How QR Codes Work

The Encoding Pipeline

Creating a QR code is a multi-step pipeline that transforms human-readable data into a precise grid of black and white modules.

Encoding Modes

QR codes support four encoding modes, each optimized for different types of data. The encoder automatically selects the most efficient mode, or can mix modes within a single code.

Structural Patterns

Before any data is placed, several structural patterns are written into the matrix at fixed positions. These patterns are what make QR codes recognizable and enable scanners to detect, orient, and decode them.

Finder Patterns

The three large squares in three corners (top-left, top-right, bottom-left) are finder patterns. Each is a 7×7 module square with a distinctive ratio of dark-light-dark (1:1:3:1:1) in any direction.

This unique ratio allows scanners to detect a QR code and determine its orientation regardless of rotation.

Alignment Patterns

For QR codes Version 2 and above, smaller 5×5 alignment patterns are placed at regular intervals. These help the decoder compensate for perspective distortion — such as when a QR code is photographed at an angle.

Timing Patterns

Alternating black-and-white modules running horizontally and vertically between the finder patterns form timing patterns. These establish the grid coordinate system, helping the decoder determine the exact row and column of each module.

Error Correction: Reed-Solomon

QR codes use Reed-Solomon error correction, the same algorithm used in CDs, DVDs, and deep-space communication (NASA's Voyager probes). This allows QR codes to remain readable even when partially damaged, obscured, or covered by a logo.

For a deep dive into each level and how to choose the right one, see QR Code Error Correction Explained.

Masking

After data and error correction modules are placed, the QR code applies one of eight masking patterns. Masking inverts specific modules to prevent problematic patterns — like large uniform areas or patterns that mimic finder patterns — that could confuse scanners.

The encoder evaluates all eight patterns against a penalty scoring system and selects the mask that produces the lowest penalty score — meaning the most scannable result. The chosen mask ID is stored in the format information area.

How Scanning Works

When your smartphone camera points at a QR code, the decoding process happens in milliseconds:

1. Detection — The camera identifies the three finder patterns and determines the code's position, size, and rotation.
2. Sampling — The decoder uses timing and alignment patterns to map each pixel to its corresponding module in the grid.
3. Unmasking — The format information is read to determine which mask was applied, and that mask is reversed.
4. Decoding — Data codewords are extracted from the module grid in the correct reading order.
5. Error correction — Reed-Solomon decoding detects and corrects any errors in the data.
6. Interpretation — The decoded bitstream is parsed according to the mode indicators to reconstruct the original text, URL, or other data.