Search

About 80-90% of accidents are caused by the unsafe actions and behaviors of employees in construction. Behavior management thus plays a key role in enhancing safety, and particularly, behavior observation is the most critical element for modifying workers' behavior in a safe manner. However, there is a lack of practical methods to measure workers' behavior in construction. To analyze workers' actions, this paper uses an advanced and economical depth sensor to collect motion data and then investigates consequent motion-analysis techniques to detect the unsafe actions of workers, which is the main focus of this paper. First, motion data are transformed onto a three-dimensional (3D) space as a preprocess, motion classification is performed to identify a typical prior, and the selected prior is used to detect the same action in a testing data set. As a case study, motion data for unsafe actions in ladder climbing (i.e., backward-facing climbing, climbing with an object, and reaching far to a side) are collected and used to detect the actions in a new testing data set in which the actions are randomly taken. The result shows that 90.91% of unsafe actions are correctly detected in the experiment. [ABSTRACT FROM AUTHOR]

This paper describes a new method of safety assessment of highway stopping and passing sight distances. Topographic data were collected by flying a plane equipped with a light detection and ranging (LiDAR) system on a section of two-lane rural highway. The collected data were manipulated in a geographic information system (GIS) to obtain three-dimensional models of the various highway subsections. These models were visually analyzed to check for stopping and passing sight distances, thus narrowing the list of locations with potential safety issues. Subsequent line-of-sight analysis in a GIS indicated the presence of several obstructions, prompting a field visit for validation. Field inspection of the locations verified the presence of the obstructions detected during the analysis. Overall, this research indicated the viability of successful visualization of LiDAR data for highway sight distance evaluations and is indicative of LiDAR’s potential for highway safety-related applications. Comments on the resources needed to embrace this technology and future research improvements are part of the paper. [ABSTRACT FROM AUTHOR]

Three-dimensional (3D) as-built geometric models are useful for many building assessment and management tasks. However, the current process of creating such models is labor-intensive. A significant amount of manual work is required to register the remote sensing data captured from multiple scans into one scene (i.e., scene registration). To automate the registration work, several research studies have been developed to automate the registration process by the detection and matching of common visual features in consecutive scans. This paper investigates the effectiveness of different combinations of common visual feature detectors and descriptors that have been widely used in the scene registration of 3D buildings. The evaluation criteria include registration accuracy and speed. The feature detectors and descriptors have been tested in a total of 31 realistic building scenarios. The results show that the combination of the scale-invariant feature transform feature detector and descriptor reached more accurate results than the others. The fastest speed is achieved by the use of an oriented binary robust independent elementary features (ORB) detector in combination with the speeded-up robust features or ORB descriptor. [ABSTRACT FROM AUTHOR]

Optical imaging techniques have been commonly used in civil engineering practice for aiding the archival of damage scenes and more recently, for image-based damage analysis. However, an evident limitation in the current practice is the lacking of real-time imaging, computing, and analytics capability for team-based visual inspection in a complex built environment. This paper explores a new computing paradigm called collaborative mobile-cloud computing (CMCC) and proposes a CMCC framework for conducting intelligent civil infrastructure condition inspection. Through software design, this framework synthesizes advanced mobile and cloud computing technologies with three innovative features: (1) context-enabled image collection, (2) interactive imaging and processing, and (3) real-time on-demand image analysis and condition analytics. Through field experiments and computational performance evaluation, this paper demonstrates the feasibility of the proposed CMCC framework, which includes verification of real-time imaging, analytics, and particularly, the mobile-cloud computational solution to two representative damage analysis problems regarding complex imagery scenes. [ABSTRACT FROM AUTHOR]

Empirical mode decomposition (EMD) is a multiresolution data analysis method recently developed to cater to the inherent nonstationarity in real-world signals. A two-dimensional (2D) extension of EMD is used in this paper as a pavement distress image analytical tool. The algorithm decomposes an image into a set of narrow band components (called bidimensional intrinsic mode function, or BIMF) that uniquely reflect the variations in the image. Although some components could have good image edge characteristics, others might hold fidelity to the shape and size of objects or trends in the image. Therefore, the complete spatial and frequency characteristic of a desired image feature might also be divided into the different components, indicating that all attributes of a desired feature might not be found in a single component. An optimal solution requires image mining from the different component resolutions to accurately extract those specific attributes without compromising certain spatial and frequency characteristics. Two major contributions to pavement image analysis are achieved. First, the paper explores pavement image denoising or enhancement by combining the EMD and a weighted reconstruction technique as a tool for background standardization of images acquired under different types of illumination effects. Second, using principal component pursuit (PCP), the authors reconstruct a composite image by selecting salient information from coarse and fine resolution BIMFs useful for accurate extraction of linear patterns in a pavement distress image. Compared with conventional image reconstruction or approximation techniques, the methodology used in this paper yields better results. [ABSTRACT FROM AUTHOR]

Vision trackers have been proposed as a promising alternative for tracking at large-scale, congested construction sites. They provide the location of a large number of entities in a camera view across frames. However, vision trackers provide only two-dimensional (2D) pixel coordinates, which are not adequate for construction applications. This paper proposes and validates a method that overcomes this limitation by employing stereo cameras and converting 2D pixel coordinates to three-dimensional (3D) metric coordinates. The proposed method consists of four steps: camera calibration, camera pose estimation, 2D tracking, and triangulation. Given that the method employs fixed, calibrated stereo cameras with a long baseline, appropriate algorithms are selected for each step. Once the first two steps reveal camera system parameters, the third step determines 2D pixel coordinates of entities in subsequent frames. The 2D coordinates are triangulated on the basis of the camera system parameters to obtain 3D coordinates. The methodology presented in this paper has been implemented and tested with data collected from a construction site. The results demonstrate the suitability of this method for on-site tracking purposes. [ABSTRACT FROM AUTHOR]

Three-dimensional visualization is getting increasingly popular. However, for realizing the full benefits of information technology (IT), we need to go beyond visualization and adopt modeling techniques that allow reuse of information for other tasks in the project life cycle. Already it is possible to use visualization models for tasks such as simulation. However, widespread use of these integrated systems is hindered by the lack of education in the fundamentals of IT in the engineering curriculum. This paper analyzes the dependencies between concepts related to visualization and lists a minimum set of topics that should be part of the curriculum. Experiences in teaching these in a new course module related to visualization are described. Feedback data collected shows that most students appreciated the use of IT tools in education. However, comments by many students indicated inherent problems related to IT and visualization. Nevertheless, it is expected that when more universities start teaching fundamentals we should see more IT adoption in the industry. [ABSTRACT FROM AUTHOR]

Computational approaches to visual data and information processing can assist asset management procedures. In this paper, we present a computerized visual pattern recognition approach to facilitate inspection and condition assessment of wastewater collection systems in realistic settings. This research aims to provide a basis for condition assessment and defect reporting in wastewater infrastructure, enabling automatic detection and recognition of defects and patterns from inspection images or videos. An approach for sensing and automatically detecting critical areas or regions of interest, and further discovering and recognizing objects of interest in large image data sets, is described. The major steps of the approach are illustrated by examples using actual inspection images acquired from wastewater pipelines. [ABSTRACT FROM AUTHOR]

Construction progress is predominantly measured with manual site surveys. These surveys are labor-intensive, on-site manual investigations. The generated survey reports are subjective and approximate because they are based on the surveyors' individual experiences. This paper presents a novel method that can automatically count the number of bricks on a façade for reducing the cost and increasing the reliability of progress surveys. The method uses video data taken from a user's mobile phone to detect bricks on a façade in each video frame by using color thresholding, edge detection, and filtering of rectangular shapes and sizes. Then, it compares the difference between consecutive frames to add counts when new bricks appear and to avoid double counting. The proposed method was implemented and tested on on-site videos of red brick façades, and resulted in 99.8% precision and 98.7% recall. The results demonstrate the suitability of this method for progress monitoring of brick façade construction. [ABSTRACT FROM AUTHOR]

Partially saturated compacted-sand specimens were characterized by using three-dimensional (3D) image registration of dualmodal (neutron and X-ray) tomography data. Neutron and X-ray imaging provide complementary information for precisely identifying the three phases (silica sand, air, and water) of a compacted sand specimen that is partially saturated. Neutron tomography provides high contrast of the water phase, whereas X-ray tomography provides high contrast of the silica sand phase due to different fundamental interaction mechanisms of neutron and X-ray with matters. X-ray interacts with the electron cloud surrounding the nucleus, whereas neutron radiation interacts with the nucleus of an atom. In this paper, a computational technique was developed to unify digital images of dual-modal data obtained at different image resolution and specimen orientations based on the maximization of the normalized mutual information to combine the information from the water phase inferred from a neutron image and information from the sand and pore phases obtained from an X-ray image. The combined microstructure obtained from registered X-ray and neutron images can thus be analyzed in 3D to precisely differentiate silica, air, and water distribution spatially. Microstructure information for granular assembly including pore size distribution and coordination number was determined from X-ray tomography data due to higher precision in identifying solid particle boundaries. Local values of saturation along the height of the specimen are obtained from the pore and water phases segmented from the registered neutron and X-ray images. Image registration of neutron and X-ray images provides the ability to obtain the microstructure of partially saturated sand in terms of the detailed variation of solid particles, air, and water phases spatially. [ABSTRACT FROM AUTHOR]

Natural disasters all too often place human lives and property at risk. Recovery efforts following a disaster can be slow and painstaking work, and potentially put responders in harm's way. A system which helps identify defects in critical building elements (e.g., concrete columns) before responders must enter a structure could save lives. In this paper we propose such a system, centered around an image-based three-dimensional (3D) reconstruction method and a new 3D crack detection algorithm. The image-based method is capable of detecting and analyzing surface damages in 3D. We also demonstrate how a robotic platform could be used to gather the set of images from which the reconstruction is created, further reducing the risk to responders. In this regard, image-based 3D reconstructions represent a convenient method of creating 3D models because most robotic platforms can carry a lightweight camera payload. Additionally, the proposed 3D crack detection algorithm also provides the advantage of being able to operate on 3D mesh models regardless of their data collection source. Our experimental results showed that the 3D crack detection algorithm performed well on several sample building elements, successfully identifying cracks, reconstructing 3D profiles, and measuring geometrical characteristics on damaged elements and not finding any cracks on intact ones. The operation and perceived benefits of the proposed method in a post-disaster situation are also discussed in detail. [ABSTRACT FROM AUTHOR]

Thermal images mosaicking, registration, and analysis ( THIMRAN) is a new MATLAB toolbox proposed for processing data provided by infrared thermography (IRT). The toolbox is able to perform a contactless recognition of damage of a large body like a masonry/concrete wall or a rock cliff, leading to useful information for health state evaluation and/or hazard assessment. In an IRT measurement session, a series of thermal images are acquired throughout one or more heating/cooling diurnal cycles. The proposed approach relies on evaluation of the time history of the thermal contrast related to one or more thermal transients. This is an extension to large bodies of a method currently used in laboratory nondestructive testing. The toolbox contains all the necessary functions for a complete IRT-based evaluation of a large body: (1) preliminary image processing (mosaicking, or stitching, and registration); (2) computation of thermal contrast for each image; and (3) damage recognition by means of pixel-by-pixel comparison of thermal images registered into the same reference frame. The toolbox is available as Supplemental Data of this paper. THIMRAN was applied to two case studies, i.e., a leaning bell tower and a rock cliff, which are also described. [ABSTRACT FROM AUTHOR]

Quick and reliable identification of energy performance problems in buildings is a critical step in improving their efficiency. The current practice of building diagnostics typically involves nonintrusive data collection using thermal cameras. This requires large amounts of unordered and nongeo-tagged two-dimensional (2D) imagery to be manually analyzed at a later stage, which makes the analysis time-consuming and labor-intensive. Because of the absence of a benchmark for energy performance, identification of performance problems also heavily relies on the auditor's knowledge, and consequently may lead to subjective and error-prone inspections. As a step towards rapid and objective identification of performance problems, this paper presents a new method for automated analysis and visualization of deviations between buildings' actual and simulated energy performances. The proposed method is based on the recently developed energy performance augmented reality (EPAR) environments. In the EPAR modeling method, actual and expected three-dimensional (3D) spatio-thermal models are generated and superimposed in a common 3D virtual environment. The method leverages unordered collections of thermal and digital imagery for actual energy performance modeling, in addition to computational fluid dynamics (CFD) analysis for expected energy performance simulation. Based on the EPAR models which store actual and simulated thermal values at the level of 3D points, two new algorithms are developed to facilitate identification of potential performance problems: (1) 3D thermal mesh modeling using trees and nearest-neighbor searching to automate calculation of temperature deviations, and (2) automated visualization of performance deviations using a metaphor based on traffic light colors. The proposed modeling method is validated on several interior locations of instructional and residential buildings. Empirical observations show that automated analysis using EPAR models enables performance deviations to be rapidly and accurately measured. The visualization of deviations in three dimensions enables auditors to easily identify potential performance problems and, in turn, enables auditors to focus more on other important tasks of analyzing retrofit alternatives. [ABSTRACT FROM AUTHOR]

Assembly is the process in which two or more objects are joined together. An assembly manual is typically used to guide the tasks required to put together an artifact. As an emerging technology, augmented reality (AR) integrates three-dimensional (3D) images of virtual objects into a real-world workspace. The insertion of digitalized information into the real workspace using AR can provide workers with the means to implement correct assembly procedures with improved accuracy and reduce errors. A prototype animated AR system was configured for assembly tasks that are normally guided by reference to documentation and was tested using a series of experiments. A LEGO model was used as the assembly and experimental tester task. Experimentation was devised and conducted to validate the cognitive gains that can be derived from using AR to assemble a LEGO model. Two formal experiments with 50 participants were conducted to compare an animated AR system and the paper-based manual system. One experiment measured the cognitive workload of using the system for assembly, whereas the other measured the learning curves of novice assemblers. Findings from the experiments revealed that the animated AR system yielded shorter task completion times, less assembly errors, and lower total task load. The results also revealed that the learning curve of novice assemblers was reduced and task performance relevant to working memory was increased when using AR training. Future work will apply the knowledge gained from the controlled assembly experiments to the real-scale construction assembly scenario to measure the productivity improvements. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

This paper describes an automated classification approach to road users. The main motivation behind road-user classification in the context of safety stems from the necessity to learn traffic scenarios and understand patterns within each road-user class. The end goal in the analysis is to identify and learn scenarios that may contribute to hazards in traffic conditions. The classification relies on video data (movement trajectories) collected in urban intersections. The approach is based on the discrimination of the shapes of the speed profiles of each road-user type, more precisely, the discrimination between the speed movement patterns of vehicles and the ambulatory characteristics of pedestrians. The collected movement-trajectory data are represented as time series. The classification is performed using singular value decomposition and reconstruction of the time series. Two complementary methods are proposed based on the quality evaluation (correlation score) of the reconstructed trajectories. In the first method, a threshold-based decision procedure is applied. This approach is complemented in the second method by a semisupervised classification procedure guided by movement prototypes. The approach is validated on real-world data collected in Oakland, California. A correct classification of around 90% was achieved using both methods. [ABSTRACT FROM AUTHOR]

Foil and vibrating wire strain gauges have an optimum strain measurement accuracy of one microstrain. However, they can only provide discrete strain readings over a single fixed-gauge length. Digital image correlation (DIC) offers an alternative to conventional strain gauges because a two-dimensional (2D) surface strain field can be obtained from a single sensor (camera). However, the benefits of 2D strain measurements are only worthwhile if a similar level of measurement accuracy to conventional strain gauges can be achieved. This paper presents the results of an investigation into the optimum strain measurement accuracy that can be achieved by using the 2D technique on artificial images (which eliminate errors associated with cameras and lighting). The principle of the 2D DIC technique and its historical development will be introduced. Then, three potential techniques for taking strain measurements will be presented and compared: single readings, averaged linear readings, and an approach on the basis of Mohr's circle. The Mohr's circle approach was found to be the most accurate and was not susceptible to image misalignment. Strain measurement accuracy was also found to be affected by the bias error of the subpixel interpolation scheme, but the use of an 8 coefficient B-spline was found to produce satisfactory results within the error of conventional strain gauges. Gauge length was also found to have a significant effect on strain measurement accuracy, indicating that measuring strains in a material in which there are variations across the strain field could result in a loss of measurement accuracy. However, overall it was found that 2D DIC offers the same strain measurement accuracy as conventional strain gauges when used under ideal conditions. [ABSTRACT FROM AUTHOR]

Earthmoving plants are essential but costly resources in the construction of heavy civil engineering projects. In addition to proper allocation, ongoing control of this equipment is necessary to ensure and increase the productivity of earthmoving operations. Captured videos from construction sites are potential tools to control earthmoving operations; however, the current practice of manual data extraction from surveillance videos is tedious, costly, and error prone. Cutting-edge computer vision techniques have the potential to automate equipment monitoring tasks. This paper presents research in the evaluation of combinations of existing object recognition and background subtraction algorithms to recognize off-highway dump trucks in noisy video streams containing other active machines. Two detection algorithms, namely, Haar-histogram of oriented gradients (HOG) and Blob-HOG, are presented and evaluated for their ability to recognize dump trucks in videos as measured by both effectiveness and timeliness. The results of this study can help practitioners select a suitable approach to recognize such equipment in videos for real-time applications such as productivity measurement, performance control, and proactive work-zone safety. [ABSTRACT FROM AUTHOR]

The automated detection of structural elements (e.g., columns and beams) from visual data can be used to facilitate many construction and maintenance applications. The research in this area is under initial investigation. The existing methods solely rely on color and texture information, which makes them unable to identify each structural element if these elements connect each other and are made of the same material. The paper presents a novel method of automated concrete column detection from visual data. The method overcomes the limitation by combining columns' boundary information with their color and texture cues. It starts from recognizing long vertical lines in an image/video frame through edge detection and Hough transform. The bounding rectangle for each pair of lines is then constructed. When the rectangle resembles the shape of a column and the color and texture contained in the pair of lines are matched with one of the concrete samples in knowledge base, a concrete column surface is assumed to be located. This way, one concrete column in images/videos is detected. The method was tested using real images/videos. The results are compared with the manual detection ones to indicate the method's validity. [ABSTRACT FROM AUTHOR]

This paper presents a new approach that allows automated recognition of three-dimensional (3D) computer-aided design (CAD) objects from 3D site laser scans. This approach provides a robust and efficient means to recognize objects in a scene by integrating planning technologies, such as multidimensional CAD modeling, and field technologies, such as 3D laser scanning. Using such an approach, it would be possible to visualize the 3D status of a project and automate some tasks related to project control. These tasks include 3D progress tracking, productivity tracking, and construction dimensional quality assessment and quality control. This paper provides an overview of the developed approach and demonstrates its performance in object recognition and project 3D status visualization, with data collected from a construction job site. [ABSTRACT FROM AUTHOR]

Based on the solid modeling, transformation operation and Boolean operation techniques widely available in computer-aided design and drafting system, first, this paper proposed a general modeling method for dynamic underground structure simulation. Then, representations of underground structures, ground, and related information in three-dimensional space were investigated. Some suggestions were put forwarded to improve visualization results, including object organization by layers, thematic viewing, level of detail for object representation, and spatial cueing for useful information. Application examples were given to illustrate the visualization effects. Finally, virtual reality technology was employed in the simulation of the underground structures. By positioning objects using real-time information, picking objects, and then accessing their real world working status through the Internet, the realistic results of a virtual system were further discussed. It has also been found that in order to obtain satisfactory visual results using an inexpensive desktop computer, a virtual reality model must be optimized from the perspective of reducing object complexity as well as efficient scene management. [ABSTRACT FROM AUTHOR]

Calibration is an essential part of the application of conceptual rainfall-streamflow models to watershed management problems for civil engineers. However, the identification of a unique set of parameters is difficult, if not impossible, for commonly used models. Most multiobjective methods and uncertainty assessment tools require substantial numbers of function evaluations and limit the intervention of experienced modelers in the calibration process. This paper demonstrates the application of an efficient user-driven calibration-support system to conceptual rainfall-streamflow models. The system is designed to assist the hydrological modeler by means of rapid sampling of solutions, clustering, and visualization, together with interactivity to exploit the expertise of the user and/or the knowledge revealed by the clustering technique. The efficiency of the multiobjective calibration is enhanced through the use of a novel objective function based on hydrograph slope. The application of the system to the calibration of the SIXPAR conceptual rainfall-streamflow model using a synthetic time series is shown to be effective. [ABSTRACT FROM AUTHOR]

Digital image correlation (DIC) is used in this paper to study two-dimensional spatial soil deformations nonintrusively. Adaptive cross correlation (ACC), which is an advanced cross-correlation algorithm that utilizes variable window sizing and window shifting methods, was used to reduce the errors associated with conventional DIC. Comparison of the two algorithms using a scheme of predefined digital and physical movements shows that ACC improves the accuracy and range of DIC. A model of a strip footing on sand is presented. A digital camera was used to capture consecutive images of soil deformations through a Plexiglas mold. The spatial deformation of the soil beneath the footing was obtained by correlating consecutive images using ACC. Shear strains and the failure surface were also calculated based on the displacements measured using ACC. The optical setup and digital image processing technique presented in this paper were proven to be an effective tool for studying soil–structure interaction nonintrusively. [ABSTRACT FROM AUTHOR]

In this paper, we examine the potential of a range camera in construction material sorting applications. By utilizing the depth information from a time-of-flight (TOF) camera, we identify and classify objects placed on a conveyor belt. The motivation stems from the fact that object manipulation and identification are two primary problems associated with automated handling and sorting of material objects moving on a conveyor belt. Such a problem may arise in the construction industry for handling objects such as formwork when it is returned to the distribution facility. We present algorithms and experimental results using the depth information from a TOF range camera for two distinct tasks: (1) object attribute measurement and (2) object identification. Experiments and results for both tasks indicate a potential benefit for the use of TOF range cameras in construction automation and robotic arm tasks as they relate, for example, to sorting applications in the construction-formwork business. [ABSTRACT FROM AUTHOR]

It is challenging to recognize dynamic three-dimensional (3D) objects in a scattered construction work environment because it is difficult to process large amounts of incomplete and complex site data from optical sensors in real time or near-real time. This paper proposes a model-based automatic dynamic object recognition and registration method, projection-recognition-projection (PRP), to help heavy equipment operators rapidly perceive the 3D working environment at construction sites. In this study, machine vision and laser scanning technologies were used to rapidly recognize and register dynamic target objects in 3D space. The proposed PRP method extracts a target object's point cloud data from the background scene to significantly reduce scan data size and modeling time. The target objects' point clouds are then processed to rapidly recognize and register 3D target models from a model database to the corresponding point clouds. Several field tests at construction equipment operation sites show that the proposed PRP framework is promising and can effectively assist heavy construction equipment operations and automated equipment control by rapidly modeling dynamic target objects in 3D views. [ABSTRACT FROM AUTHOR]

Weathering steel with a natural corrosion-resistant feature has been widely applied to the structural components of steel bridges. However, severe surface corrosion damage has been frequently observed in the weathering steels of bridges, which causes the performance degradation of the structure. Conventional visual classification approaches are time-consuming and subjective and cannot provide quantitative evaluation effectively and efficiently. This paper presents a new imaging-based intelligent method for quantitatively rating the corrosion states of weathering steel bridges. Images are characterized by image texture analysis using two-dimensional wavelet decomposition, from which both the local and global energy distributions of each detail subimage are extracted as representative features. To enhance the performance of a support vector machine (SVM) in corrosion state classification, a particle swarm optimization algorithm (PSO) is developed to obtain the optimal parameters of the SVM. A comparative study indicates that PSO-SVM can achieve better classification accuracy rates than artificial neural network. Numerical results demonstrate that this study provides an effective approach to imaging-based rating by integrating wavelet transform and PSO-SVM techniques for allocating the condition state of corroded weathering steel. [ABSTRACT FROM AUTHOR]

Industry shifts to the use of digital tools have created a need for new means and methods for approaching team interaction and planning for collaborative tasks, in particular, the shift from typical and understood interactions based on paper media to dynamic multimedia when using virtual prototypes in alignment with advanced visualization, analysis, and simulation tools. The INVOLVE framework was developed to respond to the need for planning for collaborative tasks focused on the architecture, engineering, and construction industry when using virtual prototypes. This study presents the elements of the INVOLVE framework, then demonstrates the validation in team interaction when different media configurations are used for two different collaborative tasks, a schematic site plan development, and a four-dimensional schedule review. The team interactions were implemented through a series of quasi-experiments within a collaborative interactive workspace. Observational methods were used to quantify the timing and duration of different team and media interactions. The findings show that changes in media traits have a measurable effect on a team's communication and can shift the performance of certain tasks between a cooperative process and a collaborative one. The findings suggest INVOLVE provides an effective framework for determining and planning the needed infrastructure to improve engagement and communication to support collaborative tasks. [ABSTRACT FROM AUTHOR]

Visual monitoring of construction worksites through the installation of surveillance cameras has become prevalent in the construction industry. These cameras also are useful for automatic observation of construction events and activities. This paper demonstrates the use of a surveillance camera for assessing tower crane activities during the course of a workday. In particular, it seeks to demonstrate that the crane jib trajectory, together with known information regarding the site plans, provides sufficient information to infer the activity states of the crane. The jib angle trajectory is tracked by using two-dimensional to three-dimensional rigid pose tracking algorithms. The site plan information includes a process model for the activities and site layout information. A probabilistic graph model for crane activity is designed to process the track signals and recognize crane activity as belonging to one of two categories: concrete pouring and nonconcrete material movement. The experimental results from a construction surveillance camera show that crane activities are correctly identified. [ABSTRACT FROM AUTHOR]

The negative impacts of urban traffic growth are well known, e.g., congestion, increased air pollution, and more traffic noise. Humans can readily see traffic congestion, and can hear traffic noise, but inevitably they are much less aware of odorless, invisible, silent air pollution. Therefore, quite naturally people, whether the general public or transport professionals, struggle to visualize air pollution from traffic, and will find it particularly difficult to visualize changes in air pollution levels resulting from changes in traffic conditions because of the implementation of various urban transport schemes. Hence, there is always a risk of the air quality impacts of potential urban transport schemes being underappreciated. Therefore, this paper contributes to the knowledge by devising a new three-dimensional (3D) visualization approach for modeled air quality before and after the implementation of potential urban transport schemes. Using the Dunkirk area of the City of Nottingham in the United Kingdom as a case study, research has been undertaken to integrate an air pollution dispersion model for the pollutant with a 3D digital city model. The modeled concentrations, before and after the implementation of an urban transport scheme, were represented in the 3D city model at various heights above the ground: first, as 3D point shapes; second, as 3D planar surfaces; and finally, as 3D volumetric clouds. The 3D volumetric clouds approach used the analogy of people's perceptions of gray clouds in the sky as representing undesirable weather, and hence provided an intuitive 3D visualization of all the modeled concentrations, at and above the ground surface, in a single 3D virtual scene. Benefits have been identified in enhancing the level of understanding of the pollution dispersion using this new approach to visualization. This has allowed the visualization process to be used in the development of future traffic scenarios that could be used to alter the design of a proposed transport scheme to increase its air quality benefits. This 3D visualization approach was found to be applicable to other transport schemes in different parts of the City of Nottingham. [ABSTRACT FROM AUTHOR]

Image processing is an effective tool for automated monitoring of construction projects. For the last decade, it has gained increasing acceptance in its application to progress monitoring, productivity analyses, and quality assurance. However, a notable downside exists in image processing, especially in outdoor applications such as construction project monitoring: image quality is heavily affected by ambient lighting conditions. Poor or undesirable lighting conditions produce substandard quality images, which generally lead to a high level of errors in the related image processing for information extraction. This paper presents error-correction methods that can improve the image processing results for construction progress monitoring in the postprocessing stage. The methods are applied in the postprocessing stage. The key idea behind the error-correction methods is the concept of priority to classify input images and project information into several categories based on data reliability and intelligently use the classified information for more accurate analyses of the project progress. Tests in real construction sites showed that these postprocessing methods significantly increased the accuracy of image processing-based construction progress monitoring. [ABSTRACT FROM AUTHOR]

Terrestrial laser scanning (TLS) provides a rapid, remote sensing technique to model 3D objects. Previous work applying TLS to structural analysis has demonstrated its effectiveness in capturing simple beam deflections and modeling existing structures. This paper extends TLS to the application of damage detection and volumetric change analysis for a full-scale structural test specimen. Importantly, it provides a framework necessary for such applications, in combination with an analysis approach that does not require tedious development of complex surfaces. Intuitive slicing analysis methods are presented, which can be automated for rapid generation of results. In comparison with conventional photographic and surface analysis methods, the proposed approach proved consistent. Furthermore, the TLS data provided additional insight into geometric change not apparent using conventional methods. As with any digital record, a key benefit to the proposed approach is the resulting virtual test specimen, which is available for posttest analysis long after the original specimen is demolished. Uncertainties that can be introduced from large TLS data sets, mixed pixels and parallax in the TLS analysis are also discussed. [ABSTRACT FROM AUTHOR]

The interactive multiobjective genetic algorithm (IMOGA) is a promising new approach to calibrate models. The IMOGA combines traditional optimization with an interactive framework, thus allowing both quantitative calibration criteria as well as the subjective knowledge of experts to drive the search for model parameters. One of the major challenges in using such interactive systems is the burden they impose on the experts that interact with the system. This paper proposes the use of a novel image-based machine-learning (IBML) approach to reduce the number of user interactions required to identify promising calibration solutions involving spatially distributed parameter fields (e.g., hydraulic conductivity parameters in a groundwater model). The first step in the IBML approach involves selecting a few highly representative solutions for expert ranking. The selection is performed using unsupervised clustering approaches from the field of image processing, which group potential parameter fields based on their spatial similarities. The expert then ranks these representative solutions, after which a machine-learning model (augmented with the spatial information of the selected fields) is trained to learn user preferences and predict rankings for solutions not ranked by the expert. To better mimic the “visual” information processing of human experts, algorithms from the field of image processing are used to mine information about the spatial characteristics of parameter fields, thus improving the performance of the clustering and machine-learning algorithms. The IBML approach is tested and demonstrated on a groundwater calibration problem and is shown to lead to significant improvements, reducing the amount of user interaction by as much as half without compromising the solution quality of the IMOGA. [ABSTRACT FROM AUTHOR]

Videotaping is an effective and inexpensive technique that has long been used in construction to conduct productivity analyzes. However, as schedules of modern construction projects become more and more compressed, the limitation of video-based analysis—intensive manual reviewing process—contrasts sharply with the need for effortless data analysis methods. This paper presents a study on developing a video interpretation model to interpret videos of construction operations automatically into productivity information. More specifically, this research formalizes key concepts and procedures of video interpretation within the construction domain. It focuses on designing a mechanism for furthering the crosstalk between the prior knowledge of construction operations and computer vision techniques. It uses this mechanism to guide the detection and tracking of project resources as well as work state classifications and abnormal production scenario identifications. The resulting approach has the potential to provide a common base for developing automated video interpretation procedures that can greatly improve current data collection and analyzes practices in construction. Experimental results from preliminary studies have shown the potential of the proposed video interpretation method as an improved productivity data analysis method. [ABSTRACT FROM AUTHOR]

Construction-site images that are now easily obtained from digital cameras have the potential to automatically provide the project status information. For example, once construction objects such as concrete columns are accurately identified and counted, the current level of project progress in the column installation activity can easily be measured. However, in order to identify and count the number of concrete columns installed at a particular point of time, a robust object recognition methodology is required. Without the successful recognition and extraction of the construction object of interest, it is almost impossible to understand the current level of project progress. This paper presents a robust image processing methodology to effectively extract the objects of interest from construction-site digital images. The proposed methodology makes use of advanced imaging algorithms and a three-dimensional computer aided design perspective view to increase the accuracy of the object recognition. Tests show that the methodology is promising and expected to provide a solid base for the successful, automatic acquisition of project information. [ABSTRACT FROM AUTHOR]

Dimensional analysis from single images is still a fruitful topic for investigation, even if range sensors and stereo imaging technologies are becoming a common alternative to the generation of metric information. This paper describes the scientific approach developed for the dimensional analysis of bridges based on a single image. The basic dimensions of bridges are provided using a low-cost and flexible alternative that breaks with the classical stereoscopic photogrammetric principle. Several case studies are reported to demonstrate the novelty and real capabilities of this methodology. [ABSTRACT FROM AUTHOR]

The ability to effectively communicate progress information and represent as-built and as-planned progress discrepancies are identified as key components for successful project management that allow corrective decisions to be made in a timely manner. However, current formats of reporting (e.g., textual progress reports, progress curves, and photographs) may not properly and quickly communicate project progress. Current monitoring methods also require manual data collection and extensive data extraction from different construction documents, which distract managers from the important task of decision making. Therefore, to facilitate progress monitoring, this paper proposes visualization of performance metrics that aims to represent progress deviations through superimposition of four-dimensional (4D) as-planned model over time-lapsed photographs in single and comprehensive visual imagery. As a part of the developed system, registration of the 4D model with photographs, augmenting photographs, and occlusion removal for progress images are presented. While contextual information is preserved, the as-built photographs are enhanced and augmented with 4D as-planned model in which the performance metrics are visualized. The augmented photographs provide a consistent platform for representing as-planned, as-built, and progress discrepancies information and facilitate communication and reporting processes. [ABSTRACT FROM AUTHOR]

The hundreds of traffic sign types on the road and their various shapes and colors make it difficult to develop a generalized method of traffic sign detection. Consequently, agencies performing a sign inventory must manually review millions of roadway video log images. This paper proposes an innovative image processing model that automatically detects traffic signs and dramatically reduces the sign inventory workload. In a test of the proposed model using 37,640 images provided by the Louisiana Department of Transportation and Development, 86 percent of the manual review efforts can be effectively saved. Our method is composed of (1) a generalized traffic sign model to represent the entire class of traffic signs; (2) a proposed new statistical traffic sign color model; (3) a traffic sign region of interest detection system using polygon approximation; and (4) traffic sign candidate decision rules based on shape and color distributions. [ABSTRACT FROM AUTHOR]

Described in this paper is a novel approach to four-dimensional (4D) computer-aided design (CAD). It involves a two-way symbiotic relationship between three-dimensional (3D) CAD software and a software implementation of linear planning that includes the ability to define a project product model and associate it with the process model. Strengths of the approach include the ability to readily modify construction sequences and examine their consequences using 4D CAD, and the ability to treat very large scale projects marked by significant repetition of their components. By building on a shared image of the project product model from both a design and construction perspective, the CAD model can be structured in a way that facilitates communication with the scheduling software and vice versa. Various challenges involved in making the 2-way process work are described, including consistency of product representation in the CAD and scheduling models, and the need to group CAD components at different levels of detail and locations to reflect the kinds of aggregation found in schedule representations of a project. The benefits of the approach include the ease with which different scheduling strategies can be explored and visualized, the links between 3D objects and activities can be maintained, and the completeness of the product model representations can be validated. A case study is used to illustrate the approach adopted and the challenges involved. [ABSTRACT FROM AUTHOR]

Accurate analyses of equipment idle time are crucial for the efficient utilization of construction equipment in large construction projects. The less idle time the equipment has, the higher productivity it can achieve. However, it is not feasible for field personnel to visually observe the operation of construction equipment all day. An image processing-based methodology is presented in this paper to automatically quantify the idle time of hydraulic excavators. The image color space (hue, saturation, and value), which shows significant advantages over the red, green, and blue color space in identifying and tracing hydraulic excavators, is used as the base for image segmentation and tracing algorithms. The changing centroid coordinates of an excavator in successive images taken at constant time intervals are used as indicators of movement. Experimental results show that the presented methodology has a promising application potential for effective equipment management in construction projects. [ABSTRACT FROM AUTHOR]

Optimum detailing of reinforcement bars for biaxial columns is a nontrivial task and there is no guidance in the codes of practice on this issue. Conventional methods of producing a single design solution and the use of optimization methods to find a single global optimum solution is not what practicing engineers require. This paper introduces an innovative approach to design, which emphasizes the interactive use of evolutionary computing in design and a new approach to visualization of search and solution spaces. This replaces the “black box” nature of the optimization techniques with an interactive approach in which the designer is in full control of the process. Instead of locating a single solution, clusters of good designs are identified in color, allowing the designer to conduct a more focused search in the areas of high performance design solutions. This method enables the designer to view individual solutions in the vicinity of the local peaks in order to learn more about the interrelationship between design parameters. Interactive visualization tools enable designers to develop a deeper understanding of the search and solution spaces and assist them in evaluating the merits of alternative designs and in their decision making. A by-product of visualization is knowledge discovery, which allows the designer to discover novel solutions that would not have been possible by conventional means. This approach could be easily extended to any stage of multidisciplinary design, especially to the conceptual stage of the design process. For example, the position of bars within a column is similar to the position of columns in a floor layout optimization problem. [ABSTRACT FROM AUTHOR]

The technological advancements in digital imaging, the widespread popularity of digital cameras, and the increasing demand by owners and contractors for detailed and complete site photograph logs have triggered an ever-increasing growth in the rate of construction image data collection, with thousands of images being stored for each project. However, the sheer volume of images and the difficulties in accurately and manually indexing them have generated a pressing need for methods that can index and retrieve images with minimal or no user intervention. This paper reports recent developments from research efforts in the indexing and retrieval of construction site images in architecture, engineering, construction, and facilities management image database systems. The limitations and benefits of the existing methodologies will be presented, as well as an explanation of the reasons for the development of a novel image retrieval approach that not only can recognize construction materials within the image content in order to index images, but also can be compatible with existing retrieval methods, enabling enhanced results. [ABSTRACT FROM AUTHOR]

Efficient data collection of high-quantity and low-cost highway assets such as road signs, traffic signals, light poles, and guardrails is a critical element to the operation, maintenance, and preservation of transportation infrastructure systems. Despite its importance, current practice of highway asset data collection is time-consuming, subjective, and potentially unsafe. The high volume of the data that needs to be collected can also negatively impact the quality of the analysis. To address these limitations, this paper proposes a new algorithm for semantic segmentation and recognition of highway assets using video frames collected from a car-mounted camera. The proposed set of algorithms (1) takes the captured frames and using a pipeline of structure from motion and multiview stereo reconstructs a three-dimensional (3D) point cloud model of the highway and surrounding assets; (2) using a Semantic Texton Forest classifier, each geo-registered two-dimensional (2D) video frame at the pixel-level is segmented based on shape, texture, and color of the highway assets; and finally, (3) based on the results of the 2D segmentation and a new voting scheme, each reconstructed 3D point in the cloud is also categorized for one type of asset and is color coded accordingly. The resulting augmented reality environment that integrates the color-coded point clouds with the geo-registered video frames enables a user to conduct visual walk through and query different categories of assets. Experiments were performed on a challenging video data set containing sequences filmed from a moving car on a 2.2-mi-long, two-lane highway research facility. Experimental results with an average accuracy of 76.50 and 86.75% in segmentation and pixel-level recognition of 12 types of asset categories reflect the promise of the applicability of this approach for segmentation and recognition of highway assets from image-based 3D point clouds. It also enables future algorithmic developments for 3D localization of traffic signs and other assets that are detected using the state-of-the-art vision-based methods. [ABSTRACT FROM AUTHOR]

This paper discusses the development of a general framework and software system to support automated analysis of sewer inspection closed-circuit television (CCTV) videos. The proposed system aims primarily to support the off-site review and quality control process of the videos and to enable efficient reevaluation of archived CCTV videos to extract historical sewer condition data. Automated analysis of sewer CCTV videos poses several challenges including the nonuniformity of camera motion and illumination conditions inside the sewer. The paper presents a novel algorithm for optical flow-based camera motion tracking to automatically identify, locate, and extract a limited set of video segments, called regions of interest (ROI), that likely include defects, thus reducing the time and computational requirements needed for video processing. The proposed algorithm attempts to recover the operator actions during the inspection session, which would enable determining the location and relative severity of the ROI. To ensure proper segmentation and defect detection, frames within the ROI are classified on the basis of the camera orientation using a set of Haar-like features and a multiclass support vector machine. A segmentation algorithm based on gray-level intensity analysis is also presented. Algorithms for automated detection of debris and joint displacement defects are also discussed. The debris detection algorithm employs image segmentation and texture analysis techniques to locate and verify debris objects inside the water flow lines. The joint displacement algorithm performs gray-level intensity analysis to detect joints offset. The proposed system was successfully applied to analyze a set of CCTV videos obtained from the cities of Regina and Calgary in Canada. The results were validated against actual inspection reports prepared by CCTV operators, which demonstrated the viability and robustness of the proposed algorithms. [ABSTRACT FROM AUTHOR]

Current practice utilizes two-dimensional (2D) drawings as the main visualization means to guide assembly. As an emerging technology, augmented reality (AR) integrates three-dimensional (3D) images of virtual objects into a real-world workspace. The insertion of digitalized information into the real-world workspace using AR can provide workers with the means to implement correct assembly procedures with improved accuracy and reduced errors. The limited available research concerning the applications of AR visualization means in assembly highlights the need for a structured methodology of addressing cognitive and usability issues for the application potentials of AR technology to be fully realized. Thus, the focus of the study presented in this paper is to investigate how much improvement in assembly productivity and performance can be achieved by lowering cognitive workload via AR. The AR system was developed in collaboration with Woodside Energy Ltd. following research project ECHO. Evaluation of the prototyped construction piping scenario revealed that the AR visualization yielded shorter task completion time (50% reduction), fewer assembly errors (50% reduction), and lower total task load. More specific findings also indicated that the animated AR visualization significantly shortened the original time (55% reduction) and rework time (46% reduction), payment to assemblers (50% reduction), and cost of correcting erroneous assembly (66% reduction). Questionnaire feedback revealed that the animated AR visualization better aided assembly comprehension and better facilitated information retrieval and collaboration between human and guidance medium. From a technology innovation perspective, by providing a higher level of immersion and interaction with spatial awareness, the dynamic AR guidance can be integrated with building information model (BIM) technology in helping and training practitioners, because all of the models and assembly sequences are designed in BIM and visualized in AR. [ABSTRACT FROM AUTHOR]

The objective of this study is to visualize and simulate microscale properties of asphalt concrete with three-dimensional discrete element models under mechanical loading. The microstructure of the asphalt concrete sample was composed of three ingredients: coarse aggregates, sand mastic (a combination of fines, fine aggregates, and asphalt binder), and air voids. Coarse aggregates were represented with the irregular polyhedron particles which were randomly created with an algorithm developed for this study. The gaps among the irregular particles were filled with air voids and discrete elements of sand mastic. The mechanical behaviors of the three ingredients were simulated with specific constitutive models at different contacts of discrete elements. Based on the geometric and mechanical models, visualization and simulation of asphalt mixtures were conducted in this study. [ABSTRACT FROM AUTHOR]

Control systems have proven to be beneficial in improving the productivity of earthmoving operations. A main component of these systems is the monitoring module. Computer vision algorithms are among the new methods that have been tested for real-time monitoring of earthwork activities. These methods, however, were able to detect only intraclass equipment and failed to identify individual machines, which is a key disadvantage compared to radio-based devices, namely global positioning systems (GPS). To address this issue, a pipeline framework, consisting of several computer vision algorithms, has been developed to identify individual machines. In this framework, an object detection method is used to locate construction equipment. If a detection view of a target is obtained, the camera zooms on the candidate to identify visual markers attached on the machine. The architecture of this system is optimized by employing time-consuming processes only for the most probable candidates. This system was evaluated using several real-time videos, and demonstrated promising performance in identifying excavators and dump trucks, with 89 and 84% identification rates and 64.6 and 77.1% recall rates, respectively. In addition, applying the marker-based verification step proved to be effective in rejecting false alarms as the precision was 100% in both test cases. [ABSTRACT FROM AUTHOR]

Movable bridges are unique structures due to the complex interaction between their structural, mechanical, and electrical systems with an intricate interrelation creating several challenges related to operation and maintenance. Continuous monitoring of the critical parts of these structures is essential to track and evaluate their performance for improving maintenance operations and reducing the associated costs. Open gears are one of the most critical components of movable bridges. Proper and regular maintenance of these gears is vitally important to ensure a safe, reliable, and cost-effective operation. In this study, a practical and low-cost monitoring approach is presented to track the lubrication level in an open gear of a movable bridge by using video cameras. Two unique indices are developed for monitoring of the open gear by investigating two different image processing methods in a comparative fashion. The first methodology is based on an edge detection algorithm that utilizes a Sobel gradient operator to determine the edges in the open gear image. A lubrication index (LI) based on the edge detection results is defined and extracted to determine the lubrication level. The second methodology employs a fuzzy neural network-based approach to define a lubrication anomaly parameter (LAP) for assessing the lubrication level. The analysis results from the real-life application show that both methodologies successfully identify the lubrication level of the movable bridge's open gear. [ABSTRACT FROM AUTHOR]

In the reported study the viability of using computed tomography (CT) scanning for assessing flushing defects in thin sprayed seal (chipseal) surfacings was explored. The study was undertaken to investigate the micromechanical interactions that occur within chipseal layer materials in order to examine their relationship to the origination of flushing, using CT scanning techniques. In particular, the effectiveness of using image analysis techniques to analyze the changes in air voids that occur within a chipseal layer during loading was investigated. The presented study was based on laboratory testing of chipseal pavement samples (cores) from in-service pavements in the Auckland and Waikato regions of New Zealand. The cores, with diameters of 200 mm and thicknesses ranging from 32.4 mm to 44.5 mm, were subjected to varying levels of lateral cyclic loading using a wheel-tracking machine, and the deformation that occurred on the surface of the cores was measured. Two small specimens were extracted from each loaded core, one specimen from the wheel-tracked area of the core and the other from the untracked area of the core. The specimens were scanned using a CT scanner, and the resulting scan images were analyzed using image analysis techniques to determine the distribution of air voids within each specimen. The air voids within the tracked and untracked specimens of each core were compared to examine the changes that had occurred to the distribution of air voids during loading. The results of the study showed that image analysis is an effective tool to analyze air voids within a chipseal layer. [ABSTRACT FROM AUTHOR]

The primary challenge in generating convincing augmented reality (AR) graphics is to project three-dimensional (3D) models onto a user's view of the real world and create a temporal and spatial sustained illusion that the virtual and real objects coexist. Regardless of the spatial relationship between the real and virtual objects, traditional AR graphical engines break the illusion of coexistence by displaying the real world merely as a background and superimposing virtual objects on the foreground. This research proposes a robust depth-sensing and frame buffer algorithm for handling occlusion problems in ubiquitous AR applications. A high-accuracy time-of-flight (TOF) camera is used to capture the depth map of the real world in real time. The distance information is processed in parallel using the OpenGL shading language (GLSL) and render to texture (RTT) techniques. The final processing results are written to the graphics frame buffers, allowing accurate depth resolution and hidden surface removal in composite AR scenes. The designed algorithm is validated in several indoor and outdoor experiments using the scalable and modular augmented reality template (SMART) AR framework. [ABSTRACT FROM AUTHOR]

Conducting a pavement survey is a time-consuming but necessary task to ensure the serviceability of road pavements. Many investigators have used image-processing methods to automate the survey process and to enhance the quality and accuracy of survey results. However, image-processing methods often mistakenly treat oil spillages, shadows, and road markings as distresses because their features are similar to those of distresses. Therefore, in this research, the authors proposed a dual-light inspection (DLI) method to reduce false alarms. The DLI involves four major steps: (1) image capture-a pair of images is retrieved from identical positions and orientations, but with different light setups; (2) image subtraction-the two images are subtracted pixel by pixel to obtain a subtracted image that represents the differences between the paring images; (3) image enhancement-an edge detection method is applied to retrieve the distress features; and (4) image classification-a classification algorithm is finally used to discriminate between images that include distresses and ones that do not. A field test was conducted to verify the DLI method. A total of 212 pairs of images were captured during nighttime, including images of alligator cracks (42 pairs), manholes (42 pairs), longitudinal cracks (58 pairs), spillages (34 pairs), and road markings (52 pairs). Twenty percent of the images (i.e., 45 pairs) were used as training sets to train the classification model. The remaining images were then used to test the accuracy of the classification model. The accuracy of the DLI method, which uses dual-light image pairs, was compared with that of the traditional method, which uses individual images. The DLI can significantly improve the accuracy in determining spillage (traditional: 18%, DLI: 82%) and road markings (traditional: 8%, DLI: 96%). The DLI is also reasonably accurate in determining other distresses including alligator cracks (traditional: 95%, DLI: 90%), manholes (traditional: 97%, DLI: 100%), and longitudinal cracks (traditional: 62%, DLI: 69%). These results indicate that DLI can become a reliable method for automatic pavement inspection. [ABSTRACT FROM AUTHOR]