Skip to content

README.md

Latest commit

 

History

History
104 lines (67 loc) · 2.98 KB

README.md

File metadata and controls

104 lines (67 loc) · 2.98 KB

Code-Blue

SENTINAL 🌊 - Seeing the Unseen

Team: Blueberry
Tagline: Protecting Oceans from Afar

🌍 Overview

SENTINAL is an AI-powered ocean protection system designed to enhance maritime safety and environmental conservation by providing real-time, AI-verified pollution and ship detection. It empowers citizens and marine authorities to communicate directly, eliminating bureaucratic delays and enabling swift action on marine issues.

🔍 Problem Statement

Despite increasing ocean pollution and threats to marine biodiversity, reporting systems are:

  • Ineffective and slow
  • Riddled with bureaucratic intermediaries
  • Lacking prompt verification mechanisms

SENTINAL addresses these challenges by combining real-time satellite imaging, AI-based verification, and an interactive web interface.

💡 Solution

We developed three custom-trained AI models for:

  1. Oil Spill Detection
  2. Pollution Detection
  3. Ship Detection and Monitoring

These models analyze images submitted by satellites, officials, and citizens to confirm the presence of pollution or suspicious ship activity.

🧠 How It Works

AI Models

  • Oil Spill Detection: Verifies oil spill images from various sources.
  • Pollution Detection: Detects and verifies pollution-related entities.
  • Ship Detection: Tracks ships (using 2022 AIS data for Singapore) and maps them on OpenStreetMap via Google Earth Engine.

Backend Architecture (FastAPI)

  • "Brain": Handles all requests (login, report submission).
  • "Memory": Secure database for storing user details and verified reports.
  • "Smart Checkers": AI models validating image authenticity.
  • "Messengers": Frontend (JavaScript) handling user interactions and API calls.

Report Flow

  1. User submits a report with photo and location.
  2. Backend verifies image via AI.
  3. If valid, it is stored; else, the user is prompted to re-upload.

🛠️ Tech Stack

  • AI/ML Models: Custom-trained models (Tensorflow and YOLO)
  • Backend: FastAPI (Python)
  • Frontend: JavaScript (Interactive Web Interface)
  • Mapping: Google Earth Engine + OpenStreetMap
  • Database: Secure storage system i.e. MySQL

🚧 Challenges

  • Handling 210M AIS ship records for 2022.
  • Difficulty in collecting live satellite data.
  • Integration of machine learning models into the web environment.

🔮 Future Scope

  • Live satellite data integration
  • Scalable deployment of the website
  • New features (e.g., coral reef degradation detection)
  • Real-time ship tracking
  • Access to current AIS datasets

🌊 Impact

  • Maritime security & surveillance
  • Promotes environmental responsibility
  • Fosters citizen participation
  • Drives tech-forward ecological solutions

🤝 Contribution

Feel free to fork, explore, or contribute ideas to expand SENTINAL's mission of ocean preservation.

🐋 Let's Protect Our Oceans Together!

Made with passion by Team Blueberry 💙