dsip and its application in biometrics authentication systems

13
Dynamic Signature Recognition Hardware Interface, Feature Extraction & Recognition Dr. Vinayak Ashok Bharadi Associate Professor & Deputy HOD Information Technology Dept. Thakur College of Engg. & Tech. Kandivali (East), Mumbai -400101

Upload: dr-vinayak-bharadi

Post on 18-Jul-2015

268 views

Category:

Engineering


4 download

TRANSCRIPT

Dynamic Signature Recognition

Hardware Interface, Feature Extraction & Recognition

Dr. Vinayak Ashok Bharadi

Associate Professor & Deputy HOD

Information Technology Dept.

Thakur College of Engg. & Tech.

Kandivali (East), Mumbai -400101

Outline

• Biometrics

• Hybrid Wavelets

• Dynamic Signature Recognition

• Hardware Interfacing

• Feature Extraction & Matching

Biometrics

• Biometrics comprises methods for uniquely recognizing humansbased upon one or more intrinsic physical or behavioral traits.

• Biometric characteristics can be divided in two main classes:• Physiological are related to the shape of the body. Examples include, but

are not limited to fingerprint, face recognition, DNA, palmprint, handgeometry, iris recognition, which has largely replaced retina, andodor/scent.

• Behavioral are related to the behavior of a person. Examples include, butare not limited to typing rhythm, gait, and voice.

Physiological Biometric Traits

FingerprintPalmprint

Finger Knuckle Prints

Face Iris

Other examples are Hand Vein, Hand Geometry, Facial Thermogram, Retina, DNA, Ear Geometry, Body Odour.

4

Behavioral Biometric Traits

Dynamic Signature

Keystroke Dynamics

Other Examples are Speech, Gait, Facial Emotions

5

Hybrid Wavelets

• Wavelets – How they are better than Transforms ?

• Applications : Replacing Transforms -> Compression, Steganography, Multiresolution Analysis …. And many more.

• Conventional Wavelets – Haar, Daubachies and others…

• Hybrid Wavelets – Type I & II

Hybrid Wavelets - Kronecker product

Kronecker product is also known as tensor product.

Kronecker product is represented by a sign

The Kronecker product of 2 matrices (say A and B) is computed by multiplying each element of the

1st matrix(A) by the entire 2nd matrix(B) as

Hybrid Wavelets

Feature Vector Generation using Hybrid Wavelets

𝐸𝑖ℎ =

𝑥=1

𝑀

𝑦=1

𝑁

𝐻𝑖(𝑥, 𝑦)2

𝐸𝑖𝑣 =

𝑥=1

𝑀

𝑦=1

𝑁

𝑉𝑖(𝑥, 𝑦)2

𝐸𝑖𝑑 =

𝑥=1

𝑀

𝑦=1

𝑁

𝐷𝑖(𝑥, 𝑦)2

Click Here for Demo

Signature Recognition : Hardware Interface

Hardware Interfacing

User Signature

Wacom Intuos4 System Driver

VBTabletInterface

.NET Framework

Library

Signature Recognition Application

Data Capturing