Vision Bot – mBot Starter Pack

Who it’s for

• Students and hobbyists starting with computer vision.
• Robotics and mBot users who want colour-driven control.
• Teachers who need a structured, well-explained lab exercise.

Key features

• End-to-end pipeline: capture → HSV mask → clean-up → contour → action.
• Beginner-first explanations; jargon clarified in a glossary.
• Small, modular scripts so you can progress from “minimal” to “keyboard control”.
• Sensible defaults with clear places to tune (hue windows, S/V floors, kernel size, area minimum, cooldown).

System requirements (summary)

• Python: CPython 3.9–3.12 (64-bit).
• OS: Windows 10/11, macOS 12+, or Ubuntu 20.04/22.04.
• Hardware: Laptop/desktop (“mBot computer”) with USB or built-in webcam; 2 cores, 4 GB RAM, 720p camera. GPU not required.
• Linux extra: libgl1 may be needed for OpenCV display windows.

Quick start

1. Create and activate a virtual environment.
2. python -m pip install -r requirements.txt
3. Run vision_bot_minimal.py (press q to exit).
4. Move up to vision_bot_bbox.py for overlays, then vision_bot_keyboard.py for control.

Typical uses

• Colour-based object following in a robotics demo.
• Classroom introductions to HSV and contour detection.
• Game/simulator control via camera-tracked markers.

Versioning & support

• Scripts target stable OpenCV 4.x wheels and NumPy 1.24+.
• If your platform lacks GUI wheels, use the PDF’s troubleshooting section.
• Request a “conda” environment file or exact pins if you need strict reproducibility.