Answers are grounded in this paper and cite their sources.
Quick Response (QR) codes are two-dimensional matrix symbols built from dark and light modules, the basic units that hold the encoded data, and any smartphone camera can read them within moments. In a few years they have moved from industrial labelling into everyday routines such as opening a restaurant menu or paying at a neighbourhood shop. Several intrinsic properties explain this uptake: Reed–Solomon error correction allows a symbol to be decoded even when part of it is damaged, dirty or hidden; omnidirectional scanning removes the need for careful alignment; and generation is inexpensive and open to anyone, which has encouraged use in marketing, education, healthcare and secure data exchange. The same openness, however, carries costs. A QR code can hold only a limited amount of data, cannot be read by a person, depends on external servers once it points to a URL, and offers no built-in way of proving who created it, which leaves it exposed to tampering, counterfeiting and quishing (QR phishing). This paper gives a structured overview of QR code technology. It describes the symbol structure and the encoding and decoding procedures, examines the main technical and security limitations, reviews protective measures based on machine learning, encryption, steganography and watermarking, and surveys applications in augmented reality, education, cashless payments and the Internet of Things.
No takes yet. Share an insight, caveat, or question.
Verma et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: