Bida: Design, implementation, and characterization of a low-cost biometric identification system using Hand Dorsal Vein Pattern Analysis

Date of Publication

2012

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Macario O. Cordel, II

Defense Panel Chair

Jocelynn W. Cu

Defense Panel Member



Abstract/Summary

Rapid advancements during these past few decades on computer technology also develop methods for digital impersonation and identity thefts. This requires our systems to be more secured and reliable. One of the popular methods to address this problem is through biometrics systems. However, commercially available vein pattern biometric identification systems are expensive as such this study focused on the development of a low-cost biometric identification using hand dorsal vein pattern analysis.

The BIDA System used a modified web camera (IR) to capture Hand Dorsal Vein Patterns (HDVPs) of persons. The acquired HDVPs undergone Image Pre-processing methods to enhance the image, to acquire the region of interest, to normalize, to binarize, and to skeletonize (thinning). A combined box method and Euclidean distance approach were used to extract the features from the HDVPs. The template database served as storage for enrolled templates, and feature matching decided if the user is identified in the system or not.

Tests were conducted to assess the design, implementation, characterization and performance of the system. IR Webcam image capture test results show that 3mW IR LED radiant power, 120/255 gain level, 80 /255 brightness level, 128/255 contrast level, 0 saturation, 0 white balance, 0 backlight compensation and 255/255 sharpness was the preferred camera settings to capture desirable HDVPs. ROI rotation test results show that a maximum rotation of 30˚ clockwise and 60˚ counter clockwise can be handled by the system. Based from the results of the Usability and Accuracy Metrics test, the system was able to achieve 96% Correct Match, 2% False Accept, and 2% False Reject of the total 55 HDVPs.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU16769

Shelf Location

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

Physical Description

1 v. (various foliations) ; 28 cm.

This document is currently not available here.

Share

COinS