Signature verification research
Signature verification is probably the oldest biometrical identification method, with a high legal acceptance. Even if handwritten signature verification has been extensively studied in the past decades, and even with the best methodologies functioning at high accuracy rates, there are a lot of open questions. The most accurate systems almost always take advantage of dynamic features like acceleration, velocity and the difference between up and down strokes. This class of solutions is called on-line signature verification. However in the most common real-world scenarios, this information is not available, because it requires the observation and recording off the signing process. This is the main reason, why off-line signature analysis is still in focus of many researchers. Off-line methods do not require special acquisition hardware, just a pen and a paper, they are therefore less invasive and more user friendly. In the past decade a bunch of solutions has been introduced, to overcome the limitations of off-line signature verification and to compensate for the loss of accuracy. Most of these methods have one in common: they deliver acceptable results but they have problems improving them.
To allow the continuous imporvement of a signature verifier the verification process must be broken down in clearly isolated steps with a well defined data flow. This kind of isolation allows us to independently benchmark and optimize our system modules. In addition our research focuses on features with real-world semantical meaning. For example we locate, describe and compare the loops (and other similar features) of signatures. This way we can deliver an easyily interpretable feedback to our users, describing why the comparison succeeded or failed.
Our main research areas are:
Feature extraction (identifying and quantifying the significant features of signatures)
- Feature matching (defining distance measures between features and creating mappings between them)
- Stroke reconstruction (extracting the original strokes and their physical parameters from a 2D image)
- Image registration (creating a 2D mapping between signature images)
Resources (Siganture databases and samples)