Showing papers in "Computer-aided Civil and Infrastructure Engineering in 2021"
TL;DR: Combining a progressive growing GAN along with Poisson blending artificially generates road damage images that can be used as new training data to improve the accuracy of road damage detection and the new Road Damage Dataset 2019 is released.
Abstract: Machine learning can produce promising results when sufficient training data are available; however, infrastructure inspections typically do not provide sufficient training data for road d...
131 citations
TL;DR: An automated crack detection method based on image processing using the light gradient boosting machine (LightGBM), one of the supervised machine learning methods, that can detect cracks with high accuracy and training time is shortened.
Abstract: Automated crack detection based on image processing is widely used when inspecting concrete structures. The existing methods for crack detection are not yet accurate enough due to the diff...
97 citations
TL;DR: Bayesian inference is used for deep vision SHM models where uncertainty can be quantified using the Monte Carlo dropout sampling and the concept of surrogate models is proposed to develop the models for uncertainty‐assisted segmentation and prediction quality tagging.
86 citations
TL;DR: A transfer learning pipeline is proposed to address this problem, which enables a distress detection model to be applied to other untrained scenarios and can reduce the demand for training data by at least 25% when the model is applied in a new scene.
72 citations
TL;DR: A framework for real‐time regional seismic damage assessment that is based on a Long Short‐Term Memory (LSTM) neural network architecture is proposed and can perform damage assessment in real time at regional scale with high prediction accuracy and acceptable variance.
70 citations
TL;DR: Results show that reasonable idle times can be generated by optimizing the scheduling plan, and it is helpful to stop the accumulation of stochastic volatilities.
Abstract: This paper develops a vehicle scheduling method for the electric bus (EB) route considering stochastic volatilities in trip travel time and energy consumption. First, a model for estimating the trip energy consumption is proposed based on field-collected data, and the probability distribution function of trip energy consumption considering the stochastic volatility is determined. Second, we propose the charging strategy to recharge buses during their idle times. The impacts of stochastic volatilities on the departure time, the idle time, the battery state of charge, and the energy consumption of each trip are analyzed. Third, an optimization model is built with the objectives of minimizing the expectation of delays in trip departure times, the summation of energy consumption expectations, and bus procurement costs. Finally, a real bus route is taken as an example to validate the proposed method. Results show that reasonable idle times can be generated by optimizing the scheduling plan, and it is helpful to stop the accumulation of stochastic volatilities. Collaboratively optimizing vehicle scheduling and charging plans can reduce the EB fleet and delay times while meeting the route operation needs.
68 citations
TL;DR: In this paper, the tension forces of cables and hangers are measured for ensuring the safety of long-span bridges, and the cable tensor tensor forces are estimated for each bridge.
Abstract: Cables and hangers are critical components of long‐span bridges, tension forces of them are needed to be accurately measured for ensuring the safety of bridges. Traditionally, cable tensio...
64 citations
TL;DR: A new methodology for detecting safety helmet wearing is proposed, which makes use of convolutional neural network‐based face detection and bounding‐box regression for safety helmet detection and deep transfer learning based on DenseNet is introduced and applied.
61 citations
TL;DR: A connected autonomous vehicle (CAV) network can be defined as a set of connected vehicles including CAVs that operate on a specific spatial scope that may be a road network, corridor, or highway as discussed by the authors.
Abstract: A connected autonomous vehicle (CAV) network can be defined as a set of connected vehicles including CAVs that operate on a specific spatial scope that may be a road network, corridor, or ...
61 citations
TL;DR: Compared with some well‐known deep convolutional neural networks, the FF‐BLS achieved a similar level of recognition accuracy, but the training speed was increased by more than 20 times, and this substantially reduces the training cost.
54 citations
TL;DR: A deep learning‐based automated crack evaluation technique for a high‐rise bridge pier using a ring‐type climbing robot that successfully evaluates cracks on the entire ROI of the bridge pier with precision of 90.92% and recall of 97.47%.
Abstract: This article proposes a deep learning‐based automated crack evaluation technique for a high‐rise bridge pier using a ring‐type climbing robot. First, a ring‐type climbing robot system comp...
TL;DR: A novel architecture is proposed to fully utilize the complementarity between the RS and the RE to accurately identify the RS with well‐defined boundaries and an innovative hybrid loss consisting of binary cross entropy, structural similarity index measure, and intersection‐over‐union is proposed and equipped into the RBGNet.
TL;DR: In this article, the authors aim to improve post-disaster preliminary damage assessment (PDA) using artificial intelligence (AI) and unmanned aerial vehicle (UAV) imagery.
Abstract: This study aims to improve post‐disaster preliminary damage assessment (PDA) using artificial intelligence (AI) and unmanned aerial vehicle (UAV) imagery. In particular, a stacked convolut...
TL;DR: A one‐dimensional, memory‐augmented convolutional neural network inspired by the memory‐AUgmented neural network (MANN), which has capacity to address new scenarios from unknown distributions and can effectively address the issue of new categories without retraining.
TL;DR: A semi‐supervised learning algorithm that uses only a small amount of labeled data for training, but still achieves high classification accuracy is proposed, and a novel uncertainty filter to select reliable unlabeled data for initial training epochs is developed to further improve the classification accuracy.
TL;DR: The first public rail components image database, including rails, spikes, and clips, is built and released online and a real‐time pixel‐level detection framework with improved real-time instance segmentation models is developed.
TL;DR: A new PAnoramic surface damage DEtection Network (PADENet) is presented, employing the proposed multiple projection methods to process high‐resolution images, and modifying the faster region‐based convolutional neural network and training via transfer learning on VGG‐16, which improves the precision for detecting multiple types of damage in distortion.
TL;DR: In this article, a cooperative traffic control strategy to increase the capacity of non-recurrent bottlenecks such as work zones by making full use of the spatial resources upstream of work zones is presented.
Abstract: This paper presents a cooperative traffic control strategy to increase the capacity of nonrecurrent bottlenecks such as work zones by making full use of the spatial resources upstream of w...
TL;DR: In this article, a vision-based approach using unmanned aerial vehicles (UAVs) mounted with high-resolution cameras was proposed to assess the health of civil infrastructure in the Middle East.
Abstract: Structural displacement is an important quantity to assess the health of civil infrastructure. Vision‐based approaches using unmanned aerial vehicles (UAV) mounted with high‐resolution cam...
TL;DR: Physics-informed neural networks (PINNs) as mentioned in this paper are a class of deep neural networks that are trained, using automatic differentiation, to compute the response of systems governed by partial diffe...
Abstract: Physics‐informed neural networks (PINNs) are a class of deep neural networks that are trained, using automatic differentiation, to compute the response of systems governed by partial diffe...
TL;DR: Considering the network‐wide spatiotemporal correlations, the TTI‐GAN can generate travel times for links without sufficient observations by modeling travel time distributions (TTDs) for links with rich data and performs better than other counterparts in imputing mean travel times under various data missing rates.
TL;DR: The results verify that the DRL approach used here can automatically explore and optimize the railway alignment, decreasing the construction cost by 17.65% and 7.98%, compared with the manual alignment and with the results of a method based on the distance transform, respectively, while satisfying various alignment constraints.
TL;DR: Numerical examples for aerodynamic shape optimization of a tall building are used to demonstrate that the proposed knowledge‐enhanced deep RL‐based shape optimizer outperforms both gradient‐based and gradient‐free optimization algorithms.
Abstract: Structural shape optimization plays an important role in the design of wind‐sensitive structures. The numerical evaluation of aerodynamic performance for each shape search and update durin...
TL;DR: In this article, the authors apply deep learning (DL) to assess structural damages in a vision-based structural health monitoring (SHM) system, but both data deficiency an...
Abstract: In recent years, applying deep learning (DL) to assess structural damages has gained growing popularity in vision‐based structural health monitoring (SHM). However, both data deficiency an...
TL;DR: This paper presents a new behavioral decision‐making model to achieve both safety and high efficiency and also to reduce the adverse effect of autonomous vehicles on the other road users while driving and proposes a combined spring model for assessing driving risk.
Abstract: Intelligent‐driving technologies play crucial roles in reducing road‐traffic accidents and ensuring more convenience while driving. One of the significant challenges in developing an intel...
TL;DR: In this paper, the authors considered many factors, such as drastically undulating terrain, geological hazard impacts, and railway alignment optimization for mountain railway alignment, and proposed a new method to solve the problem.
Abstract: Mountain railway alignment optimization is known as a very complex engineering problem that should consider many factors, such as drastically undulating terrain, geological hazard impacts,...