Search

Crack detection and measurement are essential for assessing the structural integrity of reinforced concrete (RC) structures, but challenges such as surface variability and class imbalance complicate accurate detection. This study introduces an approach integrating Convolutional Neural Networks (ConvNets), adaptive sliding windows, and DBSCAN-based semantic segmentation to address these challenges and enhance crack detection and quantification. The method was evaluated on various surface types, including painted masonry and concrete pavement, with a particular focus on overcoming class imbalance. To tackle this issue, the resampling (RS) technique was applied, achieving the best balance between precision and recall, with an F1 score of 0.836 during validation. The adaptive sliding window algorithm, optimized for lower magnification factors, further enhanced crack localization, improving IoU, recall, and precision. In semantic segmentation, the proposed method performed competitively on the DeepCrack dataset, achieving an IoU of 0.671, comparable to state-of-the-art models. Additionally, the measurement algorithm, which captures crack features such as length, width, and orientation, was tested on multiple surfaces. For painted masonry, it achieved a precision of 0.99, recall of 0.845, and IoU of 0.838, while for concrete pavement, it achieved a precision of 0.983, recall of 0.835, and IoU of 0.823. When applied to the DeepCrack dataset ground truth, it yielded a recall of 0.884, precision of 0.971, and IoU of 0.860. The results demonstrate the robustness and adaptability of this framework, offering an effective solution for automated crack detection and measurement across diverse surfaces. [ABSTRACT FROM AUTHOR]

The accurate detection of tunnel lining cracks and prompt identification of their primary causes are critical for maintaining tunnel availability. The advancement of deep learning, particularly in the domain of convolutional neural network (CNN) for image segmentation, has made tunnel lining crack detection more feasible. However, the CNN-based technique for tunnel lining crack detection commonly prioritizes increasing algorithmic complexity to enhance detection accuracy, posing a challenge in balancing the accuracy of detection and the efficiency of the algorithm. Motivated by the superior performance of Unet in image segmentation, this paper proposes a lightweight tunnel lining crack detection model named Mini-Unet, which refined the Unet architecture and utilized depthwise separable convolutions (DSConv) to replace some standard convolution layers. In the optimization of the proposed model parameters, applying a hybrid loss function that integrated dice loss and cross-entropy loss effectively tackled the imbalance between crack and background categories. Several models were set up to contrast with Mini-Unet and the experimental results were analyzed. Mini-Unet achieves a mean intersection over union (MIoU) of 60.76%, a mean precision of 84.18%, and a frame per second (FPS) of 5.635, respectively. Mini-Unet outperforms several mainstream models, enabling rapid detection while maintaining identified accuracy and facilitating the practical application of AI power for real-time tunnel lining crack detection. [ABSTRACT FROM AUTHOR]

To determine the extent of pavement damage and forms of pavement distress, road pavement conditions must be precisely assessed. As a result, monitoring systems are regarded as an important stage in the maintenance procedure. In recent times, numerous investigations have been carried out to track the condition of pavement and monitor road surfaces. In the undertaken study, we have proposed a novel artificial intelligent (AI) and computer vision-enabled PavementCarevision 4.0 approach to detect and classify pavement health conditions i.e., defects. In this study, a customized pavement-2000 dataset has been designed which contains more than 2,000 images of a variety of pavement defects. In the initial phase, we pre-processed and enhanced pavement images using the customized adjustable linear contrast enhancement methodology. The enhanced pavement image samples were fed to the proposed customized YOLOV8 enabled PavementHealth 4.0 framework for pavement condition detection of a variety of pavement defects such as longitudinal cracks, alligator cracks, transverse cracks, and potholes. The proposed customized YOLOV8 enabled PavementHealth 4.0 framework has achieved an accuracy of 99.20 percent; an receiver operating characteristic (ROC) value of 0.98 and outperformed existing AI-based state-of-the-art methodologies such as pose NET, YOLOv7, YOLOv5, long short-term memory network (LSTM), Mask region-based convolutional neural network (R-CNN), and decision tree. [ABSTRACT FROM AUTHOR]

The automation of bridge disease detection necessitates the time-consuming, labor-intensive manual detection process and the limitations of traditional image segmentation methods, such as inadequate denoising effects and insufficient continuity in crack segmentation. This paper proposes a rapid detection and information feedback approach based on an enhanced Convolutional Neural Network (CNN) model to tackle these issues in bridge crack width measurement and information processing. To improve efficiency and accuracy in bridge safety monitoring, the training data is constructed by the bridge image library and network crack through the refined preprocessing and image segmentation techniques applied to these images, key features of cracks are identified and extracted to enhance the capability for crack identification. For crack assessment, the maximum internal tangent circle method is employed to accurately measure the width of bridge abutment cracks. The effectiveness of our model was verified through both fixed-point detection and Unmanned Aerial Vehicle (UAV) dynamic detection, ensuring comprehensive and accurate data collection. This dual validation strategy shows that our model substantiates the wide applicability across various scenarios, and the non-contact crack measurement technique achieves a precision of 0.01 mm, demonstrating the effectiveness and accuracy of this streamlined CNN model in accurately assessing crack width.

This paper introduces a damage diagnosis method based on the reassignment method and matching networks (MNs) to study the structural health monitoring of aerospace composite material components. This aims to facilitate the mapping of signal features to complex failure modes. We introduce a signal processing technique based on the reassignment method, employing a sliding analysis window to re‐estimate local instantaneous frequency and group delay. By utilizing the short‐time phase spectrum of the signal, we correct the nominal time and frequency coordinates of the spectrum data, aligning them more accurately with the true support region of the analyzed signal. Subsequently, this paper developed a deep matching network (DMN) damage diagnosis model based on MNs. This model utilizes a convolutional neural network (CNN) to extract damage‐related features from the signal and introduces the full context embedding (FCE) method to enhance the compatibility of sample embeddings. In this process, the embeddings of each sample in the training set should be mutually independent, while the embeddings of test samples should be regulated by the distribution of training set sample data. Ultimately, the damage category of test samples is determined based on cosine similarity. We validate our model using damage sample data collected from experiments and simulations conducted under varying components and operating conditions. Comparative assessments with five mainstream methods reveal an average accuracy exceeding 96%. This underscores the exceptional recognition accuracy and generalization performance of our proposed method in cross‐operating condition fault diagnosis experiments concerning aircraft composite material components. [ABSTRACT FROM AUTHOR]

Road cracks are a common road traffic safety problem. Methods such as manual measurement of cracks do not facilitate large-scale inspections, which affects the normal operation of roads and the safety of pedestrians and vehicles. In this paper, a visual measurement technique based on convolutional neural networks and morphological operations is proposed for automated and efficient detection of road cracks. This method adopts the U-Net network as the infrastructure, changes the backbone feature extraction network to the VGG-16 network and introduces multiple Indicators to establish a new loss function to alleviate the sample imbalance problem. Finally, the crack features are combined to perform morphological operations on the image to enrich the recovered detail features. After experiments on the road crack dataset, this method has better crack segmentation capability and reliability compared to other algorithms. [ABSTRACT FROM AUTHOR]

Disaster-resilient dams require accurate crack detection, but machine learning methods cannot capture dam structural reaction temporal patterns and dependencies. This research uses deep learning, convolutional neural networks, and transfer learning to improve dam crack detection. Twelve deep-learning models are trained on 192 crack images. This research aims to provide up-to-date detecting techniques to solve dam crack problems. The finding shows that the EfficientNetB0 model performed better than others in classifying borehole concrete crack surface tiles and normal (undamaged) surface tiles with 91% accuracy. The study's pre-trained designs help to identify and to determine the specific locations of cracks. [ABSTRACT FROM AUTHOR]

Concrete 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. [ABSTRACT FROM AUTHOR]

Crack detection is significant to building repair and maintenance; however, conventional inspection is a labor-intensive and time-consuming process for field engineers. This paper proposes a three-step computer vision-based framework to quickly recognize concrete cracks and automatically identify their length, maximum width, and area in damage images. In step one, a region-based convolutional neural network (YOLOv8) is applied to train the crack localizing model. In step two, Gaussian filtering, Canny, and FindContours are integrated to extract the reference contour (a pre-designed seal) to obtain the conversion scale between pixels and millimeter-wise sizes. In step three, the recognized crack bounding box is cropped, and the ApproxPolyDP function and Hough transform are performed to quantify crack dimensions based on the conversion ratio. The developed framework was validated on a dataset of 4630 crack images, and the model training took 150 epochs. Results show that the average crack detection accuracy reaches 95.7%, and the precision of quantified dimensions is over 90%, while the error increases as the crack size grows smaller (increasing to 8% when the crack width is within 1 mm). The proposed method can help engineers to efficiently achieve crack information at building inspection sites, while the reference frame must be pre-marked near the crack, which may limit the scope of application scenarios. In addition, the robustness and accuracy of the developed image processing techniques-based crack quantification algorithm need to be further improved to meet the requirements in real cases when the crack is located within a complex background. [ABSTRACT FROM AUTHOR]

Abstract Technology offers a lot of potential that is being used to improve the integrity and efficiency of infrastructures. Crack is one of the major concerns that can affect the integrity or usability of any structure. Oftentimes, the use of manual inspection methods leads to delays which can worsen the situation. Automated crack detection has become very necessary for efficient management and inspection of critical infrastructures. Previous research in crack detection employed classification and localization-based models using Deep Convolutional Neural Networks (DCNNs). This study suggests and compares the effectiveness of transfer learned DCNNs for crack detection as a classification model and as a feature extractor to overcome this restriction. The main objective of this paper is to present various methods of crack detection on surfaces and compare their performance over 3 different datasets. Experiments conducted in this work are threefold: initially, the effectiveness of 12 transfer learned DCNN models for crack detection is analyzed on three publicly available datasets: SDNET, CCIC and BSD. With an accuracy of 53.40%, ResNet101 outperformed other models on the SDNET dataset. EfficientNetB0 was the most accurate (98.8%) model on the BSD dataset, and ResNet50 performed better with an accuracy of 99.8% on the CCIC dataset. Secondly, two image enhancement methods are employed to enhance the images and are transferred learned on the 12 DCNNs in pursuance of improving the performance of the SDNET dataset. The results from the experiments show that the enhanced images improved the accuracy of transfer-learned crack detection models significantly. Furthermore, deep features extracted from the last fully connected layer of the DCNNs are used to train the Support Vector Machine (SVM). The integration of deep features with SVM enhanced the detection accuracy across all the DCNN-dataset combinations, according to analysis in terms of accuracy, precision, recall, and F1-score.

Automatic crack detection of cement pavement chiefly benefits from the rapid development of deep learning, with convolutional neural networks (CNN) playing an important role in this field. However, as the performance of crack detection in cement pavement improves, the depth and width of the network structure are significantly increased, which necessitates more computing power and storage space. This limitation hampers the practical implementation of crack detection models on various platforms, particularly portable devices like small mobile devices. To solve these problems, we propose a dual-encoder-based network architecture that focuses on extracting more comprehensive fracture feature information and combines cross-fusion modules and coordinated attention mechanisms formore efficient feature fusion. Firstly, we use small channel convolution to construct shallow feature extractionmodule (SFEM) to extract low-level feature information of cracks in cement pavement images, in order to obtainmore information about cracks in the shallowfeatures of images. In addition,we construct large kernel atrous convolution (LKAC) to enhance crack information, which incorporates coordination attention mechanism for non-crack information filtering, and large kernel atrous convolution with different cores, using different receptive fields to extract more detailed edge and context information. Finally, the three-stage feature map outputs from the shallow feature extraction module is cross-fused with the two-stage feature map outputs from the large kernel atrous convolution module, and the shallow feature and detailed edge feature are fully fused to obtain the final crack prediction map. We evaluate our method on three public crack datasets: DeepCrack, CFD, and Crack500. Experimental results on theDeepCrack dataset demonstrate the effectiveness of our proposed method compared to state-of-the-art crack detection methods, which achieves Precision (P) 87.2%, Recall (R) 87.7%, and F-score (F1) 87.4%. Thanks to our lightweight crack detectionmodel, the parameter count of the model in real-world detection scenarios has been significantly reduced to less than 2M. This advancement also facilitates technical support for portable scene detection. [ABSTRACT FROM AUTHOR]

Technology offers a lot of potential that is being used to improve the integrity and efficiency of infrastructures. Crack is one of the major concerns that can affect the integrity or usability of any structure. Oftentimes, the use of manual inspection methods leads to delays which can worsen the situation. Automated crack detection has become very necessary for efficient management and inspection of critical infrastructures. Previous research in crack detection employed classification and localization-based models using Deep Convolutional Neural Networks (DCNNs). This study suggests and compares the effectiveness of transfer learned DCNNs for crack detection as a classification model and as a feature extractor to overcome this restriction. The main objective of this paper is to present various methods of crack detection on surfaces and compare their performance over 3 different datasets. Experiments conducted in this work are threefold: initially, the effectiveness of 12 transfer learned DCNN models for crack detection is analyzed on three publicly available datasets: SDNET, CCIC and BSD. With an accuracy of 53.40%, ResNet101 outperformed other models on the SDNET dataset. EfficientNetB0 was the most accurate (98.8%) model on the BSD dataset, and ResNet50 performed better with an accuracy of 99.8% on the CCIC dataset. Secondly, two image enhancement methods are employed to enhance the images and are transferred learned on the 12 DCNNs in pursuance of improving the performance of the SDNET dataset. The results from the experiments show that the enhanced images improved the accuracy of transfer-learned crack detection models significantly. Furthermore, deep features extracted from the last fully connected layer of the DCNNs are used to train the Support Vector Machine (SVM). The integration of deep features with SVM enhanced the detection accuracy across all the DCNN-dataset combinations, according to analysis in terms of accuracy, precision, recall, and F1-score. [ABSTRACT FROM AUTHOR]

As the backbone of transportation networks, the structural integrity of bridges is paramount for ensuring public safety and the efficient flow of goods and people. The traditional method that is Manual inspection methods for crack detection are labor intensive and often subjected to human error. This research explores an innovative approach to address this challenge by leveraging Convolutional Neural Networks for automated crack identification in bridge infrastructure. The proposed model employs a dataset encompassing concrete bridge conditions, and crack manifestations to train and evaluate the selected model. The accuracy of the prediction was verified by the test sets. The introduced model demonstrated a crack detection accuracy of 99% without relying on pre-training. Experimental results indicated that, when compared to existing classification models, the suggested model exhibited superior performance. Notably, the proposed model surpassed the ResNet50 model in terms of effectiveness. [ABSTRACT FROM AUTHOR]

Detecting cracks can be challenging, especially on rough surfaces such as masonry. This research paper focuses on the detection of surface cracks on masonry surfaces using deep-learning techniques. This study compared the performance of various networks trained using deep-learning techniques for semantic segmentation of cracks on masonry surfaces. For the semantic segmentation of cracks, the segmentation models U-Net, feature pyramid network (FPN), DeepLabV3+, and PSPNet were integrated with several convolutional neural networks (CNNs) acting as the network's backbone. Two loss functions, binary cross entropy and binary focal loss, were used in the study. Comparisons among networks using different metrics were performed to find the most promising approaches. Over the training and validation masonry data sets, a total of 23 networks were examined. The results of this study show that three networks can also accurately detect finer surface cracks on masonry surfaces. Based on performance metrics [dice coefficient, intersection over union (IoU), and F1 score], the three best networks were FPN(#2a) (86.9%, 74.9%, 59.3%), FPN(#2c) (85.6%, 75.4%, 56.3%), DeepLabV3+(#1a) (83.1%, 72.0%, 54.4%), respectively. Trained networks have demonstrated proficient performance on existing masonry culverts. This study can significantly aid the detection of cracks in the masonry substructure of old railway bridges. [ABSTRACT FROM AUTHOR]

Regular detection and repair for lining cracks are necessary to guarantee the safety and stability of tunnels. The development of computer vision has greatly promoted structural health monitoring. This study proposes a novel encoder–decoder structure, CrackRecNet, for semantic segmentation of lining segment cracks by integrating improved VGG-19 into the U-Net architecture. An image acquisition equipment is designed based on a camera, 3-dimensional printing (3DP) bracket and two laser rangefinders. A tunnel concrete structure crack (TCSC) image data set, containing images collected from a double-shield tunnel boring machines (TBM) tunnel in China, was established. Through data preprocessing operations, such as brightness adjustment, pixel resolution adjustment, flipping, splitting and annotation, 2880 image samples with pixel resolution of 448 × 448 were prepared. The model was implemented by Pytorch in PyCharm processed with 4 NVIDIA TITAN V GPUs. In the experiments, the proposed CrackRecNet showed better prediction performance than U-Net, TernausNet, and ResU-Net. This paper also discusses GPU parallel acceleration effect and the crack maximum width quantification. [ABSTRACT FROM AUTHOR]

This paper introduces a deep learning approach for accurately predicting the position and size of cracks in steam turbine blades using frequency response function data obtained from finite element analysis. Training deep neural networks for crack prediction necessitates an extensive dataset comprising various crack conditions, along with corresponding position and size information. However, obtaining sufficient high-quality and well-balanced experimental data for diverse crack conditions poses challenges. To address this issue, we leverage finite element analysis techniques to effectively amass a large volume of training data. Our methodology involves frequency response analysis, incorporating a crack modeling approach based on node sharing elimination and model order reduction (MOR). Automating this process enables us to efficiently generate a diverse dataset of crack blade frequency response data in a short timeframe, focusing on cracks in a single blade. The collected data is then transformed into two-dimensional image data, including magnitude similarity maps and magnitude frequency response maps, which serve as training data for our convolutional neural network (CNN) model. Evaluation of the CNN model's performance and accuracy for crack position and size prediction highlights its remarkable precision in determining the initiation position and size of blade cracks. Utilizing the proposed prediction model, there was a 95 % probability of accurately predicting the crack position within 3.1 % of the entire airfoil area. Furthermore, successful crack size predictions were achieved for approximately 96.4 % of the entire test dataset within an error range of ±1 mm. This underscores the potential capability of neural network training and crack prediction by leveraging training data derived from the finite element analysis and MOR. Our approach demonstrates promising implications for enhancing steam turbine blade maintenance and structural integrity assessment, as it streamlines data acquisition and provides an efficient solution for predicting critical crack characteristics. [ABSTRACT FROM AUTHOR]

Cracks on concrete surfaces are vital factors affecting construction safety. Accurate and efficient crack detection can prevent safety-related accidents. Using drones to photograph cracks on a concrete surface and detect them through computer vision technology has the advantages of accurate target recognition, simple practical operation, and low cost. To solve this problem, an improved CenterNet concrete crack-detection model is proposed. Firstly, a channel-space attention mechanism is added to the original model to enhance the ability of the convolution neural network to pay attention to the image. Secondly, a feature selection module is introduced to scale the feature map in the downsampling stage to a uniform size and combine it in the channel dimension. In the upsampling stage, the feature selection module adaptively selects the combined features and fuses them with the output features of the upsampling. Finally, the target size loss is optimized from a Smooth L1 Loss to IoU Loss to lessen its inability to adapt to targets of different sizes. The experimental results show that the improved CenterNet model reduces the FPS by 123.7 Hz, increases the GPU memory by 62 MB, increases the FLOPs by 3.81 times per second, and increases the AP by 15.4% compared with the original model. The GPU memory occupancy remained stable during the training process and exhibited good real-time performance and robustness. [ABSTRACT FROM AUTHOR]

Regular crack inspection plays a significant role in the maintenance of concrete structures. However, most deep-learning-based methods suffer from the heavy workload of pixel-level labeling and the poor performance of crack segmentation with the presence of background interferences. To address these problems, the Deformable Oriented YOLOv4 (DO-YOLOv4) is first developed for crack detection based on the traditional YOLOv4, in which crack features can be effectively extracted by deformable convolutional layers, and the crack regions can be tightly enclosed by a series of oriented bounding boxes. Then, the proposed DO-YOLOv4 is further utilized in combination with the image processing techniques (IPTs), leading to a novel hybrid approach, termed DO-YOLOv4-IPTs, for crack segmentation. The experimental results show that, owing to the high precision of DO-YOLOv4 for crack detection under background noise, the present hybrid approach DO-YOLOv4-IPTs outperforms the widely used Convolutional Neural Network (CNN)-based crack segmentation methods with less labeling work and superior segmentation accuracy. [ABSTRACT FROM AUTHOR]

Active learning is an important technology to solve the lack of data in crack detection model training. However, the sampling strategies of most existing active learning methods for crack detection are based on the uncertainty or representation of the samples, which cannot effectively balance the exploitation and exploration of active learning. To solve this problem, this study proposes an active learning method for crack detection based on subset searching and weighted sampling. First, a new active learning framework is established to successively search subsets with large uncertainty from the candidate dataset, and select training samples with large diversity from the subsets to update the crack detection model. Second, to realize the active learning process, a subset searching method based on sample relative error is proposed to adaptively select subsets with large uncertainty, and a weighted sampling method based on flow-based deep generative network is introduced to select training samples with large diversity form the subsets. Third, a termination criterion for active learning directly based on the prediction accuracy of the trained model is proposed to adaptively determine the maximum number of iterations. Finally, the proposed method is tested using two open-source crack datasets. The experimental comparison results on the Bridge Crack Library dataset show that the proposed method has higher calculation efficiency and prediction accuracy in crack detection than the uncertainty-based and representation-based active learning methods. The test results on the DeepCrack dataset show that the crack detection model trained by the proposed method has good transferability on different datasets with multi-scale concrete cracks and scenes. [ABSTRACT FROM AUTHOR]

Detection of cracks in solar photovoltaic (PV) modules is crucial for optimal performance and long-term reliability. The development of convolutional neural networks (CNNs) has significantly improved crack detection, offering improved accuracy and efficiency over traditional methods. This paper presents a comprehensive review and comparative analysis of CNN-based approaches for crack detection in solar PV modules. The review discusses various CNN architectures, including custom-designed networks and pre-trained models, as well as data-augmentation techniques and ensemble learning methods. Additionally, challenges related to limited dataset sizes, generalizability across different solar panels, interpretability of CNN models, and real-time detection are discussed. The review also identifies opportunities for future research, such as the need for larger and more diverse datasets, model interpretability, and optimized computational speed. Overall, this paper serves as a valuable resource for researchers and practitioners interested in using CNNs for crack detection in solar PV modules.

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

This study investigates image processing techniques for detecting surface cracks in spring steel components, with a focus on applications like Magnetic Particle Inspection (MPI) in industries such as railways and automotive. The research details a comprehensive methodology that covers data collection, software tools, and image processing methods. Various techniques, including Canny edge detection, Hough Transform, Gabor Filters, and Convolutional Neural Networks (CNNs), are evaluated for their effectiveness in crack detection. The study identifies the most successful methods, providing valuable insights into their performance. The paper also introduces a novel batch processing approach for efficient and automated crack detection across multiple images. The trade-offs between detection accuracy and processing speed are analyzed for the Morphological Top-hat filter and Canny edge filter methods. The Top-hat method, with thresholding after filtering, excelled in crack detection, with no false positives in tested images. The Canny edge filter, while efficient with adjusted parameters, needs further optimization for reducing false positives. In conclusion, the Top-hat method offers an efficient approach for crack detection during MPI. This research offers a foundation for developing advanced automated crack detection system, not only to spring sector but also extends to various industrial processes such as casting and forging tools and products, thereby widening the scope of applicability. [ABSTRACT FROM AUTHOR]

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

Detection of cracks in solar photovoltaic (PV) modules is crucial for optimal performance and long-term reliability. The development of convolutional neural networks (CNNs) has significantly improved crack detection, offering improved accuracy and efficiency over traditional methods. This paper presents a comprehensive review and comparative analysis of CNN-based approaches for crack detection in solar PV modules. The review discusses various CNN architectures, including custom-designed networks and pre-trained models, as well as data-augmentation techniques and ensemble learning methods. Additionally, challenges related to limited dataset sizes, generalizability across different solar panels, interpretability of CNN models, and real-time detection are discussed. The review also identifies opportunities for future research, such as the need for larger and more diverse datasets, model interpretability, and optimized computational speed. Overall, this paper serves as a valuable resource for researchers and practitioners interested in using CNNs for crack detection in solar PV modules. [ABSTRACT FROM AUTHOR]

Crack detection is an important task in bridge health monitoring, and related detection methods have gradually shifted from traditional manual methods to intelligent approaches with convolutional neural networks (CNNs) in recent years. Due to the opaque process of training and operating CNNs, if the learned features for identifying cracks in the network are not evaluated, it may lead to safety risks. In this study, to evaluate the recognition basis of different crack detection networks; several crack detection CNNs are trained using the same training conditions. Afterwards, several crack images are used to construct a dataset, which are used to interpret and analyze the trained networks and obtain the learned features for identifying cracks. Additionally, a crack identification performance criterion based on interpretability analysis is proposed. Finally, a training framework is introduced based on the issues reflected in the interpretability analysis. [ABSTRACT FROM AUTHOR]

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

The segmentation of crack detection and severity assessment in low-light environments presents a formidable challenge. To address this, we propose a novel dual encoder structure, denoted as DSD-Net, which integrates fast Fourier transform with a convolutional neural network. In this framework, we incorporate an information extraction module and an attention feature fusion module to effectively capture contextual global information and extract pertinent local features. Furthermore, we introduce a fractal dimension estimation method into the network, seamlessly integrated as an end-to-end task, augmenting the proficiency of professionals in detecting crack pathology within low-light settings. Subsequently, we curate a specialized dataset comprising instances of crack pathology in low-light conditions to facilitate the training and evaluation of the DSD-Net algorithm. Comparative experimentation attests to the commendable performance of DSD-Net in low-light environments, exhibiting superlative precision (88.5%), recall (85.3%), and F1 score (86.9%) in the detection task. Notably, DSD-Net exhibits a diminutive Model Size (35.3 MB) and elevated Frame Per Second (80.4 f/s), thereby endowing it with the potential to be seamlessly integrated into edge detection devices, thus amplifying its practical utility. [ABSTRACT FROM AUTHOR]

As a current popular method, intelligent detection of cracks is of great significance to road safety, so deep learning has gradually attracted attention in the field of crack image detection. The nonlinear structure, low contrast and discontinuity of cracks bring great challenges to existing crack detection methods based on deep learning. Therefore, an end-to-end deep convolutional neural network (AttentionCrack) is proposed for automatic crack detection to overcome the inaccuracy of boundary location between crack and non-crack pixels. The AttentionCrack network is built on U-Net based encoder-decoder architecture, and an attention mechanism is incorporated into the multi-scale convolutional feature to enhance the recognition of crack region. Additionally, a dilated convolution module is introduced in the encoder-decoder architecture to reduce the loss of crack detail due to the pooling operation in the encoder network. Furthermore, since up-sampling will lead to the loss of crack boundary information in the decoder network, a depthwise separable residual module is proposed to capture the boundary information of pavement crack. The AttentionCrack net on public pavement crack image datasets named CrackSegNet and Crack500 is trained and tested, the results demonstrate that the AttentionCrack achieves F1 score over 0.70 on the CrackSegNet and 0.71 on the Crack500 in average and outperforms the current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Crack detection plays a vital role in concrete surface maintenance. Deep-learning-based methods have achieved state-of-the-art results. However, these methods have some drawbacks. Firstly, a single-sized convolutional kernel in crack image segmentation tasks may result in feature information loss for small cracks. Secondly, only using linear interpolation or up-sampling to restore high-resolution features does not restore global information. Thirdly, these models are limited to learning edge features, causing edge feature information loss. Finally, various stains interfere with crack feature extraction. To solve these problems, a pyramid hierarchical convolution module (PHCM) is proposed by us to extract the features of cracks with different sizes. Furthermore, a mixed global attention module (MGAM) was used to fuse global feature information. Furthermore, an edge feature extractor module (EFEM) was designed by us to learn the edge features of cracks. In addition, a supplementary attention module (SAM) was used to resolv interference in stains in crack images. Finally, a pyramid hierarchical-convolution-based U-Net (PHCNet) with MGAM, EFEM, and SAM is proposed. The experimental results show that our PHCNet achieves accuracies of 0.929, 0.823, 0.989, and 0.801 on the Cracktree200, CRACK500, CFD, and OAD_CRACK datasets, respectively, which is higher than that of the traditional convolutional models. [ABSTRACT FROM AUTHOR]

Cracks in concrete cause structural damage, and it is important to identify and classify them. The objective of the research was to describe the behavior and predict the type of failure in concrete cylinders using convolutional neural networks. The methodology consisted of creating a database of 2650 images of failure types in concrete cylinders tested in compression at the Laboratory of Testing and Strength of Materials of the National University of Jaen, Cajamarca, Peru. To identify cracks on the concrete surface, the database was divided into training (60%), validation (20%), and testing (20%), and a transfer learning approach was developed using the MobileNet, DenseNet121, ResNet50, and VGG16 algorithms from the Keras library, programmed in Python. To validate the performance of each model, the following indicators were used: recall, precision, and F1 score. The results show that the models studied correctly classified the type of failure in concrete with accuracies of 96, 91, 86, and 90%, with the MobileNet algorithm being the best predictor with 96%. The novelty of the study was the development of deep learning algorithms with different architectures that can be used in structural health assessment as an automated and reliable method compared to traditional ones. In addition, these trained algorithms can be used as source code in drones for structural monitoring. [ABSTRACT FROM AUTHOR]

Crack detection is an important task for road maintenance. Currently, convolutional neural-network-based segmentation models with attention blocks have achieved promising results, for the reason that these models can avoid the interference of lights and shadows. However, by carefully examining the structure of these models, we found that these segmentation models usually use down-sampling operations to extract high-level features. This operation reduces the resolution of features and causes feature information loss. Thus, in our proposed method, a Parallel Convolution Module (PCM) was designed to avoid feature information loss caused by down-sampling. In addition, the attention blocks in these models only focused on selecting channel features or spatial features, without controlling feature information flow. To solve the problem, a Self-Gated Attention Block (SGAB) was used to control the feature information flow in the attention block. Therefore, based on the ideas above, a PSNet with a PCM and SGAB was proposed by us. Additionally, as there were few public datasets for detailed evaluation of our method, we collected a large dataset by ourselves, which we named the OAD_CRACK dataset. Compared with the state-of-the-art crack detection method, our proposed PSNet demonstrated competitive segmentation performance. The experimental results showed that our PSNet could achieve accuracies of 92.6%, 81.2%, 98.5%, and 76.2% against the Cracktree200, CRACK500, CFD, and OAD_CRACK datasets, respectively, which were 2.6%, 4.2%, 1.2%, and 3.3% higher than those of the traditional attention models. [ABSTRACT FROM AUTHOR]

Crack detection is an important component of dam safety monitoring. Detection methods based on deep convolutional neural networks (DCNNs) are widely used for their high efficiency and safety. Most existing DCNNs with high accuracy are too complex for users to deploy for real-time detection. However, compressing models face the dilemma of sacrificing detection accuracy. To solve this dilemma, an improved residual neural network (ResNet)-based algorithm for concrete dam crack detection using dynamic knowledge distillation is proposed in this paper in order to obtain higher accuracy for small models. To see how well distillation works, preliminary experiments were carried out on mini-ImageNet. ResNet18 was trained by adding additional tasks to match soft targets generated by ResNet50 under dynamic high temperatures. Furthermore, these pre-trained teacher and student models were transferred to experiments on concrete crack detection. The results showed that the accuracy of the improved algorithm was up to 99.85%, an increase of 4.92%. [ABSTRACT FROM AUTHOR]

Cracks are a major sign of aging transportation infrastructure. The detection and repair of cracks is the key to ensuring the overall safety of the transportation infrastructure. In recent years, due to the remarkable success of deep learning (DL) in the field of crack detection, many researches have been devoted to developing pixel-level crack image segmentation (CIS) models based on DL to improve crack detection accuracy, but as far as we know there is no review of DL-based CIS methods yet. To address this gap, we present a comprehensive thematic survey of DL-based CIS techniques. Our review offers several contributions to the CIS area. First, more than 40 papers of journal or top conference most published in the last three years are identified and collected based on the systematic literature review method. Second, according to the backbone network architecture of the models proposed in them, they are grouped into 10 topics: FCN, U-Net, encoder-decoder model, multi-scale, attention mechanism, transformer, two-stage detection, multi-modal fusion, unsupervised learning and weakly supervised learning, to be reviewed. Meanwhile, our survey focuses on discussing strengths and limitations of the models in each topic so as to reveal the latest research progress in the CIS field. Third, publicly accessible data sets, evaluation metrics, and loss functions that can be used for pixel-level crack detection are systematically introduced and summarized to facilitate researchers to select suitable components according to their own research tasks. Finally, we discuss six common problems and existing solutions to them in the field of DL-based CIS, and then suggest eight possible future research directions in this field.

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. [ABSTRACT FROM AUTHOR]

The visual test for cracks is a tough work for human eyes. With the help of cameras, methods for object detection based on deep learning can detect the cracks automatically from digital images. However, an anomaly is observed that the models for crack detection are obviously underestimated by the widely used mean Average Precision (mAP) standard. In this study, by theoretical analyses about the topology of cracks, we attribute the failure of mAP standard to the random fractal feature. More deeply, this issue is due to the strict box matching during the calculation. To solve this problem, we construct a practical fractal-available evaluation method adopting the idea of covering box matching. Several metrics including Extended Recall (XR), Extended Precision (XP), and Extended F -score (F ext) are defined for scoring the crack detectors. Then, we conduct experiments on several mainstream models for object detection. Results show two advantages of the proposed evaluation method: (1) the proposed metrics evaluate cases better; (2) the proposed evaluation is non-maximum suppression-unrelated. Moreover, we present a case study to optimize a faster region-based convolutional neural networks model for crack detection adopting the proposed metrics. Recall (represented by the XR) of our best model achieves an industrial-level at 95.8%. These results denote that the proposed evaluation method removes the obstacle in evaluating the methods for object detection for cracks, which show great potential for automatic industrial inspections. [ABSTRACT FROM AUTHOR]

The timely detection and recognizing of distress on an airport pavement is crucial for safe air traffic. For this purpose, a physical inspection of the airport maneuvering areas is regularly carried out, which might be time-consuming due to its size. One of the modern approaches to speeding up this process is unmanned aerial vehicle imagery followed by an automatic evaluation. This study explores the automatic detection of the transverse crack, its dimension measurement, and position determination within the slab on the concrete runway. The aerial image data were obtained from flights at the given altitude above the runway and processed using commercial multi-view reconstruction software to create a dataset for the training, verification, and testing of a YOLOv2 object detector. Once the crack was detected, the main features were obtained by image segmentation and morphological operations. The YOLOv2 detector was tuned with 3279 images until the detection metrics (average precision AP = 0.89) reached sufficient value for real deployment. The detected cracks were further processed to determine their position within the concrete slab, and their dimensions, i.e., length and width, were measured. The automated crack detection and evaluation system developed in this study was successfully verified on the experimental section of the runway as an example of practical application. It was proven that unmanned aerial vehicle imagery is efficient over broad areas and produces impressive results with the combination of artificial intelligence. [ABSTRACT FROM AUTHOR]

The study presents two crack detection techniques for a beam undergoing mode I fracture: (i) convolutional neural network approach and (ii) image processing using opensource computer vision library OpenCV and pixel count of cracks approach. The second method also includes crack segmentation and masking for visualization of the cracks. The study attempted to evaluate the accuracy of both methods at different crack mouth opening displacement of seven simply supported concrete beams being tested using three-point bend test. The novelty of the present work lies in quantification of the crack growth from captured images and evaluating the potential of applying computer vision techniques as a replacement for sensitive crack measuring instruments to create a robust computer vision based health monitoring system. Results suggest that both methods identified the cracks in the beam and are capable of generating a warning before the collapse. [ABSTRACT FROM AUTHOR]

Nowadays, acoustic and vibration signals have been widely used to detect crack and its degree on compressor blades for non-destructive evaluation and structural health monitoring. Due to complex working conditions and interference of strong noise, single signal-based methods reach unsatisfactory accuracy. To improve the reliability and accuracy of crack detection, the two-level fusion method based on Mahalanobis distance (MD) and self-attention mechanism is proposed based on multi-acoustic, vibration, and acoustic emission information. First, the MD is calculated to measure the similarity between raw samples and group, which can fuse the same type of samples in data level. Then, the processed raw and data-level fusion samples are inputted to one-dimensional convolutional neural network, and the preliminary decisions are obtained. Finally, the decision-level fusion based on self-attention mechanism and credit assignment is proposed to modify and correct the preliminary decisions. The compressor experiments are implemented to test the proposed method under single and mixed working conditions with strong noise intensity. Further, the results illustrate that the proposed method can accurately detect crack for compressor blades. By comparing with other data fusion approaches, the advantage of the proposed method is validated under complex conditions. [ABSTRACT FROM AUTHOR]

Abstract Manual investigation of damages incurred to infrastructure is a challenging process, in that it is not only labour-intensive and expensive but also inefficient and error-prone. To automate the process, a method that is based on computer vision for automatically detecting cracks from 2D images is a viable option. Amongst the different methods of deep learning that are commonly used, the convolutional neural network (CNNs) is one that provides the opportunity for end-to-end mapping/learning of image features instead of using the manual suboptimal image feature extraction. Specifically, CNNs do not require human supervision and are more suitable to be used for indoor and outdoor applications requiring image feature extraction and are less influenced by internal and external noise. Additionally, the CNN’s are also computationally efficient since they are based on special convolution layers and pooling operations that enable the full execution of CNN frameworks on several hardware devices. Keeping this in mind, we propose a deep CNN framework that is based on 10 different convolution layers along with a cycle GAN (Generative Adversarial Network) for predicting the crack segmentation pixel by pixel in an end-to-end manner. The methods proposed here include the Deeply Supervised Nets (DSN) and Fully Convolutional Networks (FCN). The use of DSN enables integrated feature supervision for each stage of convolution. Furthermore, the model has been designed intricately for learning and aggregating multi-level and multiscale features while moving from the lower to higher convolutional layers through training. Hence, the architecture in use here is unique from the ones in practice which just use the final convolution layer. In addition, to further refine the predicted results, we have used a guided filter and CRFs (Conditional Random Fields) based methods. The verification step for the proposed framework was carried out with a set of 537 images. The deep hierarchical CNN framework of 10 convolutional layers and the Guided filtering achieved high-tech and advanced performance on the acquired dataset, showing higher F-score, Recall and Precision values of 0.870, 0.861, and 0.881 respectively, as compared to the traditional methods such as SegNet, Crack-BN, and Crack-GF.

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

Copyright of Revista Iberoamericana de Ingeniería Mecánica is the property of Editorial UNED and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The tire sidewall is the weakest part of the entire tire. Although the tire sidewall is not directly in contact with the ground, it often undergoes great deformation. Weather, road conditions, and driving habits can also affect the tire life. Cracking is one of the earliest signs of tire aging and deterioration. If a driver does not regularly inspect their vehicle, damage to a tire may remain undetected and an uncontrolled tire explosion may occur. In this study, we use deeplearning-based artificial intelligence computer vision to train a deep neural network model using a large number of digital images to detect tire sidewall cracks instead of traditional sensors, inspection devices, or visual inspection methods. In this study, tire sidewall crack images were preprocessed and annotated using the annotation program VGG Image Annotator (VIA). Residual network 50 (ResNet50) is used as the backbone of mask-region-based convolutional neural networks (Mask R-CNNs). The preprocessing training and test results of our dataset show that the improved Mask R-CNN has better mean accuracy (mAP) and detection accuracy than the original Mask R-CNN and Faster-R-CNN and can not only reduce inspection costs and time, but also improve the efficiency of tire crack analysis. [ABSTRACT FROM AUTHOR]

Concrete crack detection is a significant research problem in structural safety. However, the traditional manual inspection is a laborious and time-consuming method, and the detection accuracy is greatly limited by the work experience of engineers. Hence, automatic image-based crack detection has attracted wide attention from both academia and industry. In this study, a novel crack detection method using attention convolution neural networks, ATCrack, is proposed for automatic crack identification. ATCrack uses a symmetric structure consisting of an encoder and a decoder by imposing channel-spatial attention to achieve end-to-end crack prediction. Channel attention module is introduced in the encoder to improve the effective utilization of crack features, and spatial attention is added in the decoder to suppress the background features. Combining with channel and spatial attention modules, the codec network will be more sensitive to the characteristics of cracks and increase detection accuracy and robustness. Moreover, a complex crack dataset of buildings and pavements is collected to verify the effectiveness and feasibility of ATCrack. Finally, experiment results are tested on several public datasets and self-collected (CBCrack) database, and it shows that the proposed method during the five-fold cross-validation can achieve state-of-the-art performance compared with other existing methods in terms of precision, recall, F1-score, and mIoU. [ABSTRACT FROM AUTHOR]

Crack detection is a vital component of structural health monitoring. Several computer vision-based studies have been proposed to conduct crack detection on concrete surfaces, but most cases have difficulties in detecting fine cracks. This study proposes a deep learning-based model for automatic crack detection on the concrete surface. Our proposed model is an encoder–decoder model which uses EfficientNet-B7 as the encoder and U-Net's modified expansion path as the decoder. To overcome the challenges in the detection of fine cracks, we trained our model with a new training strategy on images extracted from an open-access dataset and achieved a 96.98% F1 score for unseen test data. Moreover, we evaluated our method on CrackForest Dataset and achieved a 97.06% F1 score which outperforms all the existing methods. The robustness of the proposed model is investigated using the various numbers of training data, and the optimal data size for training this model is presented. The results show that although deep learning models acquire a large number of data, this model works with limited data, without any degradation in its performance. Furthermore, the novel training strategy used in this study, significantly improves the model's accuracy in detecting different types of cracks. [ABSTRACT FROM AUTHOR]

Two optimization methods are proposed to improve faster region-based convolutional neural network (Faster R-CNN), which are (1) restructuring Faster R-CNN's backbone network and the classification and regression (C&R) network using residual networks and (2) designing the feature ensemble structure for Faster R-CNN to combine the shallow with deep feature maps of the backbone. In addition, this paper proposed a method to evaluate the model's performance, which is pixel mean value (Pmean) distribution of different channel feature maps, and quantitatively evaluate the feature representation capability of the model. Experimental results show that mean average precision (mAP) of the model optimized by the first method can reach 86.5%, which is 1.9% higher than that of baseline. However, mAP of the model optimized by the second method reaches 87.5%, which is 2.9% higher than the baseline model. The Pmean statistics of each channel feature map extracted by different backbones show that the model accuracy is higher when the Pmean of its channel feature maps is bigger, which can effectively improve the interpretability of the model accuracy. [ABSTRACT FROM AUTHOR]

The creation and training of artificial neural networks with a given accuracy makes it possible to identify patterns and hidden relationships between physical and technological parameters in the production of unique building materials, predict mechanical properties, and solve the problem of detecting, classifying, and segmenting existing defects. The detection of defects of various kinds on elements of building materials at the primary stages of production can improve the quality of construction and identify the cause of particular damage. The technology for detecting cracks in building material samples is of great importance in building monitoring, in pre-venting the spread of defective material. In this paper, we consider the use of the YOLOv4 convolutional neural network for crack detection on building material samples. This was based on the creation of its own empirical database of images of samples of aerated concrete. The number of images was increased by applying our own augmentation algorithm. Optimization of the parameters of the intellectual model based on the YOLOv4 convolutional neural network was performed. Experimental results show that the YOLOv4 model developed in this article has high precision in defect detection problems: AP@50 = 85% and AP@75 = 68%. It should be noted that the model was trained on its own set of data obtained by simulating various shooting conditions, rotation angles, object deformations, and light distortions through image processing methods, which made it possible to apply the developed algorithm in practice. [ABSTRACT FROM AUTHOR]