Inspiration
The inspiration for Sterilysense came from the critical need to improve hygiene and infection control in healthcare facilities, especially in resource-limited areas. We observed that traditional disinfection methods, often manual and chemical-based, are inconsistent, labor-intensive, and environmentally harmful. This motivated us to develop a smart, automated disinfection system using UV-C light and TiO₂ photocatalysis to address these challenges effectively.
What it does
Sterilysense is an automated disinfection system designed to ensure cleaner, safer environments by detecting and eliminating microbial contamination on surfaces. It
How we built it
• System Architecture: We designed a layered architecture combining UV-C disinfection, TiO₂ coatings, AI-driven contamination detection, and IoT-based monitoring. • Hardware Integration: We integrated components like the ESP32 microcontroller, GUVA-S12SD UV sensor, UV-C light strip, and relay module. • AI Model: We trained a contamination detection model using TensorFlow Lite to analyze sensor readings and automate UV light activation based on contamination levels. • Software Development: We used Python, TensorFlow, and Edge Impulse for model training and deployment. Arduino IDE was used to program the ESP32, and OpenCV assisted with image processing where needed. • Prototype Testing: We simulated contamination levels, validated the system's performance, and optimized threshold values for real-world conditions.
Challenges we ran into
Sensor Calibration: Accurately calibrating the UV sensor to detect contamination without false positives. • AI Model Optimization: Training a model that could run efficiently on ESP32’s limited resources. • Power Management: Ensuring reliable power delivery for the UV light strip using a boost converter. • Data Collection: Gathering diverse datasets to improve model generalization for different surfaces and contamination types. • Environmental Constraints: Ensuring consistent performance in environments with varying light conditions and surface materials.
Accomplishments that we're proud of
We are the finalist of Hacxerve India level Hackathon 2025 and HackTu 6.0 Hackathon 2025.
What we learned
Throughout this project, we gained valuable insights into: The effectiveness of photocatalysis in microbial disinfection. The capabilities of UV-C light in breaking down pathogens. The potential of machine learning and AI models to analyze sensor data and improve contamination detection. The practical applications of IoT systems using ESP32 for real-time monitoring and automation.
What's next for Sterilysense
AI Model Enhancement: We plan to improve the contamination detection model by training it with more diverse datasets, including different surfaces and microbial strains, to increase accuracy and reliability.
Smart Dashboard Integration: Develop a web-based and mobile-friendly dashboard for real-time contamination monitoring, historical data analysis, and predictive maintenance insights.
Edge AI Integration: Integrate more advanced Edge AI models to detect contamination patterns faster and reduce dependency on threshold-based detection.
Energy Optimization: Optimize the power consumption by implementing dynamic UV light control based on contamination density, ensuring efficient energy use.
Scalability for Larger Spaces: Enhance system design to support larger, multi-floor facilities by adding distributed sensors and centralized monitoring.
Partnerships and Commercialization: Collaborate with hospitals, laboratories, and industries to pilot the system and gather real-world performance data.
Sustainability Focus: Explore eco-friendly materials for the coating and implement practices that minimize environmental impact.
Log in or sign up for Devpost to join the conversation.