Improved Detection Rates with New Generation of Biometric Signature Comparison

Share Article

The signature is the last remnant of the hand-written document in a digital world, and is considered an acceptable and trustworthy means of authenticating all written documents and business approvals. Verification of digitized signatures is the most natural solution to the problem of authenticating documents digitally. xyzmo’s top-notch version of the SIGNificant Biometric Server features a new generation of signature comparison capabilities.

The signature is the last remnant of the hand-written document in a digital world, and is considered an acceptable and trustworthy means of authenticating all written documents and business approvals. Dynamic Signature Verification is the modern equivalent to signature cards, used for subsequent identification of the customer by comparing the customer’s signature with the signature on the card.

xyzmo’s top-notch version of the SIGNificant Biometric Server - http://bit.ly/i1kp3s - features a new generation of signature comparison capabilities. This new and improved version optimally offers a balance between customer acceptance and security thanks to a versatile threshold factor, enabling a varying authentication process for different environments. Therefore bringing the correct fraud detection rates to a maximum while bringing the false positive (detecting a fraud mistakenly) and false negative (missing a fraud when there is one) to a minimum.

Dynamic Signature Verification authenticates the identity of individuals by measuring their captured handwritten signatures. The signature is treated as a series of movements that contain unique data, such as personal rhythm, acceleration and pressure. Unlike electronic signature captures that are often used today, Dynamic Signature Verification does not treat the signature as a graphic image. With graphic images, such as the scanned-in signatures we often attach to our documents, it is not possible to detect the dynamics within each individual’s signature; hence, the signatures can easily be copied. By contrast, Dynamic Signature Verification measures exactly how the signature is signed.

The basic idea behind Dynamic Signature Verification is the transformation of natural hand fluctuations into a mathematical structure called a ‘personal profile’. This transformation is one-way only; hence, the hand (pen) movements can be transformed into a personal profile but the reverse operation is virtually impossible. Pen movements are measured in up to five ways: horizontal and vertical movement, pressure, angle and tilt.

The personal profile has two important characteristics:
1. It is very stable (and comparable)
2. It occupies just a few hundred bytes, regardless of the size and complexity of the signature.

The personal profile is updated each time the user signs, and the profile’s record of the signature is highly flexible. As time passes, each person’s signature tends to change, and the xyzmo algorithm is able to adjust the personal profile to adapt to these changes continuously. These profiles enable trustworthy real-time verification of a signature by comparison of the recorded parameters of the handwritten signatures against the pre-enrolled profile.

What results can be expected with this new generation?

If a forger has no idea as to how the signatory signs, then the chance of success for such a fraud is close to 0%. If a forger knows what the signature looks like (e.g. having seen it in print), then the chance of success is less than 0.5%.

In addition to the obvious benefits of digitized signatures, deriving from their cultural acceptance and easy implementation in any organization, Dynamic Signature Verification is one of the most accurate methods, as it cannot be copied or replicated in the way that fingerprints or other physical biometrics can. Furthermore, it is intuitive, rapid and cost-effective, and involves compact data. These benefits make it a superb solution for document authentication in most business environments and workflows.

# # #

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Gerald Cäsar
Visit website