Finder pattern

Each finder pattern is 7 × 7 modules: a dark ring one module thick, a light ring inside it, and a solid dark 3 × 3 square in the middle. A line through its center crosses dark, light, dark, light and dark in the ratio 1:1:3:1:1, and because the rings are concentric squares, that ratio is the same whichever direction the line runs. Scanners search the image for exactly that ratio. A one-module light border, the separator, keeps each finder pattern apart from the data.

Three patterns are enough to fix the code’s position, scale and rotation, and the corner without one tells the scanner the orientation, so a QR code can be read upside down or sideways. The encoder also avoids accidental 1:1:3:1:1 sequences in the data area: the mask it picks is penalized for each one.

Finder patterns are not covered by error correction, which repairs data codewords only. If a finder pattern is hidden, smudged or redrawn beyond recognition, many scanners cannot locate the code at all, however strong its error correction.

Designers often restyle the finder patterns, sometimes called “eyes”: rounded corners, colors or circles. These usually still scan if the dark-light-dark structure and strong contrast survive. Never place a logo over a finder pattern, and in AI QR code art check them first: artwork that turns them into decoration is the most common reason an art code fails.

The finder patterns also tell a scanner how big the code is before it reads a single data module. Their centers sit 3 modules in from the edges, so the top-left and top-right centers are always the code’s width minus 7 modules apart: 14 modules in a version 1 code (21 across), 170 in version 40. Dividing that distance by the module size, which the finder pattern’s own 7-module width gives, yields 4V + 10, from which the decoder estimates the version V.