Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/221886
Title: AN ONTOLOGY TO DESCRIBE WORK AT HEIGHT RISKS IN PHOTOGRAPHS
Authors: ASHWIN JORDAN PEREIRA
Keywords: Building
PFM
Project and Facilities Management
Goh Yang Miang
2019/2020 PFM
Classification
Computer vision
Construction Safety
Fall from height
Ontology
Photographs
Work at Height
Issue Date: 29-May-2020
Citation: ASHWIN JORDAN PEREIRA (2020-05-29). AN ONTOLOGY TO DESCRIBE WORK AT HEIGHT RISKS IN PHOTOGRAPHS. ScholarBank@NUS Repository.
Abstract: Fall from height (FFH) is a common occurrence worldwide. In Singapore, the leading causes of fatalities and major injuries is FFH. This paper aims to develop an ontology (WaH-Onto) that provides a basis for construction safety professionals and computers to describe photos with Work at Height hazards. The WaH-onto is built on the generic ontology developed by Guo & Goh (2017) and presents an initial effort in developing a computer vision application that can detect work at height risks based on the knowledge framework developed. The WaH-onto is evaluated through criteria and tasked based evaluation and validated by senior safety professionals. This paper also presents WaH-onto-lite, a scaled-down version that excludes concepts that are not critical in retrieving the photos. Approximately 15% of photos from the donated construction safety database were work at height relevant. These photos were sorted manually. Future studies can explore using WaH-onto-lite and developing the application to detect, retrieve and reason construction safety photos automatically. While detecting hazards in photographs is useful, the industry should focus on immediate rectification and proper education of negative scenarios. Tender processes should actively look into the safety aspect of projects above price and quality. The author hopes that there will come a day with zero negative cases for WaH.
URI: https://scholarbank.nus.edu.sg/handle/10635/221886
Appears in Collections:Bachelor's Theses

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Ashwin Jordan Pereira 2019-2020_Dissertation.pdf3.44 MBAdobe PDF

RESTRICTED

NoneLog In

Page view(s)

46
checked on Dec 1, 2022

Download(s)

26
checked on Dec 1, 2022

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.