Development of a people monitoring system using convolutional neural networks/
Araullo, John Art Marie G.
Development of a people monitoring system using convolutional neural networks/ John Art Marie G. Araullo, Aldrin Merrell T. Buenaventura, Raphael Wayne C. De Vera, Sherwin S. Laguidao, and Zommel Jeff V. Laquindanum.-- - Manila: Technological University of the Philippines, 2025. - xii, 129pages: 29cm.
Bachelor's thesis
College of Science.--
Includes bibliographic references and index.
The increasing demand for dependable and effective crowd monitoring in crowded
areas had been addressed by the development of a People Monitoring System using
Convolutional Neural Networks (CNN). For safety, crowd control, and improved resource
management, precise, real-time data is frequently difficult to obtain using traditional
means. To detect and count people in real time, our system makes use of CNN's
sophisticated detection capabilities. To protect privacy, it also detects faces without
identifying the individuals. Webcams at entry points are used by the system to record real-
time video and monitor individuals as they enter the entry point. GPU-accelerated tools
such as CUDA make it run effectively. Important features include the ability to
automatically count people, compare faces based on different facial characteristics without
saving personal information, and generate timestamped PDF documents to assist with
security and privacy compliance, adhering to the Data Privacy Act. The system uses
continuous real-time monitoring and reporting to increase efficiency, improve crowd
control, and maintain the safety of both public and private areas. ISO 25010 standards have
been used for testing to make sure it functions properly, operates effectively, and is user-
friendly. This system represents a significant advancement in ethical monitoring
technology that respects privacy and performs effectively in both public and private
scenarios, even if it just detects and counts people without identifying them.
Computer science
Monitoring system
Convolutional neutral networks (cnn)
BTH QA 76 / A73 2025
Development of a people monitoring system using convolutional neural networks/ John Art Marie G. Araullo, Aldrin Merrell T. Buenaventura, Raphael Wayne C. De Vera, Sherwin S. Laguidao, and Zommel Jeff V. Laquindanum.-- - Manila: Technological University of the Philippines, 2025. - xii, 129pages: 29cm.
Bachelor's thesis
College of Science.--
Includes bibliographic references and index.
The increasing demand for dependable and effective crowd monitoring in crowded
areas had been addressed by the development of a People Monitoring System using
Convolutional Neural Networks (CNN). For safety, crowd control, and improved resource
management, precise, real-time data is frequently difficult to obtain using traditional
means. To detect and count people in real time, our system makes use of CNN's
sophisticated detection capabilities. To protect privacy, it also detects faces without
identifying the individuals. Webcams at entry points are used by the system to record real-
time video and monitor individuals as they enter the entry point. GPU-accelerated tools
such as CUDA make it run effectively. Important features include the ability to
automatically count people, compare faces based on different facial characteristics without
saving personal information, and generate timestamped PDF documents to assist with
security and privacy compliance, adhering to the Data Privacy Act. The system uses
continuous real-time monitoring and reporting to increase efficiency, improve crowd
control, and maintain the safety of both public and private areas. ISO 25010 standards have
been used for testing to make sure it functions properly, operates effectively, and is user-
friendly. This system represents a significant advancement in ethical monitoring
technology that respects privacy and performs effectively in both public and private
scenarios, even if it just detects and counts people without identifying them.
Computer science
Monitoring system
Convolutional neutral networks (cnn)
BTH QA 76 / A73 2025