[attachment=3961]

BIOMETRIC FACE AND IRIS AUTHENTICATION

---

Presented by
V. Divya
M. Janani
G. Lakshmi
J. Magdalene Milan

Project Guide
Mrs. V. Vinodha Vijayaragavan.

---

OBJECTIVE AND IDEA:
Enhance security purpose and right to vote
The face and iris images are taken and compared with the database images and turn on the polling machine if and only if it is authenticated.



IRIS IS A UNIQUE KEY IN THE WORLD
No two irises are alike.
There is no detailed correlation between the iris patterns of even identical twins, or the right and left eye of an individual.
Accuracy is greater than finger prints and DNA.
APPROVAL GIVEN BY CENTRE OF SCIENCE AND MATHEMATICS CONGREGETION OF UK
FLOW DIAGRAM:
ALGORITHM USED:



PCA: Principle Component Analysis
PCA, commonly referred to as the use of Eigen faces.
PCA approach is used to reduce the dimension of the data and removes the information that is not useful.
Thus revealing the most effective low dimensional features of facial patterns.
PCA algorithm is optimal in the sense of efficiency.



FACE RECOGNITION
Flow Diagram:
Input Image
rgb to gray

DCT
PCA
LMS
STEP1: INPUT IMAGE
STEP 2: RGB TO GRAY


STEP 3: DISCRETE COSINE TRANSFORM (DCT)
Computers store images as an NxN matrix of values that represent pixels
For example
256 gray-scale image each pixel is stored as a value between 0 “ 255
0 = black pixel
255 = white pixel
Value between are shades of gray.
Hence we going to DCT.
Two Dimensional DCT
Example
After subtracting,


STEP 4: PRINCIPAL COMPONENT ANALYSIS(PCA)



Principle
Linear projection method to reduce the number of parameters
Transfer a set of correlated variables into a new set of uncorrelated variables
Map the data into a space of lower dimensionality
Form of unsupervised learning

Properties
It can be viewed as a rotation of the existing axes to new positions in the space defined by original variables
New axes are orthogonal and represent the directions with maximum variability

STEPS IN PCA:
A.Calculate the covariance matrix
For the dataset of p variables(dimensions) X1;X2; Xp for n individuals.
Then we have a n x p data matrix X.

The covariance matrix is
Thus covariance and variance are calculated




B.TO CALCULATE EIGEN VECTOR & EIGEN VALUE
The Characteristics equation of Eigen Vectors are given as
(A - lI)X = 0
This is a homogeneous system of equations, and from fundamental linear algebra, we know that a solution exists if and only if
det (A -leigenval I) = 0
Using PCA in the Face Recognition (FR) (1)
STEP 5: LMS ALGORITHM
It calculates the difference between eigen vectors for the face.
(Eigen vector1-Eigen vector2)^2
If o/p =0 : authenticated
If o/p != 0 : unauthenticated
FACE RECOGNITION
IRIS RECOGNITION:
STEPS IN IRIS RECOGNITION:
STEP1: INPUT IMAGE




STEP 2: GAUSSIAN NOISE
Noise by definition is just unwanted sound.
It is added to reduce further addition of noise.
Gaussian coefficent
0.0113 0.0838 0.0113
0.0838 0.6193 0.0838
0.0113 0.0838 0.0113




STEP 3: EDGE DETECTION:
In our project we use Canny edge detection as it overcomes various drawbacks present in other detectors.
Canny coefficients are convolved with the filtered image.



CANNY COEFFICIENTS:
-1 -1 -1 -1 -1
-1 -1 -1 -1 -1
-1 -1 20 -1 -1
-1 -1 -1 -1 -1
-1 -1 -1 -1 -1
STEPS 4: NORMALISATION



STEP 5: IMAGE LOCALISATION
The database contains already processed normalised value.
The current normalised value is compared with the database , if found similar , the iris is recognised.


COMPARING THE RECOGNITIONS
If both comparison is positive, it is authenticated and turns on the polling machine. Or else it is unauthenticated.



Software required: MATLAB
MATLAB is a high-performance language for technical computing.
It integrates computation, visualization, and programming in an easy-to-use environment where problems and solutions are expressed in familiar mathematical notation.
MATLAB is an interactive system whose basic data element is an array that does not require dimensioning.



GUI:GRAPHICAL USER INTERFACE
How does it work
1. We can create buttons , axes( to display images) according to the task we perform and provides aesthetic look.

2. The corresponding codes will be generated in editor window and call back functions can be coded into it as per the need.
3. When we run the program, it calls the GUI automatically and we can display the output through it.



FUTURE WORK:
Last year of may 2009, only 36% voting takes place in TamilNadu and 60% voting amid violence.
Polling rate has been declined.
Thus this project helps individuals by making it available at the door step.



CONCLUSION:
PCA algorithm which helps in identifying and distinguishing the unique features of all the individuals.
It provides an enhanced security purpose to avoid the mal practices which is still taking place in our country.

We have provided a helping hand to our government to have secured method of voting.
THANK YOU