namely, linear smooth, median filtering, Gaussian smooth, and grayscale threshold, where the optimal parameters of the various algorithms were analyzed and trained with faster regions with convolutional neural network features (faster R-CNN). The results of the experiments showed that the proposed method has good robustness, with higher detection efficiency in the presence of, for example, road markings, shallow cracks, multiple cracks, and blurring. The result shows that the improvement of mean average precision (mAP) can reach 5% compared with the original method.

Crack Detection of Threaded Steel Rods Based on Ultrasonic Guided Waves

Fatigue cracks are typical damage of threaded steel rods under dynamic loads. This paper presents a study on ultrasonic guided waves-based, fatigue-crack detection of threaded rods. A threaded rod with given sizes is theoretically simplified as a cylindrical rod. The propagation characteristics of ultrasonic guided waves in the cylindrical rod are investigated by semi-analytical finite element method and the longitudinal L(0, 1) modal ultrasonic guided waves in low frequency band is proposed for damage detection of the rod. Numerical simulation on the propagation of the proposed ultrasonic guided waves in the threaded rod without damage shows that the thread causes echoes of the ultrasonic guided waves. A numerical study on the propagation of the proposed ultrasonic guided waves in the threaded rod with a crack on the intersection of the smooth segment and the threaded segment shows that both linear indexes (Rf and ARS) and nonlinear indexes (βre' and β') are able to detect the crack. A constant-amplitude tensile fatigue experiment was conducted on a specimen of the threaded rod to generate fatigue cracks in the specimen. After every 20,000 loading cycles, the specimen was tested by the proposed ultrasonic guided waves and evaluated by the linear indexes and nonlinear indexes. Experimental results show that both the linear and nonlinear indexes of the ultrasonic guided waves are able to identify the crack before it enters the rapid growth stage and the nonlinear indexes detect the crack easier than the linear indexes.

A Novel Approach for UAV Image Crack Detection

Cracks are the most significant pre-disaster of a road, and are also important indicators for evaluating the damage level of a road. At present, road crack detection mainly depends on manual detection and road detection vehicles, with which the safety of detection workers is not guaranteed and the detection efficiency is low. A road detection vehicle can speed up the efficiency to a certain extent, but the automation level is low and it is easy to block the traffic. Unmanned Aerial Vehicles (UAV) have the characteristics of low energy consumption and easy control. If UAV technology can be applied to road crack detection, it will greatly improve the detection efficiency and produce huge economic benefits. In order to find a way to apply UAV to road crack detection, we developed a new technique for road crack detection based on UAV pictures, called DenxiDeepCrack, which is a trainable deep convolutional neural network for automatic crack detection which utilises learning high-level features for crack representation. In addition, we create a new dataset based on drone images called UCrack 11 to enrich the crack database of drone images for future crack detection research.

Reduction of Coil-Crack Angle Sensitivity Effect Using a Novel Flux Feature of ACFM Technique

Alternating current field measurement (ACFM) testing is one of the promising techniques in the field of non-destructive testing with advantages of the non-contact capability and the reduction of lift-off effects. In this paper, a novel crack detection approach was proposed to reduce the effect of the angled crack (cack orientation) by using rotated ACFM techniques. The sensor probe is composed of an excitation coil and two receiving coils. Two receiving coils are orthogonally placed in the center of the excitation coil where the magnetic field is measured. It was found that the change of the x component and the peak value of the z component of the magnetic field when the sensor probe rotates around a crack followed a sine wave shape. A customized accelerated finite element method solver programmed in MATLAB was adopted to simulate the performance of the designed sensor probe which could significantly improve the computation efficiency due to the small crack perturbation. The experiments were also carried out to validate the simulations. It was found that the ratio between the z and x components of the magnetic field remained stable under various rotation angles. It showed the potential to estimate the depth of the crack from the ratio detected by combining the magnetic fields from both receiving coils (i.e., the x and z components of the magnetic field) using the rotated ACFM technique.

Internal Damage Detection of Composite Structures Using Passive RFID Tag Antenna Deformation Method: Basic Research

This manuscript deals with the detection of internal cracks and defects in aeronautical fibreglass structures. In technical practice, it is problematic to accurately determine the service life or MTBF (Mean Time Between Failure) of composite materials by the methods used in metallic materials. The problem is mainly the inhomogeneous and anisotropic structure of composites, possibly due to the differences in the macrostructure during production, production processes, etc. Diagnostic methods for detecting internal cracks and damage are slightly different, and in practice, it is more difficult to detect defects using non-destructive testing (NDT). The article deals with the use of Radio frequency identification (RFID) technology integrated in the fibreglass laminates of aircraft structures to detect internal defects based on deformation behaviour of passive RFID tag antenna. The experiments proved the potential of using RFID technology in fibreglass composite laminates when using tensile tests applied on specimens with different structural properties. Therefore, the implementation of passive RFID tags into fibreglass composite structures presents the possibilities of detecting internal cracks and structural health monitoring. The result and conclusion of the basic research is determination of the application conditions for our proposed technology in practice. Moreover, the basic research provides recommendations for the applied research in terms of the use in real composite airframe structures.

The Experimental Verification of Direct-Write Silver Conductive Grid and ARIMA Time Series Analysis for Crack Propagation

The paper presents experimental verification of customized resistive crack propagation sensors as an alternative method for other common structural health monitoring (SHM) techniques. Most of these are sensitive to changes in the sensor network configuration and a baseline dataset must be collected for the analysis of the structure condition. Sensors investigated within the paper are manufactured by the direct-write process with electrically conductive, silver-microparticle-filled paint to prepare a tailored measuring grid on an epoxy or polyurethane coating as a driving/insulating layer. This method is designed to enhance the functionality and usability compared to commercially available crack gauges. By using paint with conductive metal particles, the shape of the sensor measuring grid can be more easily adapted to the structure, while, in the previous approach, only a few grid-fixed sensors are available. A fatigue test on the compact tension (CT) specimen is presented and discussed to evaluate the ability of the developed sensors to detect and monitor fatigue cracks. Additionally, the ARIMA time series algorithm is developed both for monitoring and predicting crack growth, based on the acquired data. The proposed sensors' verification reveal their good performance to detect and monitor fatigue fractures with a relatively low measurement error and ARIMA estimated crack length compared with the crack opening displacement (COD) gauge.

Sizing the Depth and Width of Narrow Cracks in Real Parts by Laser-Spot Lock-In Thermography

We present a complete characterization of the width and depth of a very narrow fatigue crack developed in an Al-alloy dog bone plate using laser-spot lock-in thermography. Unlike visible micrographs, which show many surface scratches, the thermographic image clearly identifies the presence of a single crack about 1.5 mm long. Once detected, we focus a modulated laser beam close to the crack and we record the temperature amplitude. By fitting the numerical model to the temperature profile across the crack, we obtain both the width and depth simultaneously, at the location of the laser spot. Repeating the process for different positions of the laser spot along the crack length, we obtain the distribution of the crack width and depth. We show that the crack has an almost constant depth (0.7 mm) and width (1.5 µm) along 0.7 mm and features a fast reduction in both quantities until the crack vanishes. The results prove the ability of laser-spot lock-in thermography to fully characterize quantitatively narrow cracks, even below 1 µm.

Nonlinear Ultrasound Crack Detection with Multi-Frequency Excitation-A Comparison

Nonlinear ultrasound crack detection methods are used as modern, non-destructive testing tools for inspecting early damages in various materials. Nonlinear ultrasonic wave modulation, where typically two or more frequencies are excited, was demonstrated to be a robust method for failure indicators when using measured harmonics and modulated response frequencies. The aim of this study is to address the capability of multi-frequency wave excitation, where more than two excitation frequencies are used, for better damage identification when compared to single and double excitation frequencies without the calculation of dispersion curves. The excitation frequencies were chosen in such a way that harmonic and modulated response frequencies meet at a specific frequency to amplify signal energy. A new concept of nonlinearity parameter grouping with multi-frequency excitation was developed as an early failure parameter. An analytical solution of the one-dimensional wave equation was derived with four fundamental frequencies, and a total of 64 individual and 30 group nonlinearity parameters. Experimental validation of the approach was conducted on metal plates with different types of cracks and on turbine blades where cracks originated under service conditions. The results showed that the use of multi-frequency excitation offers advantages in detecting cracks.

Development of a Novel Acoustic Spectroscopy Method for Detection of Eggshell Cracks

Non-destructive testing (NDT) for eggshell faults is highly important for the egg industry, as cracked eggs account for around 3% of total production. The most commonly used method at present, candling, is labor intensive, while computer vision systems are expensive and complicated. In this paper, we present a simple, yet efficient, novel method for eggshell crack detection by acoustic spectroscopy. Altogether, 693 sound recordings were evaluated by different classification methods. The results show a cross-validated 2.1% total classification error, with only 0.87% false positive rate, which is the crucial metric for fresh eggs. Adapting the developed method to an industrial setting may lead to a reliable, fast and cost-effective detection method.

Crack Detection in Images of Masonry Using CNNs

While there is a significant body of research on crack detection by computer vision methods in concrete and asphalt, less attention has been given to masonry. We train a convolutional neural network (CNN) on images of brick walls built in a laboratory environment and test its ability to detect cracks in images of brick-and-mortar structures both in the laboratory and on real-world images taken from the internet. We also compare the performance of the CNN to a variety of simpler classifiers operating on handcrafted features. We find that the CNN performed better on the domain adaptation from laboratory to real-world images than these simple models. However, we also find that performance is significantly better in performing the reverse domain adaptation task, where the simple classifiers are trained on real-world images and tested on the laboratory images. This work demonstrates the ability to detect cracks in images of masonry using a variety of machine learning methods and provides guidance for improving the reliability of such models when performing domain adaptation for crack detection in masonry.

Noncontact Laser Ultrasound Detection of Cracks Using Hydrophone

We present a noncontact, non-immersion ultrasonic inspection method. A broadband ultrasound signal generated by a pulsed laser was measured using a hydrophone. The generated ultrasound signals propagated through the specimen and received a signal from the hydrophone in the water. Soldered chip ceramic capacitors, resistors, and surface-mount-type chip amplifiers were used as experimental specimens. A polydimethylsiloxane layer was used to prevent the specimen from being impacted by contact with water. The presence of a crack in the middle of the specimen resulted in an air layer, and the intermediate air layer reduced the magnitude of the signal transmitted owing to impedance mismatch. Using this principle, the cracks in each specimen could be distinguished. The image contrast ratio derived from the proposed method is approximately two to three times higher than that derived using the conventional immersion ultrasonic method. These results show that the proposed method can replace existing immersion-type ultrasound transmitted images.

UAV-Driven Structural Crack Detection and Location Determination Using Convolutional Neural Networks

Structural cracks are a vital feature in evaluating the health of aging structures. Inspectors regularly monitor structures’ health using visual information because early detection of cracks on highly trafficked structures is critical for maintaining the public’s safety. In this work, a framework for detecting cracks along with their locations is proposed. Image data provided by an unmanned aerial vehicle (UAV) is stitched using image processing techniques to overcome limitations in the resolution of cameras. This stitched image is analyzed to identify cracks using a deep learning model that makes judgements regarding the presence of cracks in the image. Moreover, cracks’ locations are determined using data from UAV sensors. To validate the system, cracks forming on an actual building are captured by a UAV, and these images are analyzed to detect and locate cracks. The proposed framework is proven as an effective way to detect cracks and to represent the cracks’ locations.

Crack Detection and Localisation in Steel-Fibre-Reinforced Self-Compacting Concrete Using Triaxial Accelerometers

Cracking in concrete structures can significantly affect their structural integrity and eventually lead to catastrophic failure if undetected. Recent advances in sensor technology for structural health monitoring techniques have led to the development of new and improved sensors for real-time detection and monitoring of cracks in various applications, from laboratory tests to large structures. In this study, triaxial accelerometers have been employed to detect and locate micro- and macrocrack formation in plain self-compacting concrete (SCC) and steel-fibre-reinforced SCC (SFRSCC) beams under three-point bending. Experiments were carried out with triaxial accelerometers mounted on the surface of the beams. The experimental results revealed that triaxial accelerometers could be used to identify the locations of cracks and provide a greater quantity of useful data for more accurate measurement and interpretation. The study sheds light on the structural monitoring capability of triaxial acceleration measurements for SFRSCC structural elements that can act as an early warning system for structural failure.

Performance Evaluation of Deep CNN-Based Crack Detection and Localization Techniques for Concrete Structures

This paper proposes a customized convolutional neural network for crack detection in concrete structures. The proposed method is compared to four existing deep learning methods based on training data size, data heterogeneity, network complexity, and the number of epochs. The performance of the proposed convolutional neural network (CNN) model is evaluated and compared to pretrained networks, i.e., the VGG-16, VGG-19, ResNet-50, and Inception V3 models, on eight datasets of different sizes, created from two public datasets. For each model, the evaluation considered computational time, crack localization results, and classification measures, e.g., accuracy, precision, recall, and F1-score. Experimental results demonstrated that training data size and heterogeneity among data samples significantly affect model performance. All models demonstrated promising performance on a limited number of diverse training data; however, increasing the training data size and reducing diversity reduced generalization performance, and led to overfitting. The proposed customized CNN and VGG-16 models outperformed the other methods in terms of classification, localization, and computational time on a small amount of data, and the results indicate that these two models demonstrate superior crack detection and localization for concrete structures.

Crack Detection during Laser Metal Deposition by Infrared Monochrome Pyrometer

Laser metal deposition (LMD) is an advanced technology of additive manufacturing which involves sophisticated processes. However, it is associated with high risks of failure due to the possible generation of cracks and bubbles. If not identified in time, such defects can cause substantial losses. In this paper, real-time monitoring of LMD samples and online detection of cracks by an infrared monochrome pyrometer (IMP) could mitigate this risk. An experimental platform for crack detection in LMD samples was developed, and the identification of four simulated cracks in a 316L austenitic stainless-steel LMD sample was conducted. Data at temperatures higher than 150 °C were collected by an IMP, and the results indicated that crack depth is an important factor affecting the peak temperature. Based on this factor, the locations of cracks in LMD-316L austenitic stainless-steel samples can be determined. The proposed technique can provide real-time detection of cracks through layers of cladding during large-scale manufacturing, which suggests its relevance for optimizing the technological process and parameters, as well as reducing the possibility of cracks in the LMD process.

An Architectural Multi-Agent System for a Pavement Monitoring System with Pothole Recognition in UAV Images

In recent years, maintenance work on public transport routes has drastically decreased in many countries due to difficult economic situations. The various studies that have been conducted by groups of drivers and groups related to road safety concluded that accidents are increasing due to the poor conditions of road surfaces, even affecting the condition of vehicles through costly breakdowns. Currently, the processes of detecting any type of damage to a road are carried out manually or are based on the use of a road vehicle, which incurs a high labor cost. To solve this problem, many research centers are investigating image processing techniques to identify poor-condition road areas using deep learning algorithms. The main objective of this work is to design of a distributed platform that allows the detection of damage to transport routes using drones and to provide the results of the most important classifiers. A case study is presented using a multi-agent system based on PANGEA that coordinates the different parts of the architecture using techniques based on ubiquitous computing. The results obtained by means of the customization of the You Only Look Once (YOLO) v4 classifier are promising, reaching an accuracy of more than 95%. The images used have been published in a dataset for use by the scientific community.

Resonant frequency tracking mode on eddy current pulsed thermography non-destructive testing

Eddy current pulsed thermography (ECPT) has been widely used in the field of non-destructive testing due to its safety, non-contact detection, high spatial resolution and intuitive results. Inductive excitation source is an important component of ECPT and provides high-frequency alternating current to drive the excitation coil. However, a resonant frequency distortion phenomenon exists in the excitation source during the detection process, which seriously affects the output power of the excitation source and the sample detection effect. This paper presents a fast resonant frequency tracking loop for full bridge series resonant inverter which is used to search the resonance frequency in real time through direct digital synthesizer (DDS) and all-digital phase-locked loop. Theoretical analysis and simulation are presented to explain the working principle of the loop. Then, an experimental prototype is manufactured which serves as an excitation source for the ECPT experimental system. Compared with traditional excitation sources, the prototype does not need a water-cooled device and the tracking speed can be adjusted by modifying the parameters of DDS. Finally, experiments have been conducted on both artificial slot of 45# steel and natural cracks of rail and stainless steel to investigate the influence of resonant frequency tracking speed on the crack detection. The results revealed that reducing the resonant frequency tracking time can efficiently improve defect detectability and the manufactured prototype showed more application potential. This article is part of the theme issue 'Advanced electromagnetic non-destructive evaluation and smart monitoring'.

Detection and Quantification of Cracking in Concrete Aggregate through Virtual Data Fusion of X-ray Computed Tomography Images

In this work, which is part of a larger research program, a framework called "virtual data fusion" was developed to provide an automated and consistent crack detection method that allows for the cross-comparison of results from large quantities of X-ray computed tomography (CT) data. A partial implementation of this method in a custom program was developed for use in research focused on crack quantification in alkali-silica reaction (ASR)-sensitive concrete aggregates. During the CT image processing, a series of image analyses tailored for detecting specific, individual crack-like characteristics were completed. The results of these analyses were then "fused" in order to identify crack-like objects within the images with much higher accuracy than that yielded by any individual image analysis procedure. The results of this strategy demonstrated the success of the program in effectively identifying crack-like structures and quantifying characteristics, such as surface area and volume. The results demonstrated that the source of aggregate has a very significant impact on the amount of internal cracking, even when the mineralogical characteristics remain very similar. River gravels, for instance, were found to contain significantly higher levels of internal cracking than quarried stone aggregates of the same mineralogical type.

Vision Measurement of Tunnel Structures with Robust Modelling and Deep Learning Algorithms

The health monitoring of tunnel structures is vital to the safe operation of railway transportation systems. With the increasing mileage of tunnels, regular inspection and health monitoring are urgently demanded for the tunnel structures, especially for information regarding deformation and damage. However, traditional methods of tunnel inspection are time-consuming, expensive and highly dependent on human subjectivity. In this paper, an automatic tunnel monitoring method is investigated based on image data which is collected through the moving vision measurement unit consisting of camera array. Furthermore, geometric modelling and crack inspection algorithms are proposed where a robust three-dimensional tunnel model is reconstructed utilizing a B-spline method and crack identification is conducted by means of a Mask R-CNN network. The innovation of this investigation is that we combine the robust modelling which could be applied for the deformation analysis and the crack detection where a deep learning method is employed to recognize the tunnel cracks intelligently based on image sensors. In this study, experiments were conducted on a subway tunnel structure several kilometers long, and a robust three-dimensional model is generated and the cracks are identified automatically with the image data. The superiority of this proposal is that the comprehensive information of geometry deformation and crack damage can ensure the reliability and improve the accuracy of health monitoring.

CNN Training with Twenty Samples for Crack Detection via Data Augmentation

The excellent generalization ability of deep learning methods, e.g., convolutional neural networks (CNNs), depends on a large amount of training data, which is difficult to obtain in industrial practices. Data augmentation is regarded commonly as an effective strategy to address this problem. In this paper, we attempt to construct a crack detector based on CNN with twenty images via a two-stage data augmentation method. In detail, nine data augmentation methods are compared for crack detection in the model training, respectively. As a result, the rotation method outperforms these methods for augmentation, and by an in-depth exploration of the rotation method, the performance of the detector is further improved. Furthermore, data augmentation is also applied in the inference process to improve the recall of trained models. The identical object has more chances to be detected in the series of augmented images. This trick is essentially a performance–resource trade-off. For more improvement with limited resources, the greedy algorithm is adopted for searching a better combination of data augmentation. The results show that the crack detectors trained on the small dataset are significantly improved via the proposed two-stage data augmentation. Specifically, using 20 images for training, recall in detecting the cracks achieves 96% and Fext(0.8), which is a variant of F-score for crack detection, achieves 91.18%.

Deep Learning-Based Feature Silencing for Accurate Concrete Crack Detection

An autonomous concrete crack inspection system is necessary for preventing hazardous incidents arising from deteriorated concrete surfaces. In this paper, we present a concrete crack detection framework to aid the process of automated inspection. The proposed approach employs a deep convolutional neural network architecture for crack segmentation, while addressing the effect of gradient vanishing problem. A feature silencing module is incorporated in the proposed framework, capable of eliminating non-discriminative feature maps from the network to improve performance. Experimental results support the benefit of incorporating feature silencing within a convolutional neural network architecture for improving the network's robustness, sensitivity, and specificity. An added benefit of the proposed architecture is its ability to accommodate for the trade-off between specificity (positive class detection accuracy) and sensitivity (negative class detection accuracy) with respect to the target application. Furthermore, the proposed framework achieves a high precision rate and processing time than the state-of-the-art crack detection architectures.

Detection and Measurement of Matrix Discontinuities in UHPFRC by Means of Distributed Fiber Optics Sensing

Following the significant improvement in their properties during the last decade, Distributed Fiber Optics sensing (DFOs) techniques are nowadays implemented for industrial use in the context of Structural Health Monitoring (SHM). While these techniques have formed an undeniable asset for the health monitoring of concrete structures, their performance should be validated for novel structural materials including Ultra High Performance Fiber Reinforced Cementitious composites (UHPFRC). In this study, a full scale UHPFRC beam was instrumented with DFOs, Digital Image Correlation (DIC) and extensometers. The performances of these three measurement techniques in terms of strain measurement as well as crack detection and localization are compared. A method for the measurement of opening and closing of localized fictitious cracks in UHPFRC using the Optical Backscattering Reflectometry (OBR) technique is verified. Moreover, the use of correct combination of DFO sensors allows precise detection of microcracks as well as monitoring of fictitious cracks’ opening. The recommendations regarding use of various SHM methods for UHPFRC structures are given.

Data-Driven Structural Health Monitoring and Damage Detection through Deep Learning: State-of-the-Art Review

Data-driven methods in structural health monitoring (SHM) is gaining popularity due to recent technological advancements in sensors, as well as high-speed internet and cloud-based computation. Since the introduction of deep learning (DL) in civil engineering, particularly in SHM, this emerging and promising tool has attracted significant attention among researchers. The main goal of this paper is to review the latest publications in SHM using emerging DL-based methods and provide readers with an overall understanding of various SHM applications. After a brief introduction, an overview of various DL methods (e.g., deep neural networks, transfer learning, etc.) is presented. The procedure and application of vibration-based, vision-based monitoring, along with some of the recent technologies used for SHM, such as sensors, unmanned aerial vehicles (UAVs), etc. are discussed. The review concludes with prospects and potential limitations of DL-based methods in SHM applications.

Crack Detection Zones: Computation and Validation

During the development of aerospace structures, typically many fatigue tests are conducted. During these tests, much effort is put into inspections in order to detect the onset of failure before complete failure. Strain sensor data may be used to reduce inspection effort. For this, a sufficient number of sensors need to be positioned appropriately to collect the relevant data. In order to minimize cost and effort associated with sensor positioning, the method proposed here aims at minimizing the number of necessary strain sensors while positioning them such that fatigue-induced damage can still be detected before complete failure. A suitable detection criterion is established as the relative change of strain amplitudes under cyclic loading. Then, the space of all possible crack lengths is explored. The regions where the detection criterion is satisfied before complete failure occurs are assembled into so-called detection zones. One sensor in this zone is sufficient to detect criticality. The applicability of the approach is demonstrated on a representative airplane structure that resembles a lower wing section. The method shows that four fatigue critical spots can be monitored using only one strain sensor in a non-intuitive position. Furthermore, we discuss two different strain measures for crack detection. The results of this paper can be used for reliable structural health monitoring using a minimum number of sensors.

Automatic Pixel-Level Crack Detection on Dam Surface Using Deep Convolutional Network

Crack detection on dam surfaces is an important task for safe inspection of hydropower stations. More and more object detection methods based on deep learning are being applied to crack detection. However, most of the methods can only achieve the classification and rough location of cracks. Pixel-level crack detection can provide more intuitive and accurate detection results for dam health assessment. To realize pixel-level crack detection, a method of crack detection on dam surface (CDDS) using deep convolution network is proposed. First, we use an unmanned aerial vehicle (UAV) to collect dam surface images along a predetermined trajectory. Second, raw images are cropped. Then crack regions are manually labelled on cropped images to create the crack dataset, and the architecture of CDDS network is designed. Finally, the CDDS network is trained, validated and tested using the crack dataset. To validate the performance of the CDDS network, the predicted results are compared with ResNet152-based, SegNet, UNet and fully convolutional network (FCN). In terms of crack segmentation, the recall, precision, F-measure and IoU are 80.45%, 80.31%, 79.16%, and 66.76%. The results on test dataset show that the CDDS network has better performance for crack detection of dam surfaces.