Online Signature Verification Based on Signer Dependent Sampling Frequency and Dynamic Time Warping

Previous studies have shown that choosing a different sample rate or signature point count provides better accuracy in online signature verification systems. However, the sampling rate that minimizes the error may vary on the database and signer levels. In this work, we studied the effect of choosing individual sampling frequencies for each signer and proposed a signature verification system based on signer-dependent sampling frequency. The system was tested on five different databases, using several features and preprocessing methods. Results showed accuracy improvement in 70% of the overall 500 tests and 92% in the chosen system where z-normalization and six samples used in the preprocessing phase.