DL-enabled web application for Singaporean residents to detect and asses building defects
Structural Health Monitoring is an essential process in civil engineering for evaluating the condition and detecting damage in civil structures. Due to the advancements in electronic devices and Artificial Intelligence technologies, Deep Learning has emerged as a potent tool in Structural Health Mon...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2024
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/176682 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Structural Health Monitoring is an essential process in civil engineering for evaluating the condition and detecting damage in civil structures. Due to the advancements in electronic devices and Artificial Intelligence technologies, Deep Learning has emerged as a potent tool in Structural Health Monitoring. However, existing Deep Learning-enabled Structural Health Monitoring techniques primarily cater to professionals, leaving normal residents unable to evaluate defects with the help of advanced Structural Health Monitoring technologies. To address this gap, this project aimed to develop a user-friendly Deep Learning-powered Web Application for Singaporean residents to detect and assess spalling defects on building surfaces. Employing the Next Pacific Earthquake Engineering Research Hub ImageNet framework by Gao et al., the Web Application enables users to upload images of defects and receive results including damage type and severity, localization and segmentation of defects, and advice. Testing and evaluations were conducted to ensure that the Web Application functions and provides satisfactory classification, localization and segmentation results. Deployment of this Web Application and more improvements could be done in future work. |
---|