Page 77 - 2617_JSSPS_C-8

Images on computers: When you view an image on your computer or phone, it is made up of
pixels. Each pixel's colour is represented by binary numbers. The combination of these binary codes
creates the image you see on the screen. For examples, a simple black-and-white image uses "0" for
black and "1" for white. For colors, we just use longer strings of binary.

Audio files: When you listen to music or watch a video, the sounds are converted into binary code.
The sound waves are sampled and then stored as a series of 0s and 1s that your computer reads and
plays.

Barcode scanners: The information on a product's barcode is converted into binary so that scanners
can read it and process the data, such as the price or product details, quickly. The black bars reflect
less light (0) and the white spaces reflect more light (1). It’s literally a physical version of binary.

Communication systems: Text messages, emails or even video calls are transmitted using binary

numbers. The data is converted into binary code, sent over networks and then decoded back into
readable formats.
These examples show how the binary number system plays a crucial role in many everyday technologies.

BINARY IN AI APPLICATIONS

In Artificial Intelligence (AI), binary is a fundamental part of how information is processed. Computers,
including those used in AI systems, work using binary code, which consists of only two values: 0 and 1.

These two numbers represent everything in a computer, from images to sounds to text. The reason for
using binary is that electronic devices like computers can easily distinguish between two states, such
as on (1) or off (0). When you use AI to perform tasks like recognising faces, translating languages or
playing music, everything gets converted into binary to be understood and processed.

How AI Reads Images as Numbers

When AI processes an image, it doesn't see it the way humans do. Instead, it converts the image into a
grid of numbers. Each pixel in an image is represented by a number, usually based on the colours that
make up the pixel. For example:

Simple logic: In a basic black-and-white (binary) image, a 0 might represent a black pixel, while a 1

represents a white one.
Complex logic: For color images, AI uses more bits to represent depth. Each pixel is typically broken
down into three channels: Red, Green, and Blue (RGB). Each channel is assigned a value (usually
0–255), which is then converted into an 8-bit binary sequence.

The goal: By turning an image into a multi-dimensional array of numbers (tensors), the AI can use

mathematical formulas to detect patterns, such as the curve of an eye or the edge of a door.
Binary Data in Voice and Text Recognition

For AI to "hear" or "read," it must turn continuous human expression into discrete digital packets. In
AI applications like voice recognition or text recognition, binary data is used to capture and interpret
sound and words.

Number Systems & Binary in Computing and AI 75