Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/148088
Full metadata record
DC FieldValueLanguage
dc.contributor.authorTan, Raymond Rui Mingen_US
dc.date.accessioned2021-04-22T13:24:26Z-
dc.date.available2021-04-22T13:24:26Z-
dc.date.issued2021-
dc.identifier.citationTan, R. R. M. (2021). Crowd monitoring using deep learning. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/148088en_US
dc.identifier.urihttps://hdl.handle.net/10356/148088-
dc.description.abstractIn the past year, the Coronavirus disease 2019 (Covid-19) has spread worldwide, leading to 120,424,082 cases and 2,665,379 deaths worldwide. Due to the fatality of virus, governments around the world, including Singapore, are enforcing rules and regulations to reduce the spread of Covid-19. One such rule is social distancing, where groups must distance themselves from each other to limit the crowds in areas. However, there is an increasing need for enforcement officer to ensure that safe distancing measures are practiced. This leads to a large number of resources being used to ensuring social distancing instead of using it for other purposes in the economy. To combat the problem, I have initiated a possible solution that will reduce the number of resources required to maintain safe distancing in the community. The Crowd Monitoring Web Application aims to automate the process mentioned above. The application could be receiving a live video feed from a device, such a closed-circuit television (CCTV) or web camera and detects the number of people in the video real-time. The real-time crowd counting is done using a deep learning model, Supervised Spatial Divide-and-Conquer network (SS-DCNet). When the crowd level exceeds the allowed number, a warning is given to the crowd to remind them to social distance, causing social distancing to be actively enforced without having the need for enforcement officers. The application was made using a client server architecture where the python with Flask with python was used for the server and HTML, CSS and JavaScript was used for the client. Further details on the architecture of the application, testing process and results, constraints and limitations as well as future improvements are also documented in the chapters below.en_US
dc.language.isoenen_US
dc.publisherNanyang Technological Universityen_US
dc.relationSCSE20-0348en_US
dc.subjectEngineering::Computer science and engineering::Computing methodologies::Image processing and computer visionen_US
dc.titleCrowd monitoring using deep learningen_US
dc.typeFinal Year Project (FYP)en_US
dc.contributor.supervisorQian Kemaoen_US
dc.contributor.schoolSchool of Computer Science and Engineeringen_US
dc.description.degreeBachelor of Engineering (Computer Science)en_US
dc.contributor.supervisoremailMKMQian@ntu.edu.sgen_US
item.grantfulltextrestricted-
item.fulltextWith Fulltext-
Appears in Collections:SCBE Student Reports (FYP/IA/PA/PI)
Files in This Item:
File Description SizeFormat 
FYP_report.pdf
  Restricted Access
1.9 MBAdobe PDFView/Open

Page view(s)

200
Updated on Jun 28, 2022

Download(s) 50

35
Updated on Jun 28, 2022

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.