Offline Signature Verification Based on Pseudo-Cepstral Coefficients

Abstract

Features representing information about pressure distribution from a static image of a handwritten signature are analyzed for an offline verification system. From gray-scale images, its histogram is calculated and used as "spectrum" for calculation of pseudo-cepstral coefficients. Finally, the unique minimum-phase sequence is estimated and used as feature vector for signature verification. The optimal number of pseudo-coefficients is estimated for best system performance. Experiments were carried out using a database containing signatures from 100 individuals. The robustness of the analyzed system for simple forgeries is tested out with a LS-SVM model. For the sake of completeness, a comparison of the results obtained by the proposed approach with similar works published using pseudo-dynamic feature for offline signature verification is presented.

In-air signature verification system using Leap Motion

“…User authentication based on signatures is a widely-studied field in biometrics since it is one of the most accepted verification techniques (Behera et al, 2017;Impedovo & Pirlo, 2008), mainly because of its usability. Signatures verification techniques are usually divided into online (Plamondon & Srihari, 2000) and offline (Vargas et al, 2009). Both approaches try to identify and characterize the features which define the signature of each user, trying to check if a signature is original or not, applying for these a wide spectrum of classifiers (Bibi et al, 2020).…”

Cosine similarity for analysis and verification of static signatures

“…For each signer, 12 genuine signatures are used for reference, whereas 12 genuine and 30 forgeries specimens are considered for testing. The system of Vargas et al [36] computes pseudo‐cepstral coefficients from the histogram of the grey‐scale signature image and uses least squares support vector machines for signature verification. In this case, 24 genuine and 24 forgeries from 100 individuals from the GPDS database are considered.…”

Sensing In-Air Signature Motions Using Smartwatch: A High-Precision Approach of Behavioral Authentication

“…Some studies have focused on specific parts of signatures. For example, based on the characteristics that pixels of signature with high pressure appear as dark zones and therefore corresponds with gray levels conforming histogram, Vargas et al [54] computed pseudo-cepstral coefficients from the histogram of the static signature images as feature vectors for verification. Using signatures from 100 individuals, the author trained a Least Squares Support Vector Machines(LS-SVM) for classification and achieved 6.20% EER.…”

“…forgeries [12]- [15]. From the aspect of data acquisition ways, signature authentication has been extended from the offline, in which signatures are mainly produced with the conventional pen and paper-based tools, to the online where signatures start to be acquired by electronic signature pads and pens.…”