检索结果-内蒙古大学图书馆

crackVision: Effective concrete crack detection With Deep Learning and Transfer Learning

IEEE ACCESS 2025年 13卷 29554-29576页

作者： Alkannad, Abdulrahman A. Al Smadi, Ahmad Yang, Shuyuan Al-Smadi, Mutasem K. Al-Makhlafi, Moeen Feng, Zhixi Yin, Zhenlong Xidian Univ Sch Artificial Intelligence Xian 710071 Shaanxi Peoples R China Zarqa Univ Dept Data Sci & Artificial Intelligence Zarqa 13100 Jordan Imam Abdulrahman Bin Faisal Univ Coll Appl Studies & Community Serv Dept Management Informat Syst Dammam 34212 Saudi Arabia Wuhan Coll Sch Informat Engn Wuhan 430212 Peoples R China

crack evaluation is critical for assessing the structural integrity of concrete, significantly impacting safety, functionality, and durability. In response to the growing demand for more advanced crack detection techniques, this paper presents crackVision, an advanced framework that leverages Transfer Learning (TL) to enhance traditional methods of concrete crack detection. The framework utilizes pre-trained Convolutional Neural Networks (CNNs), such as ResNet50, Xception, and InceptionV3, which are fine-tuned to address the unique challenges of concrete crack detection. Initially trained on extensive datasets, these models have been further enhanced to improve their accuracy and reliability. The effectiveness of these models is rigorously evaluated using two key datasets: METU and SDNET2018. Strategic resampling techniques were employed to address the data imbalance in the SDNET2018 dataset, thereby improving the effectiveness of crackVision for both binary and multi-class classification tasks. The findings reveal that the ResNet50 model achieves remarkable accuracy, reaching 99.95% on the METU dataset and consistently exceeding 97% on the balanced SDNET2018 dataset. These results demonstrate the effectiveness of the crackVision framework, particularly emphasizing the exceptional performance of the ResNet50 model through Transfer Learning in concrete crack detection.

关键词： concrete crack detection Transfer learning data imbalance classification civil engineering convolutional neural networks concrete crack detection transfer learning data imbalance classification civil engineering convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

crackVisionX: A Fine-Tuned Framework for Efficient Binary concrete crack detection

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2025年

作者： Alkannad, Abdulrahman A. Smadi, Ahmad A. L. AL-Makhlafi, Moeen Yang, Shuyuan Feng, Zhixi Xidian Univ Sch Artificial Intelligence Key Lab Intelligent Percept & Image Understanding Xian 710071 Peoples R China Zarqa Univ Dept Data Sci & Artificial Intelligence Zarqa 13110 Jordan Wuhan Coll Sch Informat Engn Wuhan 430212 Peoples R China

cracks are critical defects in concrete structures, traditionally identified through human inspection. However, computer vision techniques, especially convolutional neural networks (CNNs), offer promising solutions for automated detection. Driven by this trend, this study proposes crackVisionX, a state-of-the-art deep learning framework for classifying binary concrete cracks. crackVisionX lies in its integration of advanced CNN architectures, ResNet50, MobileNet_v3_large, DenseNet121, and EfficientNetB0, with extensive hyper-parameter tuning. This integration optimizes crack detection accuracy while maintaining low model complexity and reducing bias, making it suitable for real-time applications. Furthermore, the framework introduces a robust data augmentation strategy that effectively addresses dataset imbalances, enhancing model generalization across diverse domains. Additionally, crackVisionX employs comprehensive preprocessing on the METU and SDNET2018 datasets to create six domains: Bridge Deck, Wall, Pavement, SDNET2018, METU, and METU & SDNET2018. The framework's performance is thoroughly evaluated and benchmarked against state-of-the-art methods, utilizing diverse metrics to improve the detection of cracks in concrete structures. EfficientNetB0, a core component of the framework, demonstrated superior performance with exceptional test accuracies of up to 99.71%, 99.78%, 99.55%, 99.89%, 99.98%, and 99.92% for Bridge Deck, Wall, Pavement, SDNET2018, METU, and METU & SDNET2018, respectively. Moreover, we evaluated the robustness of crackVisionX using images contaminated with different types and intensities of noise, demonstrating its reliability and effectiveness. This balance between high accuracy and computational efficiency confirms the framework's potential for practical deployment. The experimental results emphasize the transformative potential of deep learning in construction safety and structural health monitoring.

关键词： Computational modeling Accuracy Deep learning Real-time systems Bridges Training Computational efficiency Convolutional neural networks Data models Inspection concrete crack detection binary classification deep learning crackVisionX fine-tuning structural health monitoring

来源：评论

学校读者我要写书评

暂无评论

SS-CCDN: A semi-supervised pixel-wise concrete crack detection network using multi-task learning and memory information

引用

MEASUREMENT 2025年 239卷

作者： Zhang, Xiaobo Tang, Haihao Yu, Chuanjin Zhai, Donghai Li, Yongle SouthWest JiaoTong Univ Sch Comp & Artificial Intelligence Chengdu 611756 Peoples R China SouthWest JiaoTong Univ Artificial Intelligence Res Inst Chengdu 611756 Peoples R China SouthWest JiaoTong Univ Natl Engn Lab Integrated Transportat Big Data Appl Chengdu 611756 Peoples R China Minist Educ Engn Res Ctr Sustainable Urban Intelligent Transpo Chengdu Peoples R China SouthWest JiaoTong Univ Sch Civil Engn Chengdu 611756 Peoples R China

concrete cracks pose significant challenges to infrastructure maintenance and safety. Traditional methods for detecting cracks suffer from inefficiency and subjectivity. Deep learning has shown promise recently, yet it often requires extensive labeled data. A semi-supervised concrete crack detection network (SS-CCDN) is proposed to handle the aforementioned challenge. To be specific, a multi-task model that takes edge feature detection of cracks as an auxiliary task is established. This model is then applied to both student and teacher networks to realize semi-supervised detection. Furthermore, the feature fusion networks incorporate the innovative multi-head cosine non-local attention module to achieve comprehensive acquisition and implementation of feature information at different scales. Lastly, the memory module is integrated into the network to compare the commonalities and distinctions between input sample information and memory sample data, successfully and effectively detecting concrete crack areas. SS-CCDN outperforms other baselines in detection, according to experimental results on two benchmark datasets.

关键词： concrete crack detection Mean teacher (MT) Semi-supervised learning Multi-task learning Pixel-wise detection

来源：评论

学校读者我要写书评

暂无评论

Optimizing concrete crack detection an echo state network approach with improved fish migration optimization

引用

SCIENTIFIC REPORTS 2025年第1期15卷 1-15页

作者： Fang, Zhichun Wang, Xiuhong Gao, Jiaojiao Eskandarpour, Behrooz Tongling Univ Inst Civil & Architectural Engn Tongling 244061 Anhui Peoples R China Hebei Vocat Coll Labour Relat Dept Architecture & Civil Engn Shijiazhuang 050093 Hebei Peoples R China Tongling Univ Sch Marxism Tongling 244061 Anhui Peoples R China Islamic Azad Univ Ilam Branch Ilam Iran Islamic Univ Coll Tech Engn Najaf Iraq

There are numerous reasons for concrete buildings cracks, like stress loads, material flaws, and environmental impacts. It is important to find and investigate the concrete cracks during analyzing the safety and structural soundness of buildings, bridges, and other infrastructure. However, there are many models available for concrete crack detection, an efficient approach is needed because the existing methods often have flaws like overfitting, high computational complexity, and noise sensitivity, which can lead to accurate crack detection and classification. This paper proposes an enhanced version of the fish migration optimization (IFMO) method combined with an optimized echo state network (ESN) model for concrete fracture detection using the combination form is established for improving the detection accuracy of the model by optimal arrangement of the ESN. The suggested ESN/IFMO model was tested on the SDNET2018 dataset, which comprises concrete photos with diverse fracture patterns, and the results were compared to several other state-of-the-art approaches. The suggested ESN/IFMO model shows potential as a more effective solution for concrete crack identification, increasing accuracy by 3.75-8.19% over current models such as DL, DINN, AlexNet, CNN, and LSTM, as well as increasing F1 score by 5.14-12.55%.

关键词： concrete crack detection Echo state networks Improved fish migration optimization Image classification Feature extraction SDNET2018 dataset

来源：评论

学校读者我要写书评

暂无评论

A performance-driven evaluation of deep learning for concrete crack detection with varying dataset sizes and training epochs: real-world implications for infrastructure monitoring

引用

Asian Journal of Civil Engineering 2025年 1-19页

作者： Bussa, Shashi Kumar Boppana, Narendra Kumar Deka, Bhupesh Department of Civil Engineering VNR VJIET Hyderabad India Department of CSE—AIML & IOT VNR VJIET Hyderabad India

Identification of concrete structural cracks at their early stages is essential for safety purposes because these early indicators lead to potential safety risks. A performance analysis of YOLOv8 object detection is provided for automatic concrete crack recognition through different training protocol assessments. The evaluation used actual concrete surface image sets that grew in size from 300 to 600 and finally 1500 samples with crack and non-crack content. YOLOv8n received training using each subset until 10, 30, 50, and 100 epochs for a total of twelve distinct model configurations. The evaluation metrics included Precision, Recall, F1-score, together with mAP@0.5 and mAP@0.5:0.95. Normalized confusion matrices, F1 and PR curves were used with validation predictions for analyzing detection patterns of each model. Model accuracy and generalization reach their peak performance as a result of the combined influence of the training duration and dataset size. A model training with 1500 images for 100 epochs delivered the optimal results, showing Precision 0.9700, Recall 0.9394, and mAP@0.5 of 0.9810 and mAP@0.5:0.95 of 0.8123. The research shows that extending training time past 50 epochs does not provide additional value in performance unless working with small-scale datasets. The research demonstrates that YOLOv8 functions as a viable and efficient tool for automation in crack detection tasks. It lays the groundwork for future enhancements such as severity classification, larger scale dataset integration, and edge deployment for real-time infrastructure monitoring. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： concrete crack detection Dataset size optimization Deep learning Object detection Structural health monitoring YOLOv8

来源：评论

学校读者我要写书评

暂无评论

Dataset for training neural networks in concrete crack detection: laboratory-classified beam and column images

引用

Data in Brief 2025年 61卷

作者： Alexandre Almeida Del Savio Ana Luna Torres Daniel Cárdenas-Salas Mónica Vergara Olivera Gianella Urday Ibarra Carrera de Ingeniería Civil Instituto de Investigación Científica Universidad de Lima Lima Perú Technology Innovation Program Carleton University Ottawa Canada

The construction industry is increasingly incorporating artificial intelligence into processes for the efficiency and accuracy of structural analysis and monitoring. However, obtaining high-quality datasets to train algorithms for detecting concrete cracks in structural components remains challenging, as such cracks normally develop over an extended period under real-world conditions. We introduce a curated dataset of 1,132 manually classified images of concrete cracks in beams and columns. These images were captured in a controlled laboratory environment using a static IP camera and annotated using the LabelImg tool. The dataset includes five object classes representing distinct cracks and failures in beams and columns and *** files containing classification and coordinates data. This dataset is designed to facilitate developing and validating of neural network-based computer vision models for automated crack detection. It is a very useful resource for researchers in structural engineering, which enables further developments in automated structural health monitoring and contributes to the overall use of AI in the construction industry.

关键词： concrete crack detection Neural networks Computer vision Structural health monitoring concrete beams concrete columns Image classification dataset AI in construction engineering

来源：评论

学校读者我要写书评

暂无评论

Lightweight concrete crack detection based on spiking neural networks

引用

ENGINEERING COMPUTATIONS 2024年第10期41卷 2534-2548页

作者： Ye, Wujian Huang, Hao Zhang, Boning Liu, Yijun Lin, Ziqi Guangdong Univ Technol Sch Integrated Circuits Guangzhou Peoples R China

Purpose - Most existing methods for concrete crack detection are based on deep learning techniques such as convolutional neural networks. However, these models, due to their large memory footprint, high power consumption and insufficient feature extraction capabilities, face challenges in mobile applications. To address these issues, this paper proposes a lightweight spiking neural network detection model. Design/methodology/approach - This model achieves fast and accurate crack detection. Firstly, the Gabor-Spiking (GS) module preprocesses input images, extracting texture features and edge features of crack images through Gabor filter convolution modules and spiking convolution modules, respectively. Next, the multiscale residual (MR) module is designed, composed of convolutional layers and residual modules of various scales, to process the fused features and perform crack detection. Findings - Experimental results demonstrate that the model's size can be reduced to 4.6 MB, achieving accuracy improvements to 87.3 and 96.4% on the SDNET and OCD datasets, respectively. Originality/value - This paper proposes a lightweight spiking neural network detection model based on the GS module for edge texture feature fusion and the MR module for crack detection.

关键词： concrete crack detection Spiking neural network Feature fusion Multiscale feature extraction Lightweight

来源：评论

学校读者我要写书评

暂无评论

Improving the concrete crack detection Process via a Hybrid Visual Transformer Algorithm

引用

SENSORS 2024年第10期24卷 3247页

作者： Shahin, Mohammad Chen, F. Frank Maghanaki, Mazdak Hosseinzadeh, Ali Zand, Neda Koodiani, Hamid Khodadadi Univ Texas San Antonio Mech Engn Dept San Antonio TX 78249 USA Univ Texas San Antonio Comp Sci Dept San Antonio TX 78249 USA Univ Texas San Antonio Civil & Environm Engn Dept San Antonio TX 78249 USA

Inspections of concrete bridges across the United States represent a significant commitment of resources, given their biannual mandate for many structures. With a notable number of aging bridges, there is an imperative need to enhance the efficiency of these inspections. This study harnessed the power of computer vision to streamline the inspection process. Our experiment examined the efficacy of a state-of-the-art Visual Transformer (ViT) model combined with distinct image enhancement detector algorithms. We benchmarked against a deep learning Convolutional Neural Network (CNN) model. These models were applied to over 20,000 high-quality images from the concrete Images for Classification dataset. Traditional crack detection methods often fall short due to their heavy reliance on time and resources. This research pioneers bridge inspection by integrating ViT with diverse image enhancement detectors, significantly improving concrete crack detection accuracy. Notably, a custom-built CNN achieves over 99% accuracy with substantially lower training time than ViT, making it an efficient solution for enhancing safety and resource conservation in infrastructure management. These advancements enhance safety by enabling reliable detection and timely maintenance, but they also align with Industry 4.0 objectives, automating manual inspections, reducing costs, and advancing technological integration in public infrastructure management.

关键词： maintenance inspection concrete crack detection big data waste reduction machine learning Industry 4.0 computer-based vision

来源：评论

学校读者我要写书评

暂无评论

Optimizing concrete crack detection: An Attention-Based SWIN U-Net Approach

引用

IEEE ACCESS 2024年 12卷 77575-77585页

作者： Sarhadi, Ali Ravanshadnia, Mehdi Monirabbasi, Armin Ghanbari, Milad Islamic Azad Univ Dept Civil Engn Sci & Res Branch Tehran *** Iran Payame Noor Univ Dept Civil Engn Tehran 193954697 Iran Islamic Azad Univ Dept Civil Engn East Tehran Branch Tehran *** Iran

Utilizing convolutional neural network (CNN) models, computer vision technology has become a reliable and powerful tool for detecting potential damage in concrete structures at the pixel level. In this study, an advanced SWIN U-Net architecture was introduced to detect concrete cracks. The model integrated attention-based convolutional neural networks to enhance the speed and accuracy of crack detection significantly. The distinctive features of the SWIN Transformer made the application of the model to images of varying sizes possible while the computational resources were used efficiently. To train the model, a dataset consisting of crack images, each accompanied by a corresponding mask that highlighted the relevant regions within the image, was used. The training data were augmented using Flip, Rotate, Random Contrast, Random Gamma, Random Brightness, Elastic Transformation, Grid Distortion, and Optical Distortion to counter potential overfitting. Additionally, L2 and spatial dropout regularization techniques were applied to the proposed model. The model was fine-tuned using stochastic gradient descent with the Adam optimizer, employing the binary cross-entropy loss function, and a learning rate of 0.001. The model was trained over 100 epochs with an adjustment scheduler. The performance of the model was evaluated using various metrics. Then, it was compared with three benchmarks and impressive results were achieved. Notably, the Dice loss and IOU values were 93% and 79%, respectively. The trained model had the exceptional performance score of 0.99 in accuracy, precision, recall, F1, and sensitivity.

关键词： Feature extraction Convolutional neural networks Image segmentation Computer architecture Computational modeling Training Surface cracks Transfer learning Computer vision Structural engineering Attention-based mechanism concrete crack detection generalizable model SWIN U-Net transfer learning

来源：评论

学校读者我要写书评

暂无评论

Visualization analysis of concrete crack detection in civil engineering infrastructure based on knowledge graph

引用

CASE STUDIES IN CONSTRUCTION MATERIALS 2024年 21卷

作者： Chen, Wei Hou, Jia Wang, Yanhua Yu, Mingyu Wuhan Univ Technol Sch Civil Engn & Architecture Wuhan 430070 Peoples R China Wuhan Univ Technol Sanya Sci & Educ Innovat Pk Sanya 572000 Peoples R China Cent & Southem China Municipal Engn Design & Res I Wuhan 430010 Peoples R China

Civil engineering infrastructure (such as buildings, roads, and underground tunnels) inevitably sustains varying degrees of damage over time, which can pose significant threats to both structural integrity and human safety. cracks are among the most common and concerning types of damage encountered. Therefore, timely and accurate inspection and evaluation of cracks are essential to ensure the safety and maintainability of these structures. Proactive crack detection and assessment enable early identification of potential issues, allowing for effective interventions that can prevent catastrophic failures and extend the lifespan of critical infrastructure. In recent years, researchers have been devoted to the exploration of concrete crack detection, which has produced a large number of papers. To better promote the development of the field of concrete crack detection, it is very important to review the literature in this field. Therefore, by using VOSviewer and CiteSpace text mining tools, this study analyzes and visualizes the scientometrics data extracted from the article from the perspective of bibliometrics. Based on 1592 papers published in 23 years, the paper analyzes the number of papers published, the sources of papers, the frequency of citations, the main authors, the research institutions, and the research hotspots of the concrete crack detection of Civil engineering infrastructure. The results of this study show that Song, GB, Aggelis, DG, Zhu, HP, et al. are the main contributors to the research of concrete crack detection. It can be seen from the types of issuing institutions that the research institutions engaged in concrete crack detection are mainly universities. The publishing institutions are predominantly located in Asia, reflecting the region's contributions to research output. The research focuses on the keywords "deep learning", "structural health monitoring", "acoustic emission" and "finite element". In the future, it is essential to strengthen c

关键词： concrete crack detection CiteSpace VOSviwer Visualization Literature review

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：