Hongjo Kim

Vision-based object-centric safety assessment using fuzzy inference: Monitoring struck-by accidents with moving objects

AbstractDue to the dynamic environment of construction sites, workers are continuously confronted with the potential for safety accidents. Although various safety guidelines have been developed, workers cannot always be aware of everything that occurs around them when they focus on their work on noisy and congested job sites. Therefore, it is difficult for workers to conform to guidelines to protect themselves when confronting dangerous situations. To address this safety issue, this paper presents an on-site safety-assessment system for monitoring struck-by accidents with moving entities based on computer vision and fuzzy inference. Computer vision is used to monitor a construction site and extract spatial information for each entity (workers and equipment). Next, fuzzy inference is used to assess the proper safety levels of each entity using spatial information. A safety level represents the potential hazard or the integrating danger that a person encounters at a particular moment. Struck-by accidents ar...

On-site Safety Management Using Image Processing and Fuzzy Inference

Construction is generally conducted in highly changeable site conditions due to operation of machinery, transfer of materials, moving workers, and changing progress status. Such dynamic characteristics can always lend themselves to the potential for safety accidents. However, it is not easy to have a generalized safety management system that can universally apply to every job site. Although there are some guidelines for safety management, their application to a construction site has to be tailored to satisfy the unique nature of the particular project. Based on the good understanding of the particular site, effective and appropriate safety management process should be derived for workers to be protected from potential dangers of the site. This paper presents an on-site safety management methodology based on image processing and fuzzy inference. Image processing is used to extract spatial information of workers; fuzzy inference is then used to provide the workers a proper level of safety warning based on the spatial information. Contextual site information, such as equipment operations and density of workers, also constitute the on-site safety management methodology. The proposed methodology is expected to provide a safer working environment for construction workers, through its capability to easily customize the safety management system for a construction site.

Detecting Construction Equipment Using a Region-Based Fully Convolutional Network and Transfer Learning

AbstractFor proper construction site management and plan revisions during construction, it is necessary to understand a construction site’s status in real time. Many vision-based construction site-...

Vision-based 2D map generation for monitoring construction sites using UAV videos

This paper presents a system to generate a 2D map using unmanned aerial vehicles (UAV) videos. The system can process the collected images for lens distortion, unstable flight, and image redundancy. The system is demonstrated using an UAV video captured on road construction site in Namyangju city, Korea. The experimental result is a high-resolution 2D map of which resolution is 6010 by 9465. This study is expected to help site managers understand site conditions and make a decision timely in diverse

Algorithm for Economic Assessment of Infrastructure Adaptation to Climate Change

Climate change, along with the increase of severe weathers and natural disasters, is becoming an important factor to consider for infrastructure investments. To adapt infrastructure to the effects of climate change, new design, construction, or rehabilitation methods – so-called adaptation methods – can be deployed. However, it is crucial to understand the impact of adaptation methods on infrastructure before they are actually implemented. When the economic benefit and cost are clear, asset managers can confidently make informed decisions about the priority of investment alternatives. This paper proposes an integrated algorithm to assess the benefit and cost of adaptation methods. The “integrated” aspect of the algorithm is derived from the fact that climate change effects on infrastructure can be divided into two categories. One is sudden extreme weather events caused by climate change; this sudden event leads to swift and disruptive damages to infrastructure. The other is a gradual climate change of which effects are shown over a long period of time. The algorithm combines the two different aspects of climate change to estimate the net benefit of adaption methods in an integrated manner. Future climate scenarios are first assumed and their input variables are determined for further procedures. With extreme events such as supertyphoon, the procedure for sudden failure of infrastructure is used to estimate the cost and benefit of the rehabilitation effort. Maintenance cost under gradual climate change is also estimated with the climate change adjusted deterioration curve for the infrastructure of interest. Finally, the above three steps are repeated for each year to estimate the life cycle cost infrastructure adaptation to climate change for the comparison of the costs with and without adopting the adaptation method.

Detection of Construction Equipment Using Deep Convolutional Networks

Vision-based monitoring methods have been investigated for understanding construction site contexts. However, detection capabilities of such methods are still insufficient to be utilized in general construction sites due to dynamic outdoor conditions and appearance variances of construction entities. To improve performance of a construction entity detector, we propose a detection method using a region-based fully convolutional network (R-FCN). R-FCN consists of two main parts, which are a fully convolutional network and a region proposal network. The fully convolutional network extracts hierarchical object features through a supervised learning process, while a region proposal network generates a set of object candidate regions in an image to localize target objects. To evaluate the generalization performance of the detection method, a benchmark dataset is collected from ImageNet for five classes (dump truck, excavator, loader, concrete mixer truck, and road roller), having various object appearances within a class in different backgrounds. A state-of-the-art performance, mean average precision of 95.61%, was recorded from the experiment. The proposed method shows a potential for the universal detector that can detect construction equipment on every construction site.

Prediction of Flexible Pavement Deterioration in Relation to Climate Change Using Fuzzy Logic

AbstractAn understanding of the impacts of climate change on infrastructure is important in the context of reducing future socioeconomic losses. Previous research has introduced models based on emp...

Valuation of Adaptation Technology to Climate Change Based on Target Classification

Increase of unpredictable and severe climate events has led to development of technologies for climate change adaptation. Types of adaptation technologies for climate change vary in terms of technological characteristics and the range of their adaptation impact as well. For accurate economic assessment, this paper presents a method to value adaptation technologies based on classification of the target. In this study, the target is classified into two groups: 1) a single structure and 2) a range of area. For a single structure, adaptation technology is focused on the specific infrastructure subject to the climate change. For a range of area, the local region is the focus to consider climate events such as flood, landslide, and sea level rise. These two classes of damages are estimated in two valuation modules: one for extreme climate change and the other for gradual climate change. Climate change scenarios with and without the adaptation technology are used to determine the total value of the adaptation technology.

An interactive progress monitoring system using image processing in mobile computing environment

A timely progress monitoring is essential for the success of a large-scale construction project. It enables a site manager to properly prepare resources and make plans for the remaining part of the construction activities. Recently, mobile computing and image processing have been investigated as a means of automating progress monitoring. Mobile computing is advantageous in wireless data recording, retrieval, and transfer whereas image processing is able to analyze the site images to extract progress information. However, their potential applications in construction monitoring were rather separately studied; synergistic effect of both techniques has not yet been fully materialized. This paper presents an interactive mobile progress monitoring system to enhance the existing progress monitoring practices. The system utilizes the interactive feature of tablet computer to combine the strength of image processing and mobile computing. When a user selects an object on a construction site image in a mobile computing environment, the system provides a list of attributes for the object of interest. In this interactive environment, the user can easily match the object to the right attributes such as location, material type, and relation with other objects. This initial matching can then allow for automatic matching of other objects on the site to the right attributes of their own. The method can raise accuracy of image processing and significantly reduce the effort required to do the pre-processing for progress monitoring. A cable-stayed bridge construction project, a case study, is used to validate the proposed system.

BIM-based mobile system for facility management

Technological advances in mobile systems, such as smartphones and tablet computers, provide engineers with new opportunities to enhance the current facility management processes. Equipped with powerful system processors, touch screens, and wireless communication capabilities, mobile systems enable the experience of omnipresence of facility information. Building information modeling (BIM) also enables improvements in facility management. With an advanced data structure and visualization capacity, BIM can be used in numerous applications for facility management. This study aims to incorporate the strength of mobile systems and BIM in facility management. The proposed system, which utilizes smartphone technology, allows engineers to conduct facility management functions, such as maintenance planning, inspection, and assessment, irrespective of a user’s location. The integration of building information modeling in a system provides detailed visual information of the building components. A case study was conducted to verify the applicability of the mobile system in terms of the hardware functions, system design, and user interface. This proposed system, with its study results, provides a guideline as to how mobile systems can be effectively integrated with BIM for improved management of facility information.