Navigation

Search This Site

Digital Imaging and Photogrammetry in Creating Virtual Environments for Structural Health Monitoring

Presenters Name: 
Hong Liang
Primary Research Mentor: 
Devin Harris
Secondary Research Mentor: 
Time: 
2:30 - 2:45
Time of Presentation: 
2019 - 2:30pm to 2:45pm
Session: 
4
Location: 
Commonwealth Room
Presentation Type: 
Oral
Presentations Academic Category: 
Engineering
Grant Program Recipient: 
USOAR Program
Abstract: 

Life cycle management and performance monitoring of infrastructure is vital for gauging the conditions of said infrastructure. The specific technologies for three dimensional modeling and virtual reality environments have progressed to the point where they are popular choices for documenting a structure. The interactivity of virtual reality game engines provide a potential space for highly detailed analysis of photo-realistic models, with higher analysis programs being able to integrate themselves as analysis tools. Photogrammetry uses purely two dimensional images taken from a standard DSLR camera to virtually reflect the true infrastructure scale and data. The work presented here is a methodology and digital workflow for transitioning from structure to virtual model using photogrammetric software and finally interactive environment in virtual reality. This project was tested on several real-world case studies, including McCormick Bridge over Emmet Street in Charlottesville, and various the Virtual Reality and Structures Laboratory spaces in Thornton Hall. All images were captured using a Nikon D610 DSLR camera from which Agisoft Photoscan created a virtual model using the images. The model was then made interactive inside the Unity Game Engine and integrated with an automatic crack detection algorithm developed by Mohamad Alipour, a PhD Candidate in the Link Lab and the Department of Civil and Environmental Engineering. The algorithm outlines and eventually measures cracking damage in structures on the model. The results of this research will be recommendations for how models like these can be further optimized for analysis and other data collection methods for structural documentation.