QR Code Scanner

Most people reach for a QR scanner tool like this one specifically because their phone's built-in camera scanner failed on something — usually a code displayed on someone else's screen (a laptop during a video call, a projector slide, another phone). Screen-to-screen scanning is genuinely harder than paper-to-camera, and it's worth knowing why so you're not just repeatedly failing the same way.

How it works

The decoder here (jsQR) starts by hunting for the QR code's three large square "finder patterns" in the corners of the image — a very specific dark-light-dark-light ratio that's designed to be recognizable at almost any rotation or skew. Once it locks onto those three squares, it uses their positions to figure out exactly how the whole grid is oriented and scaled, then reads the small data modules inside according to that geometry. This corner-pattern approach is why QR codes tolerate rotation so well compared to 1D barcodes — you don't need to line the code up perfectly, just get all three finder squares clearly visible.

Where screen scanning goes wrong

A code on a laptop screen photographed by a phone camera introduces two problems paper doesn't have: moiré interference between the screen's pixel grid and the camera sensor's own pixel grid (which can visibly distort the fine grid lines of the QR pattern), and glare from the screen's own backlight reflecting the room's lighting back into the lens. The fix for both is usually the same — increase the screen's brightness, tilt the screen or the phone slightly off a perfectly perpendicular angle to break up the reflection, and back up a bit rather than filling the frame edge-to-edge, which gives the decoder more surrounding context to lock onto the finder patterns.

Common mistakes

• Getting too close. Filling the frame with just the QR code, with no margin, can actually hurt — the decoder benefits from a bit of surrounding space (the "quiet zone") around the code, and getting too close can also throw off focus on some phone cameras.
• Not accounting for high error-correction codes that are mostly logo. Some QR codes have a logo overlaid in the center, which works because QR error correction can reconstruct missing data — but only up to a point. If the logo is large relative to the code, or the code was already low-resolution, scanning can fail even though it looks fine to your eye.
• Assuming a scan failure means a bad QR code. If live camera scanning keeps failing on a screen, try taking a plain screenshot of the code and uploading that as an image instead — a single sharp static image often succeeds where a live reflective feed doesn't.