Mask Detection & People Flow Systems

Project Overview

This project focused on the development of an Integrated Mask Wearing Classification and People Counting System for public health monitoring. Recognized with the CYCU College Student Research Creativity Award, the system addresses the need for automated compliance checks in public spaces.

Core Techniques

The proposed solution utilizes YOLOv5s for high-speed object detection and mask classification (Correct, Incorrect, None), integrated with Deep SORT for robust multi-object tracking. This combination allows for accurate Crowd Counting and individual tracking even in dense environments.

Related Publications