Protecting the Great Barrier Reef with Real-Time AI
The Great Barrier Reef, one of the most diverse ecosystems on the planet, faces numerous threats, including climate change, pollution, and outbreaks of the coral-eating crown-of-thorns starfish (COTS). To combat these challenges, a collaborative project between Google and Australia’s Commonwealth Scientific and Industrial Research Organization (CSIRO) has developed a real-time AI model to monitor and protect the reef.
Summary
This article explores how AI technology is being used to help protect the Great Barrier Reef. The project employs computer vision detection models to identify damaging outbreaks of COTS through live camera feeds, enabling scientists to address growing populations quickly and protect the valuable ecosystem.
The Challenge
Coral reefs support about 25% of sea species, including fish, invertebrates, and marine mammals, despite covering less than 1% of the ocean floor. However, they are under significant threat from COTS outbreaks, which can cause irreversible damage if left unchecked.
The Solution
The project developed an edge ML platform built on top of the NVIDIA Jetson AGX Xavier, which can analyze underwater image sequences and map out detections in near real time. The researchers used an annotated dataset from CSIRO to develop an accurate object detection model that uses a live camera feed rather than a snorkeler to detect the starfish.
How It Works
The model processes images at more than 10 frames per second with precision across a variety of ocean conditions, including lighting, visibility, depth, viewpoint, coral habitat, and the number of COTS present. When a COTS starfish is detected, it is assigned a unique ID tracker, linking detections over time and video frame.
Key Features
- Real-time detection: The model can detect COTS in real time, enabling scientists to address growing populations quickly.
- High accuracy: The model achieves a sequence-based F2 score of 0.80 for the 1080p model and 0.78 for the 720p model.
- Efficiency: The model can process images at more than 10 frames per second.
Technical Details
| Model | Resolution | FPS | F2 Score |
|---|---|---|---|
| 1080p | 1080p | 11 | 0.80 |
| 720p | 720p | 22 | 0.78 |
Benefits
The use of AI in monitoring and protecting the Great Barrier Reef offers several benefits:
- Improved accuracy: AI models can detect COTS more accurately than traditional methods.
- Increased efficiency: AI models can process images much faster than human annotators.
- Real-time monitoring: AI models enable real-time monitoring, allowing scientists to address growing populations quickly.
Future Directions
The project aims to showcase the capability of machine learning and AI technology applications for large-scale surveillance of ocean habitats. Future work includes exploring the relationship between species composition, reef health, and halo presence and size.
Additional Information
For those interested in learning more about AI applications in marine conservation, the following resources are available:
- CoralNet: A web portal hosting nearly 225,000 coral reef images.
- Google Colab: An open-source platform for accessing the crown-of-thorns starfish detection pipeline.
Final Thoughts
The integration of AI technology in marine conservation is a promising development. As these technologies continue to evolve, they will play an increasingly important role in protecting our planet’s most vulnerable ecosystems.
Conclusion
The use of AI in protecting the Great Barrier Reef is a significant step forward in conservation efforts. By leveraging real-time AI models, scientists can monitor and address threats to the reef more effectively, helping to preserve this valuable ecosystem for future generations.