4.6 Article

Automated Pothole Distress Assessment Using Asphalt Pavement Video Data

Journal

JOURNAL OF COMPUTING IN CIVIL ENGINEERING
Volume 27, Issue 4, Pages 370-378

Publisher

ASCE-AMER SOC CIVIL ENGINEERS
DOI: 10.1061/(ASCE)CP.1943-5487.0000232

Keywords

Asphalt pavements; Potholes; Imaging techniques; Data collection; Pavement assessment; Pothole recognition; Remote sensing; Imaging techniques; Vision tracking

Funding

  1. German Academic Exchange Service (DAAD)
  2. Engineering and Physical Sciences Research Council [EP/K000314/1, EP/I019308/1] Funding Source: researchfish
  3. EPSRC [EP/K000314/1, EP/I019308/1] Funding Source: UKRI

Ask authors/readers for more resources

Potholes, as a severe type of pavement distress, are currently identified and assessed manually in pavement-maintenance programs. This manual process is time-consuming and labor-intensive. Existing methods for automated pothole detection either rely on expensive and high-maintenance range sensors or make use of acceleration data, which only apply when the pothole is on the tires' path. The authors' previous work has proposed and validated a camera-based pothole-detection method. However, this method is limited to single frames and cannot determine the severity of potholes. This paper presents a novel method that addresses these issues by incrementally updating a representative texture template for intact pavement regions and using a vision tracker to reduce the computational effort, improve the detection reliability, and count potholes efficiently. The improved method was implemented and tested on real data. The results indicate a significant capability and performance increase of this method over its predecessor.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.6
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

Article Construction & Building Technology

Vision-based excavator pose estimation using synthetically generated datasets with domain randomization

Amin Assadzadeh, Mehrdad Arashpour, Ioannis Brilakis, Tuan Ngo, Eirini Konstantinou

Summary: This study introduces a framework for synthetically generating large and accurately annotated images for excavator pose estimation, utilizing a game engine and domain randomization. Experimental results show that the model trained on synthetic data can achieve comparable performance to the one trained on real images.

AUTOMATION IN CONSTRUCTION (2022)

Article Computer Science, Artificial Intelligence

A graph-based approach for unpacking construction sequence analysis to evaluate schedules

Ying Hong, Haiyan Xie, Vahan Hovhannisyan, Ioannis Brilakis

Summary: Construction schedules are crucial for project success, but they often require experienced schedulers. This study proposes a graph-based method to find the most time-efficient construction sequence from historic projects, improving scheduling productivity and accuracy. Results indicate that earthwork sequences are the least time-efficient and frequent sequences learned from past projects are closer to the actual schedule.

ADVANCED ENGINEERING INFORMATICS (2022)

Article Construction & Building Technology

Analysis of User Needs in Time-Related Risk Management for Holistic Project Understanding

Haiyan Xie, Ying Hong, Ioannis Brilakis

Summary: This research investigates the obstacles in project execution and proposes a decision tree model to map user needs and trace workflows. The results show that the most urgent needs in time-related risk management are decision support tools, preparation assistance for risk communication, and generation of risk mitigation scenarios.

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT (2022)

Article Chemistry, Multidisciplinary

A BIM Based Framework for Damage Segmentation, Modeling, and Visualization Using IFC

Mathias Artus, Mohamed Said Helmy Alabassy, Christian Koch

Summary: This study proposes a framework that combines automatic damage data acquisition and transfer with a damage information model for data exchange, addressing the issues of paper-based data acquisition and manual transfer between incompatible software or data formats during bridge inspections.

APPLIED SCIENCES-BASEL (2022)

Article Computer Science, Artificial Intelligence

Improving the accuracy of schedule information communication between humans and data

Ying Hong, Haiyan Xie, Gary Bhumbra, Ioannis Brilakis

Summary: This study proposes an ontology-based Recurrent Neural Network approach to bi-directionally translate between human written language and machinery ontological language. Experimental results show that the proposed approach has good performance in text generation accuracy, machine readability, and human understandability.

ADVANCED ENGINEERING INFORMATICS (2022)

Article Construction & Building Technology

Enriching geometric digital twins of buildings with small objects by fusing laser scanning and AI-based image recognition

Yuandong Pan, Alexander Braun, Ioannis Brilakis

Summary: This paper introduces a novel method to enrich geometric digital twins of buildings by capturing important entities from the electrical and fire-safety domain. The method fuses laser scanning and photogrammetry to capture relevant objects, recognize them in 2D images, and map them to a 3D space. The resulting digital twin contains geometric information, semantic information, and useful text information, and can be used for condition monitoring, facility maintenance, and management.

AUTOMATION IN CONSTRUCTION (2022)

Article Computer Science, Interdisciplinary Applications

Object-Oriented Damage Information Modeling Concepts and Implementation for Bridge Inspection

Mathias Artus, Christian Koch

Summary: This study explores the concept of modeling damage information for bridges, defines a data model for damage information independently from the implementation, and explains the implementation process using an established standard. Functional tests show that the standard is suitable for delivering damage information, although several software programs lack proper implementation of the standard.

JOURNAL OF COMPUTING IN CIVIL ENGINEERING (2022)

Article Construction & Building Technology

Scientometric mapping of global research on green retrofitting of existing buildings (GREB): Pathway towards a holistic GREB framework

Mershack O. Tetteh, Amos Darko, Albert P. C. Chan, Amirhosein Jafari, Ioannis Brilakis, Weiwei Chen, Gabriel Nani, Sitsofe Kwame Yevu

Summary: Green retrofitting of existing buildings (GREB) is an effective approach to reduce energy consumption and carbon emissions and improve people's well-being. However, there is a lack of comprehensive investigation into the value of research in this field. This study uses a scientometric review technique to analyze global research on GREB and provides insights for future research. The findings reveal the focus on energy efficiency retrofitting and the declining importance of occupant behavior and indoor environmental quality in current GREB practices.

ENERGY AND BUILDINGS (2022)

Article Construction & Building Technology

Graph-Based Automated Construction Scheduling without the Use of BIM

Ying Hong, Haiyan Xie, Eva Agapaki, Ioannis Brilakis

Summary: The construction industry has long struggled with delays and cost overruns. This paper proposes a graph-based automated scheduling (GAS) method to capture, store, and reuse the tacit knowledge in construction schedules. The GAS method was validated on two case studies and proved to be more accurate in generating construction schedules compared to planned schedules.

JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT (2023)

Article Computer Science, Interdisciplinary Applications

ConSLAM: Construction Data Set for SLAM

Maciej Trzeciak, Kacper Pluta, Yasmin Fathy, Lucio Alcalde, Stanley Chee, Antony Bromley, Ioannis Brilakis, Pierre Alliez

Summary: This paper introduces a periodically collected data set on a construction site, aiming to evaluate the performance of SLAM algorithms used by mobile scanners or autonomous robots. The data set includes ground-truth scans, spatially registered and time-synchronized images, lidar scans, and inertial data. The paper also demonstrates how to measure the accuracy of SLAM algorithms against the ground-truth trajectory using a popular software package. This is the first publicly accessible data set of sequentially collected data on a construction site.

JOURNAL OF COMPUTING IN CIVIL ENGINEERING (2023)

Review Chemistry, Analytical

Construction and Maintenance of Building Geometric Digital Twins: State of the Art Review

Viktor Drobnyi, Zhiqi Hu, Yasmin Fathy, Ioannis Brilakis

Summary: Most existing buildings were built based on 2D drawings, but building information models have become prevalent in recent years. However, it will take a long time for these models to be widely adopted in all existing buildings. This paper reviews the state-of-the-art practice and research for constructing and maintaining geometric digital twins, and proposes a new geometry-based object class hierarchy to prioritize automation.

SENSORS (2023)

Article Construction & Building Technology

Assessing a thermoelectric radiative cooling partition as a personalised comfort system using empirical experiments enhanced by digital shadow visualisation

Ammar Hassan Osman, Mathias Artus, Hayder Alsaad, Christian Koch, Conrad Voelker

Summary: This study investigates the cooling effect of the thermoelectric cooling partition (Thecla) as a personalised comfort system and its impact on thermal comfort. The results show that Thecla has a better effect on the body under higher temperatures, and the integration of virtual reality enhances the visualization of thermal impact.

BUILDING AND ENVIRONMENT (2023)

Article Computer Science, Interdisciplinary Applications

Dense 3D Reconstruction of Building Scenes by AI-Based Camera-Lidar Fusion and Odometry

Maciej Trzeciak, Ioannis Brilakis

Summary: In this paper, a dense 3D reconstruction pipeline is proposed to improve the resolution of point clouds captured by handheld scanners. Time-synchronized and spatially registered images and lidar sweeps are fused using spatial AI methods to generate higher resolution dense scans for progressive reconstruction. The results showed a reduction of 11% in overall point cloud noise and an increase in density by approximately six times.

JOURNAL OF COMPUTING IN CIVIL ENGINEERING (2023)

Proceedings Paper Computer Science, Artificial Intelligence

CLOI: An Automated Benchmark Framework for Generating Geometric Digital Twins of Industrial Facilities

Eva Agapaki, Ioannis Brilakis

Summary: This paper presents a framework called CLOI, which accurately generates labeled point clusters of important shapes in existing industrial facilities with minimal manual effort. CLOI combines deep learning and geometric methods to segment points into classes and individual instances. Experimental results show that CLOI can reliably segment complex and incomplete point clouds of industrial facilities, achieving 82% class segmentation accuracy. Compared to current practices, the proposed framework can save an estimated 30% of time on average.

COMPUTING IN CIVIL ENGINEERING 2021 (2022)

Article Construction & Building Technology

Framework for BIM-Based Simulation of Construction Operations Implemented in a Game Engine

Carlos A. Osorio-Sandoval, Walid Tizani, Estacio Pereira, Jelena Ninic, Christian Koch

Summary: This article demonstrates how BIM and a game engine can be used to facilitate the development and application of construction simulation models, reducing the complexity and improving the reuse of the models.

BUILDINGS (2022)

No Data Available