Jin Gang Lee

Dr. Jin Gang Lee


Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2022 in Journal of Computing in Civil Engineering
Verified
Journal Article
Closed access
PDF file
JLJin Gang Lee
Jin Gang Lee
JHJeongbin Hwang
Jeongbin Hwang
SCSeokho Chi
Seokho Chi
JSJoonOh Seo
JoonOh Seo
This paper presents a systematic method to create universally applicable synthetic training image datasets for computer vision-based construction object detection. The synthetic images created by inserting a virtual object of interest into a real site image allows us to minimize the time and effort for training image data collection and annotation. In addition, the use of synthetic images has an additional benefit that training images can be easily customized for a target construction site by considering the context of the site (e.g., different background scenes, camera positions, and angles) and the possible variability of target objects to be detected (e.g., different sizes, locations, rotation angles, and postures) on images. An automated approach proposed in this study attempts to systematically create the synthetic images using the Unity game engine in which context- and variability-related parameters can be controlled. The proposed method was validated by training a deep learning-based object detection algorithm [i.e., a faster regions with convolutional neural network (R-CNN) model] with synthetic images and testing it on real images from earthwork construction sites to detect an excavator. The CNN models trained with synthetic images showed an average precision value of more than 90%; in particular, the classifier using synthetic images outperformed the one using real site images. The detection results also demonstrated an improved capability to capture the high irregularity of a construction object on images when using techniques of context customization and variability randomization. The findings from this study demonstrate the feasibility and practicality of the use of synthetic images for vision-based approaches in a construction domain. Ultimately, the proposed approach serves as an alternative way to build comprehensive image datasets for construction entities, contributing to facilitating vision-based studies on construction.

3 August 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
JLJin Gang Lee
Jin Gang Lee
JSJoonOh Seo
JoonOh Seo
AAAli Abbas
Ali Abbas
MCMinji Choi
Minji Choi
To effectively use augmented reality (AR) technology for end-user involved design collaboration, it is necessary to ensure the effectiveness of the AR system from the end-user’s perspective. However, most efforts have mainly focused on technological development, and as such, limited attention has been paid to the end-user’s application of the AR system. Therefore, this study investigates how the AR system affects architectural design review based on the user’s perspectives. Three different display systems presenting a 3D model including a 2D screen, VR, and AR were tested, and a total of 76 participants evaluated visual presentation quality, perceived acceptability, and user experience according to their usage of the visualization platform types during the design review activities. Compared to other systems, the results indicated that the AR system could be more effective in reviewing the visual elements of a building. Furthermore, AR showed the highest ratings for acceptance level and user experience. The innovation provided by AR created a positive user experience, despite its remaining challenges to be resolved in terms of functionality. Since it is expected that the use of AR can be promoted by overcoming certain technological limitations, this study contributes to guiding AR system applications for end-users involved in the design review process.

26 August 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
YJYixuan Jin
Yixuan Jin
JSJoonOh Seo
JoonOh Seo
JLJin Gang Lee
Jin Gang Lee
JAJun Ahn
Jun Ahn
SHSangUk Han
SangUk Han
Three-dimensional (3D) visualization technology, such as augmented reality (AR), has served as the display for building information modeling (BIM)-based architectural design collaboration to provide more effective design observation and communication for stakeholders. That said, AR has several technical limitations in terms of personal device issues, user experience, and visualization quality. A new form of AR called spatial augmented reality (SAR) has been introduced to address these issues, which uses a digital projector to present graphics on physical objects for augmenting real-world objects. Therefore, SAR has great benefits and potentials to combine with BIM for design collaboration. This paper introduces a BIM-based SAR operational framework, where 3D building models generated from BIM software are imported to projection mapping tools to display building surface textures on physical white building models. A case study using Revit and 3ds Max as the BIM software, and MadMapper as the projection mapping tool, was conducted to demonstrate the feasibility of the proposed framework and to evaluate the projection performance of SAR. The case study showed that the texture of BIM models could be projected on the objects clearly and realistically. Additionally, the proposed SAR method potentially offers intuitive observation of building models and comfortable wear-free experience for collaborative design, and the qualitative analysis by changing the parameters was conducted to test the different projection conditions. Since it is expected that the use of SAR can be promoted by overcoming the discussed technical limitations and possible solution application, this study aims to traceability provide the whole process of BIM-based SAR for architectural design collaboration.

1 November 2022 in Sustainability
Verified
Journal Article
Open access
PDF file
DSDharmaraj Sivasubramanian
Dharmaraj Sivasubramanian
JLJin Gang Lee
Jin Gang Lee
In the construction industry, decision-making plays an important role in achieving sustainability. Decision-making processes have significantly expanded to consider detailed product information when selecting sustainable alternatives in order to make more accurate choices; however, it has also turned out to be a barrier due to the knowledge intensiveness of the process and unreliable product information, especially in SMEs (small and medium-sized enterprises). This research focuses on developing a simple decision-making framework for selecting sustainable products as alternatives. The proposed framework is developed with analysis of two transparent factors: (1) life cycle cost and (2) technology growth rate. A decision-making model also provides a visual comparison of the alternatives in consideration with quadrant model. The resulting framework provides a distinct anticipated outcome description for each product considered; thus, enabling the decision-makers to identify and choose the most suitable alternative. The framework is expected to influence the spread of sustainable products in SMEs.

22 August 2023 in Applied Sciences
Verified
Journal Article
Open access
PDF file
JCJaehyun Choi
Jaehyun Choi
MHMinhui Ha
Minhui Ha
JLJin Gang Lee
Jin Gang Lee
As improper inspection of construction works can cause an increase in project costs and a decrease in project quality, construction inspection is considered a critical factor for project success. While traditional inspection tasks are still mainly labor-intensive and time-consuming, computer vision has the potential to revolutionize the construction inspection process by providing more efficient and effective ways to monitor the progress and quality of construction projects. However, previous studies have also indicated that the performance of vision-based site monitoring heavily relies on the volume of training data. To address the issues of challenging data collection at construction sites, this study developed models using transfer learning-based object detection models incorporating data augmentation and transfer learning. The performance of three object detection algorithms was compared based on average precision and inference time for detecting T/S bolt fastening of steel structure. Despite the limited training data available, the model’s performance was improved through data augmentation and transfer learning. The proposed inspection model can increase the efficiency of quality control works for building construction projects and the safety of inspectors.

30 September 2026 in Journal of Construction Automation and Robotics
Verified
Journal Article
Open access
PDF file
HJHwa Sung Jang
Hwa Sung Jang
SMSi On Min
Si On Min
YLYong Chan Lee
Yong Chan Lee
JSJae Hun Sim
Jae Hun Sim
JLJin Gang Lee
Jin Gang Lee