In biometrics, signature recognition (also called dynamic signature recognition) authenticates identity by measuring handwritten signatures. The signature is treated as a series of movements that contain unique biometric data, such as personal rhythm, acceleration, and pressure flow. Unlike electronic signature capture, which treats the signature as a graphic image, signature recognition technology measures how the signature is signed.
How it works Edit
In a signature recognition system, a person signs his or her signature on a digitized graphics tablet or personal digital assistant. The system analyzes signature dynamics such as speed, velocity, acceleration, timing, pressure, and direction of the signature strokes — all analyzed along the X, Y, and Z axes. The technology can also track each person’s natural signature fluctuations over time.
The signature dynamics information is encrypted and compressed into a template that can range from slightly larger than 1,000 bytes to approximately 3,000 bytes. These templates are large by biometric standards and reflect the variety of data available in a typical signature.
The characteristics used for signature recognition are almost impossible to replicate. Unlike an image of the signature, which can be replicated by a trained human forger and/or basic imaging technologies, dynamic characteristics are complex and unique to the handwriting style of the individual. Despite this major strength, the characteristics have a large intra-class variability (an individual's own signature may vary from one collection point to another) and this often makes recognition difficult.
Dynamic signature verification holds value as a widely-usable biometric because it can easily be integrated into existing systems based on the availability and prevalence of signature digitizers and the public's acceptance of the collection process. Across this broad scope of potential applications, signature recognition technology is actually very limited ‐ it can be used only for verification purposes due to its limited uniqueness and variations in the individual's performance. Continued research and development will help to drive to full maturity the development and application of this technology.