Article
Computer Science, Artificial Intelligence
Tin Barisin, Christian Jung, Franziska Muesebeck, Claudia Redenbach, Katja Schladitz
Summary: This paper reviews and compares automatic crack segmentation methods for 3D images, including classical image processing methods and learning methods. Learning methods perform the best in thin cracks and low grayvalue contrast.
PATTERN RECOGNITION
(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
Chemistry, Analytical
Jie Kang, Shujie Feng
Summary: A pavement cracks segmentation method based on conditional generative adversarial network is proposed in this paper, which achieves accurate crack detection through training of generator and discriminator, showing superior performance in complex background.
Article
Construction & Building Technology
Qinghua Han, Xuan Liu, Jie Xu
Summary: This paper proposes an image-based detection and location method for cracks on the surface using unmanned aerial vehicle, which can detect and avoid potential safety accidents caused by cracks. The method consists of steps such as super-pixel segmentation, damage area detection, pixel-level identification, and crack location.
JOURNAL OF BUILDING ENGINEERING
(2022)
Article
Construction & Building Technology
Shuwei Li, Xingyu Gu, Xiangrong Xu, Dawei Xu, Tianjie Zhang, Zhen Liu, Qiao Dong
Summary: This study introduces an effective method utilizing 3-D ground penetrating radar (GPR) and deep learning models to automatically detect concealed cracks in asphalt pavement. The results demonstrate the feasibility of the proposed method, with YOLOv4 and YOLOv5 models showing significant advancements in crack detection even with a small dataset.
CONSTRUCTION AND BUILDING MATERIALS
(2021)
Article
Automation & Control Systems
Wangzhe Du, Hongyao Shen, Jianzhong Fu
Summary: In this article, three methods were proposed to improve the segmentation of X-ray images, including enhancing contrast using CLAHE, establishing a two-stream CNN for image processing, and introducing a weighted IOU loss function. Experimental results show that these methods outperform the baseline in terms of different metrics, indicating better performance and effectiveness in object segmentation.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Engineering, Multidisciplinary
Li Song, Hongshuo Sun, Jinliang Liu, Zhiwu Yu, Chenxing Cui
Summary: The article introduces a method for automatic detection and quantification of cracks on concrete structure surfaces. The method uses an electric driving platform for close-range scanning and shooting, applies convolutional neural networks for automatic segmentation, and employs crack matching and property calculation methods for automatic quantification. Experimental results demonstrate high accuracy of the method.
Article
Engineering, Multidisciplinary
Behrouz Mataei, Fereidoon Moghadas Nejad, Hamzeh Zakeri
Summary: This paper presents a 3D automatic device based on a cumulative imaging technique for measuring pavement texture. The device generates 3D point cloud models and extracts evaluation indices for assessing pavement texture, including a new method for evaluating texture in rainy conditions. Experimental tests demonstrate a high correlation between the results of this system and traditional sand patch tests.
Article
Radiology, Nuclear Medicine & Medical Imaging
Chengyin Li, Hassan Bagher-Ebadian, Rafi Ibn Sultan, Mohamed Elshaikh, Benjamin Movsas, Dongxiao Zhu, Indrin J. Chetty
Summary: The purpose of this research was to develop and optimize a new architecture for automatically segmenting the prostate gland and normal organs in medical images. The researchers combined a shifted-window transformer with a convolutional U-Net to create the SwinAttUNet network, which achieved high accuracy in segmenting multi-organ anatomy.
Article
Chemistry, Multidisciplinary
Pang-jo Chun, Tatsuro Yamane, Yukino Tsuzuki
Summary: The technology uses a convolutional neural network to automatically detect and evaluate cracks in pavement images, improving accuracy through retraining with previously misanalyzed images, with experiments confirming high system performance.
APPLIED SCIENCES-BASEL
(2021)
Article
Construction & Building Technology
Niannian Wang, Jiaxiu Dong, Hongyuan Fang, Bin Li, Kejie Zhai, Duo Ma, Yibo Shen, Haobang Hu
Summary: A low cost and automatic 3D reconstruction and segmentation system for potholes is proposed in this study. It consists of pavement pothole Structure-from-motion (PP-SFM) and a 3D point cloud segmentation network. PP-SFM is used for 3D reconstruction of multi-view 2D pothole images, and Trans-3DSeg with transformer module is developed for effective segmentation of 3D point cloud data. Experimental results show that the proposed system has better segmentation performance with low cost.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Construction & Building Technology
Zhen Liu, Xingyu Gu, Jiaqi Chen, Danyu Wang, Yihan Chen, Lutai Wang
Summary: This study proposes a YOLOv3 model with four-scale detection layers (FDL) to detect cracks in GPR images. Results show that the YOLOv3-FDL model achieves higher F1 score and mAP on the GPR dataset, indicating improved detection performance compared to the YOLOv3 model.
AUTOMATION IN CONSTRUCTION
(2023)
Article
Construction & Building Technology
Zhong Zhou, Longbin Yan, Junjie Zhang, Yidi Zheng, Chenjie Gong, Hao Yang, E. Deng
Summary: In order to tackle the challenges of complex environmental interference and multiscale targets in deep learning-based tunnel lining defect identification, a novel segmentation algorithm called MC-TLD is proposed. MC-TLD utilizes a context-enhanced feature encoder to extract global context information, a multiscale attention-based atrous spatial pyramid pooling module to improve feature extraction for different scales of defects, and parameter learnable DUpsampling in the feature decoding module to output more accurate pixel prediction results. Experimental results demonstrate that MC-TLD achieves higher segmentation accuracy compared to other models, making it suitable for tunnel lining multiscale defect identification in complex environments.
CONSTRUCTION AND BUILDING MATERIALS
(2023)
Article
Engineering, Civil
Duo Ma, Hongyuan Fang, Niannian Wang, Chao Zhang, Jiaxiu Dong, Haobang Hu
Summary: The proposed system utilizes a PCGAN to generate realistic crack images and a YOLO-MF network for crack detection and tracking, achieving high accuracy and improved detection speed. The system includes a calculating module, an automated unmanned aerial vehicle, and other components for on-site measurement and detection.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Agriculture, Multidisciplinary
Hengxiang He, Yulong Qiao, Ximeng Li, Chunyu Chen, Xingfu Zhang
Summary: Measuring pig weight based on depth images can avoid the time-consuming and stressful process of weighing with scales. By designing preprocessing algorithms and a regression network based on BotNet, we achieve good accuracy in predicting pig weight.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2021)
Article
Construction & Building Technology
Runhua Guo, Donglei Fu, Giuseppe Sollazzo
Summary: This paper introduces an ensemble learning model utilizing Gradient Boosting Decision Tree to predict rut depth and International Roughness Index, determines optimal hyperparameters through grid search and cross-validation, interprets model results and importance of factors using SHAP, and outperforms other AI methods in prediction quality, providing practical applications for pavement maintenance and budget optimization.
INTERNATIONAL JOURNAL OF PAVEMENT 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
Construction & Building Technology
G. Bosurgi, O. Pellegrino, G. Sollazzo
Summary: The article discusses the challenges of data management arising from the increasing capability for road pavement condition inspections in recent years. It proposes a solution for presenting and elaborating pavement condition information in an I-BIM environment and tests its preliminary effectiveness on an existing highway.
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
(2022)
Article
Construction & Building Technology
Gaetano Bosurgi, Dario Bruneo, Fabrizio De Vita, Orazio Pellegrino, Giuseppe Sollazzo
Summary: Recently, the road industry has been transitioning towards big-data and Internet of Things concepts inspired by 'Industry 4.0'. Smart roads are using modern instrumented vehicles and sensor networks to provide useful data, but face challenges in data management and processing.
STRUCTURAL CONTROL & HEALTH MONITORING
(2022)
Article
Construction & Building Technology
G. Bosurgi, M. Modica, O. Pellegrino, G. Sollazzo
Summary: The authors propose an algorithm for automated pothole detection through the processing of 3D data of pavement surfaces, which shows remarkable performance compared to traditional approaches, allowing road administrators to optimize maintenance and road functionality.
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
(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
Chemistry, Analytical
Gaetano Bosurgi, Orazio Pellegrino, Alessia Ruggeri, Giuseppe Sollazzo
Summary: The presence of sensors in modern vehicles has led to the development of new driving assistance tools, but their effectiveness depends on the environmental context. This study proposes a method to quantify the effectiveness of warnings produced by an On-Board Unit (OBU) in a specific environmental context. The experiment was conducted with a sample of young users using a driving simulator, and the results showed that the OBU provided useful information to the driver without affecting their performance.
Article
Green & Sustainable Science & Technology
Gaetano Bosurgi, Orazio Pellegrino, Alessia Ruggeri, Giuseppe Sollazzo
Summary: Traffic loads and environmental factors can cause distress on road pavements, leading to hazardous conditions for drivers, especially during rain. Current methods for evaluating road surfaces focus on the detection of distresses through synthetic indicators, but fail to consider the hydraulic efficiency of the carriageway. In this paper, a synthetic indicator called Hydraulic Condition Index (HCI) is proposed to evaluate the hydraulic quality of road pavement surfaces, which can help road agencies prioritize maintenance needs and improve traffic safety.
Article
Green & Sustainable Science & Technology
Gaetano Bosurgi, Orazio Pellegrino, Alessia Ruggeri, Giuseppe Sollazzo
Summary: This study aims to determine whether drivers receive data appropriately during specific maneuvers and whether the data provided creates unnecessary burden. The results from an experimentation in a simulated environment show that in complex situations, drivers rely more on a smaller number of information sources rather than the driving aid device inside the cockpit.
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
Construction & Building Technology
Gaetano Bosurgi, Clara Celauro, Orazio Pellegrino, Alessia Ruggeri, Giuseppe Sollazzo
Summary: The use of waste materials in the construction of HMA pavements is an important opportunity for environmental sustainability. Experimental analysis compared the performance of traditional surface mixture with two alternative mixtures containing basalt aggregates and steel slags, respectively. The laboratory results showed comparable structural performance for the alternative mixtures and remarkable benefits in terms of environmental sustainability.
CONSTRUCTION AND BUILDING MATERIALS
(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, Industrial
Gaetano Bosurgi, Stellario Marra, Orazio Pellegrino, Giuseppe Sollazzo
Summary: When there are insurmountable constraints in a road design, a rational procedure based on vehicles telemetry data in a simulated environment is proposed to achieve a certain level of safety for drivers. The procedure allows comparison of two road geometries through synthetic indices, indicating that a certain deviation from standards did not affect driver performance. Compared to existing literature, this study provides a fully objective procedure based on a new indicator that can be easily adapted to different contexts involving drivers, roads, and vehicles.
COGNITION TECHNOLOGY & WORK
(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
Engineering, Civil
Gaetano Bosurgi, Stellario Marra, Orazio Pellegrino, Giuseppe Sollazzo, Massimo Villari
Summary: Exhaust gas emissions from motor vehicles are influenced by road features, vehicle type, and driving behavior. This study examines the relationship between longitudinal acceleration, CO2 emissions, and road geometrical features to improve road design quality. Experimental results reveal that emissions are influenced by driving consistency, including variables such as visibility, understanding of road alignment, and similarity of adjacent elements. The study raises questions about existing models for speed forecasting and suggests the importance of considering environmental impacts in road design.
TRANSBALTICA XII: TRANSPORTATION SCIENCE AND TECHNOLOGY
(2022)