HerdNet is a deep learning-based aerial animal detection and counting system tailored for wildlife and livestock monitoring. It processes high-resolution imagery to identify and count herds using object detection models trained on diverse habitats like savannas and deserts. The system supports bounding box annotation formats, pretrained models, and reproducible pipelines via conda environments. Originally developed for academic research, it has been applied to species including buffalo, elephants, and goats using aerial and drone imagery.
HerdNet
HerdNet
HerdNet is a deep learning-based aerial animal detection and counting system tailored for wildlife and livestock monitoring. It processes high-resolution imagery to identify and count herds using object detection models trained on diverse habitats like savannas and deserts...
Organization
GitHub user
Deployment Model
Edge
Agricultural Production Step
Livestock Monitoring, Livestock Tracking
End User
Farmers, Researchers
Category
Used for
Livestock
TRL
5-6
Community Engagement
Low
Number of Forks
13
Country
Universal
Licence
Disclamer
While every effort has been made to accurately record the licensing status and all other information presented, this catalogue is provided for informational purposes only. Users are strongly encouraged to verify these details—particularly the license terms and conditions—before adopting or integrating any solution, and to take special care when no explicit license is listed in the repository. The partners of the OpenAgri project make no representations or warranties of any kind, express or implied about the completeness, accuracy, reliability or suitability with respect to the interactive catalogue or the information, products, services or related graphics contained in the catalogue for any purpose. Any reliance you place on such material is therefore strictly at your own risk.