Biometric Feasibility Analysis3

Biometric Feasibility Analysis (I) Final Year Project – Part 1 PRESENTED BY : DAYANI A/P MURALI : 1031125472 : B.Eng. (Hons) Electronics majoring in Bio-Instrumentation SUPERVISOR : MS.HIDAYATI ABDUL AZIZ MODERATOR : MR.KHAIR RAZLAN OTHMAN

OBJECTIVES 

To study various types of biometric identification method



To do experiment and analysis on the strength and weakness of physiological signal biometric compared to physical biometric method

OVERVIEW OF PROJECT •

Biometric

•

Biometric method

•

Project aim

PART 1

OUTLINE OF PRESENTATION         

Introduction on biometrics Types of biometric identification Characteristics of biometrics Biometric system Feature Selection Feature Extraction Implementation on Part 1- Fingerprint Conclusion on Part 1 Implementation on Part 2- ECG

INTRODUCTION TO BIOMETRICS • •

a way of analyzing problem in biological sciences. to recognize or identify the difference between two individual. Biometrics used In two major ways

Identification (determining who a person is)

Verification/ Authentication (determining if a person is who they say they are)

3 Basic types of classification

Physical

•Fingerprint •Iris •Hand Geometry •Face •Vein Patterns

Behavioral

•Signature •Voice •Key stroke

Physiological

•ECG •EEG

Physical Characteristics 

Physical biometrics measures the inherent physical characteristics on an individual.



It can be used for either identification or verification.

FINGERPRINT

BERTILLONAGE

VEIN PATTERN

HAND GEOMETRY

IRIS

FACE

Physiological -measures the characteristics acquired from the internal body condition.

Electrocardiogram

Electroencephalogram (EEG) A

B

Behavioral Characteristics Behavioral biometrics basically measures the characteristics which are acquired naturally over a time. It is generally used for verification 

SIGNATURE



VOICE



KEYSTROKE

Characteristics of Biometrics       

Unique Measurable Universality Permanence Performance Acceptability Circumvention

3 Steps on application of Biometrics ENROLLMENT

DATA STORAGE

COMPARISON /MATCHING

Biometrics System SENSOR

SIGNAL PROCESSING

MATCHING ALGORITHM

TRANSMISSION DATA STORAGE

FEATURE SELECTION 

Selecting a subset of relevant features for building robust learning models.



A process done before feature extraction

FEATURE EXTRACTION 

 

Process of developing a representation for or a transformation from the original data Also a special form of dimensionality reduction. Process of transforming/ filtering/ reducing the input data

FEATURES

GENERAL FEATURES •COLOR •TEXTURE •SHAPE

DOMAIN-SPECIFIC FEATURES •FACE •FINGERPRINT

Methods of Feature Extraction 

Pre-processing –This is an image enhancement process.



Feature extraction – extraction of the unique feature



Post processing – Removal of false details

IMPLEMENTATION OF PART 1

Basics of fingerprints

2 1

LEFT LOOP

WHORL

RIGHT LOOP

ARCH

DOUBLE LOOP

TENTED ARCH

Preprocessing stage (image enhancement)

Enhance color Brightness, Contrast

Remove noise

Minutiae extraction using Binarization and skeletoning enhanced image is binarised.  skeleton of image is obtained. - To obtain ridge which is one pixel wide  A minutiae is extracted. - Minutiae point is either the ridge ending or ridge bifurcation of the one pixel ridge 

Post processing 

The features obtained after extraction may contain false details.

Types of false details Spike

Bridge

Hole

Break

Spur

Ladder

3 levels of involved in the filtering of false minutiae: 





Level 1: Removes the false ridge ending. Level 2: Removes the first five types of minutiae . Level 3:limiting the maximum number of minutiae present in the thinned image

Experimental Results on Image Enhancement 1

2

3

Experimental Results on Feature Extraction

4

5

6

7

8

CONLUSION (PART 1) By studying the various types of biometric identification method, we are able to identify the strength and weaknesses of the specified method in the second part of the final year project and thus, will be able to produce a more reliable result at the end of the project.

Implementation for FYP Part 2

THANK YOU