A Tool for Robust Offline Signature verification

Abstract: Hand written signature verification is an important utility in real time applications. It is required to verify identify fake signatures. Many existing techniques are based on the storke pixels and the underlying gray level values. Local Binary Patterns are used to obtain good results in offline signature verification. The experiments were made using GPDS corpus containing offline signatures. The corpus signatures are with uniform white background in invoices or cheques. Ferrero et al. proposed a novel offline signature verification technique that makes use of gray level features. Their technique was tested with offline signatures of various kinds in invoices and checks. In this paper, we implement that technique. We build a prototype application that demonstrates the proof of concept. The empirical results revealed that the prototype is useful in real time applications.

Keywords: Signature verification, local binary patterns, texture features, and biometrics