Online mobile-based system for signature verification

Other Title(s)

نظام التحقق من التواقيع باستخدام الهواتف الذكية

Dissertant

al-Tarawinah, Islam Hasan

Thesis advisor

al-Tarawinah, Raghad Muhammad
al-Abbadi, Muhammad

University

Mutah University

Faculty

Information Technology College

Department

Computer Science Department

University Country

Jordan

Degree

Master

Degree Date

2017

English Abstract

In this thesis we investigated the possibility of verifying user’s signatures using gesture-based features on smartphones.

Three shapes were used to implement a primary seed to signature verification system; lines, triangles, and circles.

These shapes were analysed to extract the feature that we used it in our study.

The features extracted were X-Y coordinates, finger’s pressure, and the number of stored points that reflects the speed of drawing.

These four features are common for the three shapes; also many other features are derived from these common features, such as: length of line, direction of motion, and the longest straight line between the point of origin and any point on the circumference of a circle or a triangle and its direction.

Decision Tree algorithm was use as a classifier with three scenarios of threshold calculations.

The classification phase came at two stages: first one is for each shape separately and the second is to take the shapes classification results then make a decision.

The main aim of this thesis is to get a good balancing between False Acceptance Rate (FAR) and True Rejection Rate (TRR) ratios and between precision and sensitivity of the system.

This leads us to get a good accuracy for the system.

The 22 participants involvedin this research included 20genuine users and 2 foreign users.The 20 genuine users also included 9 females and 11 males and they have ages between ten and seventy years.

The results showed a good accuracy and balancing between FAR and TAR ratios.

Also, higher classification accuracy was achieved when combining the three shapes’ gesture features as opposed to using a single shape gesture feature.

The results also showed that the average of finger pressure of females is less than males’ one.

Keywords: Signature verification, touch-screen, gesture-based, accuracy

Main Subjects

Information Technology and Computer Science

Topics

• Smartphones
• Digital signatures

No. of Pages

66

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-749304