Enhanced method for handwritten signature recognition

Other Title(s)

طريقة محسنة لتمييز التوقيع اليدوي

Thesis advisor

al-Ani, Muzhir Shaban

Comitee Members

al-Bahadili, Husayn
Kakish, Malik

University

Amman Arab University

Faculty

Collage of Computer Sciences and Informatics

Department

Department of Computer Science

University Country

Jordan

Degree

Master

Degree Date

2011

English Abstract

Electronic biometric systems became available in last few decades.

Handwriting signature recognition as a behavioral biometric plays an important role in our life, because of its widespread acceptance among people, especially in security transactions.

In this thesis, an enhanced method for handwritten signature recognition has been developed and implemented in a comprehensive system for handwritten signature recognition.

The system includes many elements that are integrated to generate the entire system.

Database of 1000 Arabic and English signatures has been introduced.

The system has been separated into two important phases: training and testing. In the training phase, preprocessing process is applied firstly for signatures enhancement, then feature extraction process is applied using discrete wavelet transform (DWT) and vectors fusion.

In enrollment process, after support vector machine (SVM) is trained, an individual template is stored in the database.

In the testing phase, two recognition processes have been introduced: verification and identification.

In each process, SVM classifier makes a decision for classification. The system indicates good results for verification and identification compared to other approaches

Main Subjects

Electronic engineering

Topics

• Writing
• Data processing

No. of Pages

63

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-528949