Computer VisionLesson 1
Lesson 16 min
Images as Data
How a computer 'sees' a picture: pixels and channels.
What you will learn
- ✓What a pixel actually is
- ✓RGB channels
- ✓Why images are 3D grids of numbers
Explanation
To a computer, an image is just numbers. Each pixel holds a brightness value, and colour images have three channels — Red, Green, Blue — stacked together.
So an image is a 3D grid: height × width × channels. A 224×224 colour photo is 224×224×3 numbers.
Everything in computer vision is math on these grids (often stored as NumPy arrays or tensors).
Real-world use
When you crop or brighten a photo, an app is just doing arithmetic on this grid of pixel numbers.
Common mistakes
- • Forgetting that colour images have 3 channels, not 1.
Practice
Work out how many numbers are in a 100×100 RGB image.
Knowledge check
0/1 answered1. A colour image is best described as...
Answer all questions to check.