Project Mufasa

Wild animal tracking system

Comment

Inspiration

Human-animal conflict creates a lot of negative impacts for both humans and wild animals. Injury and loss of life of humans and wildlife, damage to human property, crop damage, and habitat destruction are some of the significant impacts of these conflicts. So there is a need to develop a system that detects any presence of interactions of wild animals in any region without causing any harmful effect to human beings and wild animals the interference

What it does

system that detects any presence of interactions of wild animals in any region without causing any harmful effect to human beings and wild animals the interference

Key Features

  • Real-time Animal Detection: Instantly identifies wild animals in images and videos.
  • Geospatial Visualization: Displays animal locations on an interactive map.
  • Notification System: Sends real-time alerts to local communities and authorities about animal presence.

How we built it

To create That system steps are following 1) Create a Detection Model: Computer Vision ML model that detects Trained Classes like Tiger, Lion, Leopard, Hyena, Fox, Wolf Data Collection --> Data Cleaning --> Data Annotation --> Model Training --> Validation/ Testing

2) Use Trained Model In Remote Device: Remote Devices mean Hardware That uses to installed in remote areas and detects the trained classes Here we use Raspberry pi (But for now use experimental scenario)

3) Notification System: Installed Hardware when detects an animal it notifies our server or app about predator/intruder

4) Dashboard App: An app that shows us where and when animals detected with image proof

Challenges we ran into

Not having Powerful GPU Not having Remote Device (Raspberry pi) for hardware testing Folium Integration

Accomplishments that we're proud of

  • Train Custom Detection Model
  • Without Hardware create real life scenario for testing
  • Created first project in Online Hackathon

What we learned

Folium, Streamlit, Computer vision

What's next for Project Mufasa

  • Integration of Drone Surveillance: Expand the system's reach with aerial monitoring.
  • Species Recognition: Train the model to identify specific animal species for more detailed insights.
  • Mobile Application: Develop a mobile app for on-the-go notifications and reporting.

Built With

Share this project:

Updates