Satellite Identification of Illegal Fishing

Inspiration

Significant areas of marine importance and high biodiversity have high rates of illegal fishing. These areas are hard to track even with platforms like Global Fishing Watch. High temporal resolution satellite imagery with meter level resolution is available to catch these illegal and shadow fishing fleets.

What it does

Use deep learning to analyze satellite imagery of areas of high conservation importance for potentially illegal boating traffic.

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for Satellite Identification of Illegal Fishing

Built With

• keras
• opencv
• python
• scikit-learn
• tensorflow