Signature verification system

Date of Publication

1993

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

Subject Categories

Computer Sciences

College

College of Computer Studies

Department/Unit

Computer Science

Abstract/Summary

The Signature Verification System (SVS) is a hardware and software system which determines the authenticity of a signature. It takes the signature from an input device, preprocesses the signature and passes it to a computing process. The computing procedure will compare the previously authenticated signatures to the sample being validated. Given the fact that no two signatures are exactly alike, a tolerance value is specified to accept or reject a signature sample. The differences in the image and in the pressure exerted in writing signatures are the characteristics considered to distinct an authentic signature with an invalid one. The signature of two individuals will surely be different because the image and the pressure are unique factors for each individual. This is due to the fact that man's manner of writing has been developed from forms adopted since his early age. Preprocessing is applied on the signature image and on the downward pressure exerted in writing the signature on a digitizing tablet. It includes the normalization adn grayscaling of the signature images and pressures. The measurement of the pressure is facilitated by incorporating a circuitry designed by the proponents on the input device. Its main component is a sensor called the force transducer. The processing method that has been used to validate the signature samples is neurocomputing. The time elapsed to write the individual's signature is also regarded wherein if the time drops or exceeds the allowable, error, the signature is automatically invalid. The system provides a graphics editor where the user sees his signature after writing it on the input device. After processing, the system indicates whether the signature is genuine or not based on a specified tolerance. The whole system is tested, in an environment with a normal condition.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU07904

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

Physical Description

60 numb. leaves ; Computer print-out.

Keywords

Signature (Writing); Computer systems; Input design, Computer; Pattern recognition systems; Machine translating

This document is currently not available here.

Share

COinS