A Signature Verification System for Class Attendance Records

Abstract

The project presents design and implementation of an artificial intelligent based signature verification system for class attendance management. The aim was to solve the problem of impersonation and ensure that unauthorized persons do not have access to academic proceedings. The study used the agile methodology. The methods used in the system development are data collection, data processing, feature extraction, clustering and verification of outcome. In the data processing, techniques such as binarization and edge detection were used to process the signature data, before they are extracted with graphical feature extraction method for training with clustering algorithm and then verify the student. The system was implemented with Matlab programming language and the evaluated. The result showed that the software was able to verify authenticities of students during class attendance verification correctly.