EcoLens AI

Inspiration

With the growing climate crisis, deforestation, and rapid urbanization, we realized that satellite imagery holds immense potential to reveal environmental changes that often go unnoticed. However, analyzing such imagery requires domain expertise and heavy tools — something most communities, NGOs, and students don’t have access to. That’s what inspired us to build EcoLens AI — an intelligent, accessible platform that empowers anyone to visually and quantitatively understand environmental transformations over time.

What it does

EcoLens AI allows users to: i) Upload or automatically fetch “before” and “after” satellite images of a region. ii) Analyze visible changes such as vegetation loss, urban expansion, or water depletion. iii) Generate AI-powered visual overlays and metrics that quantify environmental impact. iv) Provide insights to support sustainable planning, research, and awareness. Essentially, EcoLens AI acts as an eco-detection lens — spotting environmental transformations with clarity and speed.

How we built it

i) Frontend: Built with React, providing an intuitive interface for image upload, location selection, and visualization. ii) Backend: Developed using FastAPI — handling file uploads, analysis requests, and data management. iii) Storage: Images are securely stored in Google Cloud Storage (GCS), with metadata managed through Firestore. iv) APIs: Integrated Google Earth Engine API for fetching satellite images based on coordinates and date ranges. v) AI Analysis (Upcoming Phase): Uses computer vision techniques (OpenCV + TensorFlow) to detect changes in terrain, vegetation, and water cover. vi) Deployment: Containerized backend deployed on Google Cloud Run for scalability and low-latency performance.

Challenges we ran into

i) Integrating Google Earth Engine API with FastAPI and authentication setup. ii) Managing large geospatial image files efficiently within limited runtime memory. iii) Handling CORS issues during frontend-backend communication. iv) Ensuring consistent performance in serverless deployment using Cloud Run. Despite the hurdles, each challenge strengthened our system design and cloud deployment understanding.

Accomplishments that we're proud of

i) Successfully built a fully functional image upload and management module with cloud integration. ii) Achieved seamless satellite image retrieval using Google Earth Engine. iii) Created a scalable backend architecture ready for real-time AI analysis. iv) Designed a modular system that can easily expand to environmental monitoring at national or global scale.

What we learned

i) Hands-on experience with Google Cloud ecosystem (GCS, Firestore, Cloud Run). ii) Practical exposure to geospatial data processing and satellite imagery analysis. iii) Building efficient asynchronous APIs with FastAPI for handling concurrent image uploads. iv) Importance of clear data pipelines and model modularity in scalable AI projects.

What's next for EcoLens AI

i) Integrate AI-based semantic segmentation for automated land cover classification. ii) Add a trend dashboard to visualize environmental health metrics over time. iii) Enable community-driven datasets for collaborative environmental tracking. iv) Expand to policy and education sectors to drive sustainable action through data transparency.