OpenDartboard
[](https://github.com/OpenDartboard/OpenDartboard/blob/main/LICENSE)
[](https://github.com/OpenDartboard/OpenDartboard/releases)
[](https://github.com/opendartboard/opendartboard)
[](https://github.com/opendartboard/opendartboard)
[](https://github.com/opendartboard/opendartboard)
[](https://discord.gg/b8YwrbN2ju)
**OpenDartboard** is a hobby-friendly, fully FOSS toolkit for building an automatic _steel-tip_ dart–scoring station at home or in a bar.
- **Headless Scorer** — Use a Raspberry Pi Zero 2 W with 3 cameras to detect darts and output scores in standard notation (`T20`, `S5`, `D12`, …) via a easy to consume [WebSocket API](docs/api.md).
- **Computer Vision** — Includes a lightweight OpenCV detector to run efficiently on a Pi Zero 2 W or build your own detector via a [Custom Detector](docs/detectors.md) plugin.
The goal: **drop-in freedom** for home tinkerers or bar owners who want "Automatic darts scoring" without closed hardware, subscriptions, or vendor lock-in.
## Table of Contents
- [Quick Start](#quick-start)
- [Recommended Clients](#recommended-clients)
- [Project Status & Roadmap](#project-status--roadmap)
- [Tech Stack](#tech-stack)
- [Hardware Reference](#hardware-reference)
- [Development Environment](#development-environment)
- [API Documentation](#api-documentation)
- [Custom Detectors](#custom-detectors)
- [Contributing](#contributing)
---
## Quick Start
```shell
# 1 Flash MicroSD with Rasbperi Pi Imager (https://www.raspberrypi.com/software/) | Image: Raspberry Pi OS Lite (64-bit)
# 1.1 - set hostname to `opendartboard`.
# 1.2 - set username as `pi`, and any password.
# 1.3 - set Wi-Fi name and password.
# 1.2 - enable SSH in services.
# 1.4 - Flash MicroSD - once complete power on Raspberry Pi
# 1.5 - Connect to it with terminal/cmd using SSH:
ssh pi@opendartboard.local
# 2. Update the package lists
sudo apt-get update
# 3. Download & install the latest .deb release
wget https://github.com/OpenDartboard/OpenDartboard/releases/download/v0.1.3/opendartboard_0.1.3-1_arm64.deb
sudo apt install -y ./opendartboard_0.1.3-1_arm64.deb
# 4. Run it and Watch the scores
opendartboard --autocams
```
> **Tip**: Need a quick debug dashboard? Run [`debug.opendartboard.org`](http://debug.opendartboard.org) in any modern browser to see the score output, camera feeds, calibrations images, and more.
## Clients
🚧 We are working on a set of official clients to make it easier to use OpenDartboard. These will be available for various platforms, and will allow you to view scores, play, manage settings, and more.
```yml
- **Web**: [COMMING SOON] - A web client for viewing scores and managing settings.
- **Android**: [COMMING SOON] - A native Android app to connect to your OpenDartboard server.
- **iOS**: [COMMING SOON] - An iOS app to connect to your OpenDartboard server.
- **Tablet**: [COMMING SOON] - A tablet client for viewing scores and managing settings.
- **Windows**: [COMMING SOON] - A Windows client for viewing scores and managing settings.
- **Linux**: [COMMING SOON] - A Linux client for viewing scores and managing settings.
- **macOS**: [COMMING SOON] - A macOS client for viewing scores and managing settings.
- **TV's**: [COMMING SOON] - A TV client for viewing scores on your big screen.
```
For now, you can try out the [Debug Dashboard](http://debug.opendartboard.org) or integrate with your own applications using the [WebSocket API](docs/api.md).
## Project Status & Roadmap
| Phase | Target | Status (ETA) |
| --------------- | -------------------------------------------------- | ------------ |
| **Packaging** | Debian package (.deb) for easy installation -- | ✅ released |
| **Development** | Docker‑based dev environment for consistent builds | ✅ ready |
| **MVP** | Basic Auto‑calibration via OpenCV | ✅ done |
| **CI/CD** | Basic CI pipelines & tagged releases | ✅ live |
| **MMR** | Scoring via OpenCV (`geometry_detector`) | 🚧 WIP |
| **API** | WebSocket API for real‑time score streaming | ✅ v0.1 |
| **Polish** | Improved accuracy & self‑calibration | 🗓 T.B.D. |
---
## Tech Stack
| Layer | Tech | Notes |
| ------------------- | ---------------------------- | :----------------------------------- |
| **Computer Vision** | `OpenCV` | INT8‑optimised model for ARM |
| **Runtime** | C++ | High‑performance dart detection |
| **API** | `WebSocket` · `JSON` | Real‑time score streaming on `13520` |
| **Infrastructure** | `systemd` · `udev` | Robust auto‑start & camera hot‑swap |
| **Development** | `Docker` · `Debian Bullseye` | Reproducible builds |
| **Distribution** | `.deb` package | CI‑checked, one‑command install |
---
## Hardware Reference
| Item | Minimum spec | Example |
| ------------ | -------------------------------------------------- | ----------------------------- |
| SBC | Raspberry Pi Zero 2 W (+ self-powered USB 2 hub) | Waveshare USB HUB HAT (B) |
| Cameras (×3) | USB 2.0 webcams outputting MJPEG @ 1280x720 30 fps | HBVCAM OV2710 100° |
| Lighting | 360° LED ring | DIY SmartLite 12 V LED 6000 K |
| Power | 5 V / 3 A PSU | Any USB PD brick + adapter |
---
## Development Environment
```sh
# Clone repository
git clone https://github.com/OpenDartboard/OpenDartboard.git
cd OpenDartboard
# Build the dev image (once)
docker compose build opendartboard
# Open an interactive shell with everything mounted
# Run `$env:PWD = (Get-Location).Path` #FOR WINDOWS
docker compose run --rm --service-ports opendartboard /bin/bash
# Builds and installs binary
make build
# Run the binary with mocks
opendartboard --debug --cams mocks/cam_1.mp4,mocks/cam_2.mp4,mocks/cam_3.mp4 --width 1280 --height 720
```
## API Documentation
See [`docs/api.md`](docs/api.md) for the full WebSocket specification & client examples.
## Custom Detectors
Want to experiment with your own CV pipeline? Check out [`docs/detectors.md`](docs/detectors.md) for implementation guides and examples.
## Contributing
Pull requests are welcome — but please open an issue first so we can discuss design & approach. 🎯
