Better Air, Better Health Creating an indoor air quality...
summary
A home system that actually measures CO2, temperature and humidity (particles are planned) uses recycled microcontrollers then settles on a Raspberry Pi 4 collecting sensor readings every ~10 minutes, serving a Flask dashboard, and running an on‑device CNN 1D with rolling windows that retrains as new data arrive. Predictions appear hourly to alert before CO2 hits concerning ranges (~1,000 ppm good, ~2,000 ppm problematic). Experiments report RMS error and intermittent accurate overlap with labels, with obvious spikes when visitors arrive and seasonality when windows stay closed.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:27Health impact of air quality
- 3:58Scientific context of CO2
- 6:38Hardware development challenges
- 10:05Data collection setup
- 13:12Predictive modeling approach
- 17:57Machine learning implementation
- 24:35Final system and conclusions
related talks
34mWhat's that smell? Detecting air quality with Python, Raspberry Pi, and Redis 🌳
23mRaspberry Pi Weather Station
33mSmart and Simple Data Driven Hydroponic Sensing Applications Using Python, Flask & Raspberry Pi
36mAnalyzing Air Quality with Python
39mPython and IoT: From Chips and Bits to Data Science
26mPython & Internet of Things
16mPython in Smart Farm
32mData Science in the Internet of Things with Python and Spark