Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for AquaGuard AI

AquaGuard AI: Intelligent Underwater Waste Detection and Environmental Analysis Assistant

Inspiration

Oceans cover more than 70% of our planet, yet millions of tons of plastic waste enter marine ecosystems every year. Plastic bottles, bags, fishing nets, and other debris threaten marine life, disrupt ecosystems, and eventually impact human health through the food chain.

The inspiration behind AquaGuard AI came from the idea of combining Artificial Intelligence, Computer Vision, and Underwater Robotics to create a smart assistant capable of detecting underwater waste and providing meaningful environmental insights. Instead of relying solely on manual underwater surveys, we envisioned an intelligent system that can automatically identify marine debris and assist in cleanup operations.

Our goal is simple:

"Seeing Beneath the Surface, Protecting the Future."

What It Does

AquaGuard AI is an AI-powered underwater waste detection assistant that:

Captures real-time underwater video using OpenCV.
Detects waste objects using the YOLO object detection model.
Identifies marine debris such as plastic bottles, cans, cups, and other waste materials.
Uses a Large Language Model (LLM) to generate:
- Waste type explanation
- Environmental impact analysis
- Cleanup recommendations
Displays all information through an interactive ocean-themed dashboard.

The system helps researchers, environmental organizations, and underwater robotic platforms monitor marine pollution efficiently.

How We Built It

The project consists of four major modules:

1. Computer Vision Layer

We used OpenCV to:

Capture live video from the camera.
Process image frames.
Stream video to the web interface.

2. Object Detection using YOLO

We integrated YOLO (You Only Look Once) for real-time object detection.

The model analyzes every frame and predicts:

[ Object = f(Image) ]

where:

(Image) = Camera frame
(f) = YOLO detection model
(Object) = Detected waste category

Examples:

Bottle
Cup
Can
Plastic Bag
Marine Debris

3. AI Environmental Analysis

After detection, the object name is sent to a Large Language Model.

The AI generates:

Environmental consequences
Marine ecosystem risks
Suggested cleanup methods

For example:

Detected Waste: Plastic Bottle

Impact: Plastic bottles may remain in oceans for hundreds of years and harm marine species.

Cleanup Strategy: Robotic collection and recycling.

4. Interactive Dashboard

The frontend was developed using:

HTML
CSS
JavaScript

Features include:

Live camera feed
AI Scan button
Waste detection panel
Environmental impact report
Ocean-themed animated interface

Challenges We Faced

Building AquaGuard AI involved several technical challenges:

1. Camera Integration

Streaming the OpenCV camera feed directly into a web application required creating a Flask video streaming endpoint and synchronizing frames efficiently.

2. Waste Classification

The default YOLO model was trained on general objects such as:

Person
Chair
Car

This sometimes caused incorrect detections.

To address this, we implemented:

Waste class filtering
Custom waste categories
AI-based post-processing

3. LLM Integration

We faced issues such as:

Invalid API keys
Model deprecation
API response errors

We resolved these by:

Updating the Groq model
Implementing exception handling
Providing fallback environmental reports

What We Learned

This project helped us gain hands-on experience in:

Computer Vision using OpenCV
Real-time Object Detection with YOLO
Flask API development
Frontend development using HTML, CSS, and JavaScript
Large Language Model integration
AI-powered environmental applications
End-to-end deployment of intelligent systems

Future Scope

In future versions, AquaGuard AI can be integrated with:

Autonomous underwater robots
GPS-based marine pollution mapping
Custom-trained underwater waste datasets
Robotic grippers for automatic waste collection
IoT-based ocean monitoring systems

Impact

AquaGuard AI demonstrates how Artificial Intelligence can contribute to sustainable ocean conservation by making underwater waste detection smarter, faster, and more accessible.

By combining Computer Vision, AI, and Environmental Awareness, AquaGuard AI takes a step toward cleaner oceans and a healthier planet for future generations.