Course 2, lesson 16 of 100, Ages 6+
How AI sees
Pictures are numbers
Like I’m 5
A picture is like a giant dot-to-dot made of tiny squares. AI looks at all the squares to work out what’s in the picture.
The big idea
To a computer, a picture is a grid of tiny squares called pixels. Each pixel is just numbers saying how much red, green and blue it has. A phone photo has millions of them.
AI learns which patterns of numbers make edges, which edges make shapes, and which shapes make a face or a cat. That's how it can find you in a photo or read the words on a sign.
Examples
- Face unlock: Your phone compares the pattern of your face with the one it saved.
- Reading signs: Translation apps find the letters in the pixels and turn them into text.
- Doctor's helper: AI looks for patterns in X-ray pixels that could mean a broken bone.
How it works
- To a computer, a picture is a grid of tiny dots called pixels. Each pixel is just a number.
- AI looks for patterns in those numbers: edges first, then shapes, then whole things like eyes or wheels.
- That’s how phones find faces in photos and how cars spot road signs.
Check your understanding
- What is a pixel?
- Options: A tiny dot in a picture; A kind of robot; A camera lens.
Answer: A tiny dot in a picture. Pictures are made of thousands of pixels, and each one is stored as numbers. - What does a computer really see in a photo?
- Options: Numbers for every tiny square of colour; Tiny people inside the screen; Nothing at all.
Answer: Numbers for every tiny square of colour. Pictures are grids of pixels, and pixels are numbers.
Remember
To AI, pictures are grids of numbers called pixels.
Talk about it
What clues help you spot a friend from far away?
Go deeper
Computer vision models learn layered features: early layers detect edges, later layers combine them into parts and objects. They power medical scan analysis, factory quality checks and tools that describe images aloud for people who are blind.