CASA: Smart Smoking and Vaping Detection System
A Raspberry Pi computer vision system that detects smoking and vaping in public no-smoking areas and alerts enforcers.
Overview
CASA is a smart monitoring system for the automated detection of smoking and vaping in designated public no-smoking areas. When it detects a violation, it captures a single still evidence image and sends it to enforcers, while an operator monitors the system from a dashboard.
My role
I am the Primary Investigator on a six-member capstone team. I:
- Contribute to the design, development, and evaluation of the system
- Authored the review of related literature and the methodology chapter (research design, privacy safeguards, data gathering protocol, and data storage and disposal plan)
- Prepared the research ethics application, informed consent forms, and data privacy documentation in compliance with the Data Privacy Act (Republic Act No. 10173); ethics clearance was granted in October 2026
- Coordinated with local government units to secure permission and site arrangements for field testing in public areas
Tools
- Python 3, YOLOv8 Nano with ByteTrack for detection and tracking
- Raspberry Pi 5 with PoE IP cameras
- SQLite for records, Tkinter for the operator dashboard
Status
System testing is underway, with the final defense in November 2026.
