Farmers in India often face challenges in identifying and tracking their cattle and buffaloes.
Traditional methods like branding, tagging, or manual monitoring are either inaccurate, costly, or inconvenient.
This leads to issues in livestock management, breeding, insurance, and overall productivity.
Pashu-Drishti is a YOLOv8-powered vision system that can accurately detect and identify different Indian cattle and buffalo breeds.
The solution is optimized to run on low-end smartphones, making it accessible to rural farmers without requiring expensive hardware.
- 🔍 Identification: Detects individual cattle & buffalo breeds with high accuracy.
- 📱 Lightweight: Built with YOLOv8, and being optimized further with YOLOv8-nano for better performance on resource-constrained devices.
- 🚜 Farmer-first approach: Tailored to support livestock owners in rural India.
👨🌾 Farmers are the primary beneficiaries of this system.
It aims to help them in:
- Tracking livestock efficiently
- Preventing cattle mix-ups
- Improving breeding records
- Facilitating insurance and health monitoring
The dataset used for training is self-collected from Indian farms and cattle shelters, ensuring real-world diversity and applicability.