Campus canteen crowd monitoring and occupancy level visualization using YOLOv8 / Marri Grace Morata, Christian Jay Tibi, and Adrian James Relles.

By: Contributor(s): Material type: TextTextDescription: 133 leavesContent type:
  • text
Media type:
  • unmediated
Carrier type:
  • volume
Dissertation note: Undergraduate Theses (BSCS) - Computer Science. Camarines Sur Polytechnic Colleges, 2024. Abstract: This study introduces a real-time Crowd Monitoring and Occupancy Level System de signed for the CSPC campus canteen. The primary objective is to address the challenges of traditional crowd monitoring methods by implementing the YOLOv8 algorithm for au tomated detection and counting of individuals. The system processes video streams from a designated area, quantifies the detected crowd, and categorizes the occupancy level as less crowded, crowded, or overcrowded. The prototype was evaluated for its accuracy and func tionality in the campus canteen, demonstrating high efficiency in detecting crowds with a precision score of 97.20%, recall of 96.08%, and accuracy of 93.50%. The results indicate that the system successfully automates crowd monitoring and occupancy level detection, making it suitable for applications in public safety, resource management, and planning. Future work could explore enhancements such as multi-camera integration, server integra tion, and predictive analytics for crowd movement trends
Tags from this library: No tags from this library for this title. Log in to add tags.
Holdings
Item type Current library Shelving location Call number Copy number Status Date due Barcode
Unpublished Materials Unpublished Materials Main Library Undergraduate Thesis Section UTH BSCS M831c 2024 (Browse shelf(Opens below)) 1-1 Not for loan 007624UM

Undergraduate Theses (BSCS) - Computer Science. Camarines Sur Polytechnic Colleges, 2024.

Includes bibliographical references.

This study introduces a real-time Crowd Monitoring and Occupancy Level System de signed for the CSPC campus canteen. The primary objective is to address the challenges of traditional crowd monitoring methods by implementing the YOLOv8 algorithm for au tomated detection and counting of individuals. The system processes video streams from a designated area, quantifies the detected crowd, and categorizes the occupancy level as less crowded, crowded, or overcrowded. The prototype was evaluated for its accuracy and func tionality in the campus canteen, demonstrating high efficiency in detecting crowds with a precision score of 97.20%, recall of 96.08%, and accuracy of 93.50%. The results indicate that the system successfully automates crowd monitoring and occupancy level detection, making it suitable for applications in public safety, resource management, and planning. Future work could explore enhancements such as multi-camera integration, server integra tion, and predictive analytics for crowd movement trends

There are no comments on this title.

to post a comment.