Dev · 2025
Coop Counter.
An AI-powered monitoring system that automatically detects and counts chickens inside a backyard coop using YOLO object detection. Built with a Raspberry Pi camera, FastAPI backend, and a React Native mobile app.
AI detection in action with bounding boxes and confidence scores
Overview
Counting chickens shouldn't require a spreadsheet.
Coop Counter started as a school project. The idea was simple: keep track of how many chickens are inside the coop at any given moment. A Raspberry Pi camera captures images every five minutes and streams live video over WebSocket to a FastAPI backend, where a YOLO object detection model counts them.
Built in collaboration with @kaelangraham, who developed the React Native mobile app while I focused on the backend, AI model, and hardware integration.
Features
What it does.
AI-Powered Counting
YOLO object detection model trained to accurately count chickens with confidence scores.
Live Camera Streaming
Real-time WebSocket video feed from a Raspberry Pi camera mounted in the coop.
Automated Scans
Schedule automatic chicken counts throughout the day, with results stored to a database.
Mobile App
React Native (Expo) app for iOS and Android with dashboard, analytics, and door control.
Gallery
Screens & hardware.
Source
Take a closer look.
The full project is open source on GitHub. Backend, mobile app, and hardware scripts all included.