OpenCV Workshop

A hands-on introduction to computer vision for first-year engineering students.

Folder Structure

opencv-workshop-local-v2/
├── assets/                          # Input images used by the scripts
├── basics/                          # All workshop files, run in order
│   ├── constants.py                 # Shared file paths
│   ├── 1io.py                       # Reading, writing, and webcam capture
│   ├── 2shape_colours.py            # Image dimensions and color spaces
│   ├── 3bitwise_threshhold.py       # Bitwise ops, masking, thresholding, live trackbar
│   ├── 4drawing_shapes.py           # Drawing shapes and text on images
│   ├── 5convulation_kernel.py       # Kernels, filters, and padding
│   ├── 6contours.py                 # Edge detection and contour analysis
│   ├── 7morphology.py               # Erosion, dilation, and morphological ops
│   ├── 8rotating_hexagon_maths.py   # Animated hexagon using trigonometry
│   ├── 9rotating_cube_maths.py      # 3D cube with rotation matrices
│   ├── 10mediapipe_hand_tracking.py # Hand detection and tracking with MediaPipe
│   ├── 11filtering.py               # All filters side by side in one window
│   ├── 12particle_systems.py        # Interactive particle explosion on click
│   ├── video_pipeline.py            # Live webcam with FPS display and clip recording
│   └── histogram_demo.py            # Histogram viewer with live brightness trackbar
├── projects/                        # Workshop project files
│   └── fruit_ninja_template.py      # Fruit Ninja starter template
├── setup_check.py                   # Run this first to verify your environment
├── requirements.txt                 # Pinned Python dependencies
└── README.md

File Reference

Day 1

File	Concept	Description
1io.py	Image and Video I/O	Load an image, show it, save it, capture from webcam
2shape_colours.py	Shapes and Color Spaces	Images are NumPy arrays; color depends on the color space
3bitwise_threshhold.py	Bitwise Ops and Thresholding	Combine or isolate image regions; live threshold trackbar
4drawing_shapes.py	Drawing	Paint rectangles, circles, text, and polygons on any image
5convulation_kernel.py	Convolution Kernels	Slide a small matrix over the image to sharpen, blur, or emboss
6contours.py	Contours and Edges	Find object outlines and describe their shape
7morphology.py	Morphological Ops	Shrink or grow white regions to clean up binary images
8rotating_hexagon_maths.py	Trig Animation	Use sin and cos to spin a hexagon on the webcam feed
9rotating_cube_maths.py	3D Projection	Rotate a 3D cube and project it onto a 2D screen

Day 2

File	Concept	Description
10mediapipe_hand_tracking.py	Hand Tracking	Detect and track hand landmarks using MediaPipe
11filtering.py	Image Filtering	Gaussian, Median, Bilateral, and Sharpened filters side by side
12particle_systems.py	Particle Systems	Click to trigger particle explosions on the webcam feed

Day 3

File	Concept	Description
video_pipeline.py	Video Pipeline	Live webcam with FPS counter; press R to record a 5-second clip
histogram_demo.py	Histogram	Side-by-side image and histogram with live brightness trackbar
fruit_ninja_template.py	Project: Fruit Ninja	Starter template to build the Fruit Ninja game

Setup (Windows)

1. Download and install Python 3.12

Open PowerShell as Administrator and run:

winget install -e --id Python.Python.3.12

2. Open the repo in VS Code and navigate to the folder

Clone the repo, open it in VS Code, then in the terminal run:

cd opencv-workshop-local-v2

3. Create a virtual environment and activate it

py -3.12 -m venv .venv
.venv\Scripts\activate

4. Install dependencies

pip install mediapipe==0.10.21
pip install opencv-python==4.8.0.74

5. Verify the installation

python -c "import cv2; import mediapipe as mp; import numpy; print('OpenCV:', cv2.__version__); print('MediaPipe:', mp.__version__); print('NumPy:', numpy.__version__)"

Or run the environment checker:

python setup_check.py

6. Verify MediaPipe is working

Run file 10 from inside the basics/ folder:

cd basics
python 10mediapipe_hand_tracking.py

A window should open showing your webcam feed with hand landmarks drawn on it.

How to Run

All scripts in basics/ import from constants.py. Run them from inside the basics/ folder:

cd basics
python 1io.py

Quick Reference

Script	Key to Exit	Notes
1io.py	Q	Opens webcam and saves a clip
2shape_colours.py	any key	Shows color space conversions
3bitwise_threshhold.py	any key	Multiple windows; ends with live trackbar
4drawing_shapes.py	any key	Static drawing demo
5convulation_kernel.py	any key	Kernel application demo
6contours.py	any key	Contour detection on an image
7morphology.py	any key	Multiple windows for each morphological op
8rotating_hexagon_maths.py	ESC	Live webcam, spinning hexagon
9rotating_cube_maths.py	ESC	Live webcam, spinning 3D cube
10mediapipe_hand_tracking.py	ESC	Live hand landmark detection
11filtering.py	any key	All filters in one side-by-side window
12particle_systems.py	Q	Click on webcam feed to spawn particles
video_pipeline.py	Q	R to record a 5-second clip
histogram_demo.py	any key	Trackbar adjusts brightness live

If a window seems frozen, click on it and press a key. OpenCV needs window focus to register keypresses.

Webcam Note

Scripts that open the webcam: 1io.py, 8rotating_hexagon_maths.py, 9rotating_cube_maths.py, 10mediapipe_hand_tracking.py, 12particle_systems.py, video_pipeline.py. Make sure the webcam is connected and not in use by another app.

Prerequisites

• Basic Python: variables, loops, functions
• Some familiarity with NumPy arrays is helpful but not required