Skip to content

01-Overview.md

Latest commit

 

History

History
127 lines (87 loc) · 4.94 KB

01-Overview.md

File metadata and controls

127 lines (87 loc) · 4.94 KB

Aztec Codes: An Overview

What is an Aztec Code?

An Aztec code is a type of 2D barcode—a square grid of black and white squares (called "modules") that encodes information. You've probably seen similar codes like QR codes on products, tickets, or advertisements.

    ██████████████████
    ██              ██
    ██  ██████████  ██
    ██  ██      ██  ██
    ██  ██  ██  ██  ██
    ██  ██      ██  ██
    ██  ██████████  ██
    ██              ██
    ██████████████████

The name "Aztec" comes from the central pattern that resembles a top-down view of an Aztec pyramid.

Why Use Aztec Codes?

Aztec codes have several advantages:

  1. No quiet zone required - Unlike QR codes, Aztec codes don't need a white border around them
  2. Efficient for small data - Very compact for short messages
  3. Built-in error correction - Can still be read even if partially damaged
  4. Variable size - Automatically scales to fit your data

Anatomy of an Aztec Code

Every Aztec code has these parts:

1. The Finder Pattern (Bull's Eye)

The center of every Aztec code has a distinctive "bull's eye" pattern of alternating black and white squares. This helps scanners locate and orient the code.

Compact (9×9 finder):        Full (13×13 finder):
    █████████                  █████████████
    █       █                  █           █
    █ █████ █                  █ █████████ █
    █ █   █ █                  █ █       █ █
    █ █ █ █ █                  █ █ █████ █ █
    █ █   █ █                  █ █ █   █ █ █
    █ █████ █                  █ █ █ █ █ █ █
    █       █                  █ █ █   █ █ █
    █████████                  █ █ █████ █ █
                               █ █       █ █
                               █ █████████ █
                               █           █
                               █████████████

2. Orientation Marks

Small marks near the finder pattern tell the scanner which way is "up." This lets the code be scanned from any angle.

3. Mode Message

A ring of modules around the finder pattern contains the "mode message"—information about the symbol itself:

  • How many layers of data the symbol has
  • How many data codewords are encoded

4. Data Layers

The actual encoded information surrounds the center, arranged in layers that spiral outward. More data = more layers = larger symbol.

5. Reference Grid (Full Symbols Only)

Larger "full" symbols have a grid of alternating black and white modules every 16 rows/columns. This helps scanners stay aligned when reading large symbols.

Two Types of Symbols

Compact Symbols

  • Smaller: 15×15 to 27×27 modules
  • 1 to 4 data layers
  • 9×9 finder pattern
  • Best for short messages (up to ~50 characters)

Full Symbols

  • Larger: 31×31 to 151×151 modules
  • 4 to 32 data layers
  • 13×13 finder pattern
  • Can encode thousands of characters

How Data Gets Encoded

Here's the journey your data takes:

"Hello" → [text encoding] → [bit stream] → [codewords] → [+ error correction] → [symbol matrix]

  1. Text Encoding: Characters are converted to numbers using different "modes" (explained in Mode Encoding)

  2. Bit Stream: The numbers become a stream of bits (0s and 1s)

  3. Codewords: Bits are grouped into fixed-size chunks called "codewords"

  4. Error Correction: Extra codewords are calculated and added to protect against damage (explained in Reed-Solomon)

  5. Symbol Matrix: Codewords are arranged in the spiral pattern to create the final barcode

Error Correction

One of the most important features of Aztec codes is error correction. Using clever mathematics (Reed-Solomon codes), the barcode includes extra information that allows it to be read even if:

  • Part of the code is torn or scratched
  • There's dirt or smudges
  • Some modules are obscured

The error correction level is configurable. Higher levels:

  • Can recover from more damage
  • Require a larger symbol for the same data

Typical levels range from 5% to 95% of codewords dedicated to error correction. The default in this library is 23%.

Next Steps