Article
Construction & Building Technology
Shuo Ding, Hong Lang, Jiang Chen, Ye Yuan, Jian Lu
Summary: This study proposes a novel pixel-level pavement crack segmentation network, PCSNet, to address the challenge of crack detection in intelligent pavement surveys. The network integrates richer attention and hybrid pyramid structures, enhancing crack feature representation and scene parsing. By capturing crack spatial dependence information and integrating contextual information at multiple receptive field scales, the network achieves superior performance with F1-score, mean intersection of union, and mean pixel accuracy of 81.21%, 77.13%, and 87.17%, respectively. The method reconstructs complete crack geometry, preserves crack edges well, and optimizes the detection of shallow and complex cracks.
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
(2023)
Article
Engineering, Multidisciplinary
Youzhi Tang, Allen A. Zhang, Lei Luo, Guolong Wang, Enhui Yang
Summary: This paper introduces an encoder-decoder network (EDNet) for crack segmentation to address the quantity imbalance issue between crack and non-crack pixels. Experimental results demonstrate that EDNet outperforms other state-of-the-art models in crack detection.
Article
Construction & Building Technology
Anzheng He, Zishuo Dong, Hang Zhang, Allen A. A. Zhang, Shi Qiu, Yang Liu, Kelvin C. P. Wang, Zhihao Lin
Summary: This paper proposes an improved HRNet-OCR model, named EJSNet, for automated pixel-level detection of expansion joints on asphalt pavement. EJSNet modifies the residual structure of the first stage by conducting a Conv. + BN + ReLU operation for each shortcut connection to avoid network degradation. It incorporates the feature selection module (FSM), receptive field block (RFB) module, convolutional block attention module (CBAM), and modifies the shared multilayer perceptron (MLP) architecture of the channel attention module (CAM) for enhanced feature extraction and refinement. Experimental results show that EJSNet outperforms four state-of-the-art models for semantic segmentation in terms of detection accuracy on both private and public datasets.
STRUCTURAL CONTROL & HEALTH MONITORING
(2023)
Article
Computer Science, Interdisciplinary Applications
Allen A. Zhang, Kelvin C. P. Wang, Yang Liu, You Zhan, Guangwei Yang, Guolong Wang, Enhui Yang, Hang Zhang, Zishuo Dong, Anzheng He, Jie Xu, Jing Shang
Summary: This paper proposes a deep-learning model named ShuttleNet for simultaneous pixel-level detection of multiple distresses and surface design features on complex asphalt pavements. Experimental results demonstrate that the proposed ShuttleNet model outperforms four state-of-the-art models in terms of detection accuracy.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2022)
Article
Engineering, Multidisciplinary
Bei Zhang, Haoyuan Cheng, Yanhui Zhong, Xianghua Tao, Guanghui Li, Shengjie Xu
Summary: This paper proposes a method based on feature pixel points to quantify and calculate the vertical height of concealed cracks in asphalt pavements. By conducting numerical simulations, the characteristics of ground-penetrating radar (GPR) images of concealed cracks in asphalt pavement with varying lengths and widths were studied. The relationship between the pixel value of the crack area and the two-way travel time was established to obtain the relationship between the vertical height of the crack and the pixel. This method achieved a minimum error of only 2.9% in estimating the vertical height of cracks and can be applied in practical engineering applications.
Article
Construction & Building Technology
Tian Wen, Shuo Ding, Hong Lang, Jian John Lu, Ye Yuan, Yichuan Peng, Jiang Chen, Aidi Wang
Summary: This paper proposes an efficient deep learning framework called PDSNet for automated asphalt pavement distress segmentation. By utilizing both global and local features, PDSNet is able to produce precise segmentation results under complex circumstances. The experimental results show that PDSNet achieves the best MIoU compared to other state-of-the-art networks while maintaining a reasonable parameter count and inference time.
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
(2023)
Article
Construction & Building Technology
Jinchao Guan, Xu Yang, Ling Ding, Xiaoyun Cheng, Vincent C. S. Lee, Can Jin
Summary: An automated pixel-level pavement distress detection framework integrating stereo vision and deep learning is developed in this study, which establishes multi-feature pavement image datasets based on a multi-view stereo imaging system and proposes a modified U-net deep learning architecture introducing depthwise separable convolution for efficient crack and pothole segmentation. The results show that the 3D pavement image achieves millimeter-level accuracy, and the enhanced 3D crack segmentation model outperforms other models in terms of accuracy and speed, enabling high-precision automated pothole volume measurement.
AUTOMATION IN CONSTRUCTION
(2021)
Article
Engineering, Multidisciplinary
Aravinda S. Rao, Tuan Nguyen, Son T. Le, Marimuthu Palaniswami, Tuan Ngo
Summary: This study proposes a deep learning framework for detecting and measuring cracks in concrete structures, which outperforms traditional methods in terms of efficiency and accuracy. The framework can handle images of different sizes and predict crack widths, making it a powerful tool for structural health monitoring.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2022)
Article
Construction & Building Technology
Krisada Chaiyasarn, Apichat Buatik, Hisham Mohamad, Mingliang Zhou, Sirisilp Kongsilp, Nakhorn Poovarodom
Summary: This paper proposes an advanced inspection reporting system based on an integrated CNN-FCN crack detection system, which enables crack inspection and display for larger structures. The system utilizes a trained CNN to detect crack patches and a trained FCN system to segment cracks at the pixel-level. The system shows promising results with high accuracy and precision.
AUTOMATION IN CONSTRUCTION
(2022)
Article
Chemistry, Analytical
Gui Yu, Juming Dong, Yihang Wang, Xinglin Zhou
Summary: In this paper, we proposed a U-shaped encoder-decoder semantic segmentation network, called RUC-Net, combining Unet and Resnet for pixel-level pavement crack image segmentation, to address the challenges of complex backgrounds, uneven illumination, irregular patterns, and various types of noise interference in crack detection. We introduced the spatial-channel squeeze and excitation (scSE) attention module and the focal loss function to improve the detection effect and handle the class imbalance problem. Evaluation on three public datasets, CFD, Crack500, and DeepCrack, showed that our methods achieved better results compared to FCN, Unet, and SegNet. Additionally, we conducted ablation studies on the CFD dataset, comparing the differences of various scSE modules and their combinations in improving crack detection performance.
Article
Chemistry, Analytical
Youheng Guo, Xuesong Shen, James Linke, Zihao Wang, Khalegh Barati
Summary: Aging infrastructure is a global concern due to its potential economic and social destruction if it collapses. This paper proposes an efficient approach that utilizes 3D point cloud reconstruction and deep learning technology to accurately detect and quantify minor defects on complicated infrastructures.
Article
Construction & Building Technology
Aohui Ouyang, Vanessa Di Murro, Martin Cull, Roddy Cunningham, John Andrew Osborne, Zili Li
Summary: This study presents a remote and automated system for crack monitoring in concrete tunnel linings using robot-mounted imaging technology. The system collects crack images remotely and stitches them together to create a panorama image of the tunnel surface. Transfer learning is employed to optimize the state-of-the-art semantic segmentation model for automatic crack detection. Field trials conducted in tunnels at CERN demonstrate the feasibility and effectiveness of the proposed crack monitoring system.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2023)
Article
Engineering, Civil
Bin Yu, Xiangcheng Meng, Qiannan Yu
Summary: A two-step convolutional neural network method is proposed for pavement crack detection on pixel-levels, reducing time consumption while maintaining accuracy. Experimental results demonstrate that this method significantly decreases processing time with minimal loss of accuracy compared to pure-segmentation networks.
JOURNAL OF TRANSPORTATION ENGINEERING PART B-PAVEMENTS
(2021)
Article
Construction & Building Technology
Yuchuan Du, Shan Zhong, Hongyuan Fang, Niannian Wang, Chenglong Liu, Difei Wu, Yan Sun, Mang Xiang
Summary: This study proposes a lightweight pavement crack-detection model that combines object detection and semantic segmentation tasks. It utilizes a modified YOLOv4-Tiny model to predict crack bounding boxes and proposes a segmentation threshold. Additionally, an attention feature pyramid network and a denoising autoencoder network are introduced to compensate for accuracy loss and remove background noise. The proposed model achieves equivalent evaluation index values with significantly fewer parameters than conventional models.
AUTOMATION IN CONSTRUCTION
(2023)
Article
Construction & Building Technology
Xu Xia, Dongdong Han, Yongli Zhao, Yichang Xie, Ziyue Zhou, Jinming Wang
Summary: Cracking is a common issue in expressway pavements in Shanxi Province, with top-down cracks and reflection cracks observed as the primary types. This study simulates the initiation and propagation processes of these cracks using the finite element method, aiming to distinguish between the two. The findings show that top-down cracks exhibit multi-point cracking on the pavement surface, while reflection cracks display significant stress concentration at the crack tip, following a single propagation path.
CASE STUDIES IN CONSTRUCTION MATERIALS
(2023)
Article
Construction & Building Technology
Enhui Yang, Jiaqiu Xu, Haoyuan Luo, Bing Huang, Kelvin C. P. Wang, Joshua Qiang Li, Yanjun Qiu
Summary: This study evaluates the performance of warm mix asphalt (WMA) with different additives in high-altitude plateau regions. Solid additives are suitable for improving the high-temperature performance in high-temperature areas, while surface-active liquid additives are suitable for low-temperature plateau areas.
ROAD MATERIALS AND PAVEMENT DESIGN
(2022)
Article
Engineering, Civil
Guangwei Yang, Kelvin Wang, Joshua Qiang Li, Matt Romero, Wenyao Liu
Summary: The study demonstrates that Warm Mix Asphalt (WMA) incorporating Reclaimed Asphalt Pavement (RAP) and Reclaimed Asphalt Shingles (RAS) outperforms traditional Hot Mix Asphalt (HMA) in pavement construction, providing sustainable benefits.
KSCE JOURNAL OF CIVIL ENGINEERING
(2022)
Article
Materials Science, Composites
Chuanqi Yan, Quan Lv, Allen A. Zhang, Changfa Ai, Weidong Huang, Dongya Ren
Summary: The study introduces a kinetic model to describe the relationship between PMB modulus and temperature, proposes a modified model for better fitting results, and uses nonlinear least squares regression to determine kinetic parameters. The method could be a promising approach to study the temperature-dependent properties and state transition behaviors of PMB composite.
COMPOSITES SCIENCE AND TECHNOLOGY
(2022)
Article
Engineering, Multidisciplinary
You Zhan, Cheng Liu, Qiangsheng Deng, Qi Feng, Yanjun Qiu, Allen Zhang, Xianlin He
Summary: This study developed an integrated FFT and XGBoost framework to predict pavement skid resistance with high accuracy, with Skewness, Mean Profile Depth, and Power Spectral Density being identified as key parameters for characterizing pavement micro-texture. The XGBoost model outperformed other models with an R-2 value of 0.88.
Article
Transportation Science & Technology
Xue Yang, Joshua Q. Li, Wenyao Liu, Kelvin C. P. Wang, Jim Hatt, Jared Schwennesen
Summary: This study developed a data-driven hierarchical evaluation process for identifying and prioritizing potential at-grade crossings as candidates for grade separations. The results provide comprehensive and practical analysis for the selection and prioritization of grade separation projects.
INTERNATIONAL JOURNAL OF RAIL TRANSPORTATION
(2023)
Article
Engineering, Multidisciplinary
Jing Shang, Jie Xu, Allen A. Zhang, Yang Liu, Kelvin C. P. Wang, Dongya Ren, Hang Zhang, Zishuo Dong, Anzheng He
Summary: In this study, a Multi-fusion U-Net network based on U-Net is proposed for pixel-level detection of sealed cracks. The network utilizes a multi-fusion module, dual attention mechanism, and Atrous Spatial Pyramid Pooling to efficiently capture the details of sealed cracks. Experimental results show that the proposed network achieves higher detection accuracy compared to seven state-of-the-art models and enables real-time detection.
Article
Computer Science, Interdisciplinary Applications
Wenlong Ye, Juanjuan Ren, Allen A. Zhang, Chunfang Lu
Summary: This study developed a systematic pixel-level crack segmentation-quantification method for nighttime detection of cracks in slab tracks. The proposed method accurately detects and repairs cracks, providing a new method and theoretical support for slab track maintenance and repair.
COMPUTER-AIDED CIVIL AND INFRASTRUCTURE ENGINEERING
(2023)
Article
Engineering, Civil
Enhui Yang, Youzhi Tang, Allen A. A. Zhang, Kelvin C. P. Wang, Yanjun Qiu
Summary: The paper proposes a fine-tuning method, called policy gradient-based focal loss (focal-PG loss), for trained convolutional neural networks (CNNs). The experimental results show that focal-PG loss greatly improves the crack recognition rate of the trained encoder-decoder network (EDNet) and can also improve the performance of other networks.
JOURNAL OF INFRASTRUCTURE SYSTEMS
(2023)
Article
Engineering, Civil
Xue Helen Yang, Joshua Qiang Li, Chaohui Wang, Kelvin Wang, Jared Schwennesen
Summary: This study applied data envelopment analysis (DEA) to analyze high-risk highway-rail grade crossings in Oklahoma in order to select the most efficient grade separation projects. The results showed that DEA can identify the top 9 most efficient crossings, which were further narrowed down to 4 using the enhanced DEA method.
JOURNAL OF TRANSPORTATION ENGINEERING PART A-SYSTEMS
(2023)
Article
Engineering, Civil
Wenyao Liu, Joshua Qiang Li, Guangwei Yang, Kelvin C. P. Wang, Bryan Wilson, Xue Yang
Summary: This paper presents and analyzes 3 years of monitoring results on 36 HFST sites located in 12 states in the US. Most HFST sites exhibited superior friction values compared with the neighboring nontreated surfaces. Cracking and surface delamination were observed as major failure types on a few sites. The study also developed a multivariant friction prediction model and a safety performance function to quantify the benefits of HFST in reducing roadway departure crashes.
JOURNAL OF TRANSPORTATION ENGINEERING PART B-PAVEMENTS
(2023)
Article
Engineering, Civil
Guolong Wang, Kelvin C. P. Wang, Guangwei Yang, Joshua Qiang Li, Walt Peters
Summary: The evaluation of bumps at bridge approaches is currently limited and varies across different state DOTs. This study aims to propose a general classification criterion for bridge approach bumps and develop a formula for dynamic load allowance estimation. With the use of sub-mm 3D laser imaging technology, 98 bridges in Oklahoma were evaluated for roughness and distress analysis. The results demonstrate the efficiency and effectiveness of this technology in evaluating bridge approach bumps and dynamic load allowance.
JOURNAL OF BRIDGE ENGINEERING
(2023)
Article
Engineering, Multidisciplinary
Hang Zhang, Allen A. A. Zhang, Anzheng He, Zishuo Dong, Yang Liu
Summary: Concurrently detecting multiple objects of interest can save a lot of time and improve the efficiency and uniformity of the detection system. This paper proposes an improved architecture called ShuttleNetV2, which enhances global modeling and fine detail retrieval capabilities. ShuttleNetV2 introduces a self-attention mechanism to capture long-range dependencies and adopts various sampling scales to combine different receptive field characteristics.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Proceedings Paper
Green & Sustainable Science & Technology
Joshua Qiang Li, Stephen A. Cross, Guangwei Yang, Kelvin Wang
Summary: This paper documents the construction and evaluation of an airport runway developed by Oklahoma State University for lightweight unmanned aerial vehicles (UAV). The study includes sampling and testing different asphalt mixtures, as well as evaluating the performance of the runway surface through data collection and analysis. The results demonstrate that the tested UAV runway meets material requirements and shows superior short-term field performance.
INTERNATIONAL CONFERENCE ON TRANSPORTATION AND DEVELOPMENT 2022: PAVEMENTS
(2022)
Proceedings Paper
Engineering, Civil
Lu Gao, Pan Lu, Fengxiang Qiao, Joshua Qiang Li, Yunpeng Zhang, Yihao Ren
Summary: This study evaluates the impact of the COVID-19 pandemic on transportation infrastructure funds in the United States by analyzing motor fuel consumption data. The findings show that the reduced fuel consumption during the pandemic leads to a decrease in gas tax revenue, affecting the rehabilitation and maintenance of transportation infrastructure.
INTERNATIONAL CONFERENCE ON TRANSPORTATION AND DEVELOPMENT 2022: INFRASTRUCTURE SYSTEMS
(2022)
Article
Construction & Building Technology
Wenying Yu, Joshua Qiang Li, Guangwei Yang, Kelvin C. P. Wang, Nii Attoh-Okine
Summary: This study assessed the influences of various characteristics on Continuous Friction Measurement Equipment (CFME) measurements through comprehensive field data collection on testing sites with different preventive maintenance treatments. The results revealed that CFME measurements are most sensitive to changes in water film depth and pavement surface texture properties.
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
(2022)