Inspiration
Urban areas face recurring civic issues—water leaks, potholes, garbage overflow—but reporting them is often slow and unorganized. Citizens need a simple, effective way to report issues while local authorities need data-driven insights to respond faster. CivicSense AI was inspired by the idea of bridging citizens and smart city governance using AI.
What it does
CivicSense AI allows citizens to report civic issues via text or images. It automatically: Classifies issues (Water Leak, Road Damage, Garbage, Street Light) Detects issues from images using lightweight AI heuristics Tracks status in a dashboard Provides city-wide analytics & heatmaps for authorities This ensures faster response, better monitoring, and improved urban management.
How we built it
Frontend & Dashboard: Streamlit for interactive UI and tabs (Report Issue, Dashboard, Analytics) Data Handling: Pandas to manage issue records AI Logic: Lightweight computer-vision heuristics with PIL + NumPy for image detection Analytics & Maps: Streamlit built-in charts + map visualization Randomized geo-coordinates simulate city locations for demo heatmaps
Challenges we ran into
Simulating realistic image-based detection without heavy ML models Ensuring charts and heatmaps work with Python 3.10+ safely Maintaining session-state for dynamic issue tracking Designing intuitive dashboard without overwhelming the user
Accomplishments that we're proud of
Fully functional, hackathon-ready app with text + image-based reporting Real-time dashboard with status tracking Heatmap & analytics showing city-wide problem areas Lightweight AI image detection that works without complex ML installation
What we learned
How to combine AI heuristics with citizen data to create a practical solution Building user-friendly dashboards that update dynamically with session-state Importance of realistic sample data and visuals for demo presentations Handling Python version & library compatibility issues in real projects
What's next for CivicSense AI
Integrate real ML/CNN or YOLO models for more accurate image detection Add user accounts & admin roles for authorities vs citizens Store persistent data in databases (SQLite/PostgreSQL) Enable notifications to authorities via email/SMS for urgent issues Expand coverage to new civic domains like traffic congestion, pollution monitoring, street lights
Log in or sign up for Devpost to join the conversation.