Biometrics A Seminar Research and compilation by: Anand M S 1PI08TE017 Dept. of Telecommunications PESIT

 INTRODUCTION  Biometrics is a rapidly evolving technology  Facilitates the automatic identification based on physical or behavioral characteristics  Eliminates the need to memorize a password or carry some form of token

 Biometrics : A definition  Science and technology of measuring and statistically analyzing biological data that is represented in human by patterns unique to every individual

Figure describing the process involved in using the biometric system

 SUBSYSTEMS OF BIOMETRIC MODEL  DATA COLLECTION  TRANSMISSION  SIGNAL PROCESSING  DECISION MAKING  DATA STORAGE

 TYPES OF BIOMETRIC SYSTEM  Fingerprint Verification  Hand Geometry  Retinal Scanning  Iris Scanning  Facial Recognition  Signature Verification  Voice Verification

 DETAIL ANALYSIS OF IRIS SCANNING  Iris scan takes about 30sec & 2sec for verification  A camera located at about 3 feet focuses on & scans iris from one side  This scan is converted into a template  Only phase information is used for recognizing iris as amplitude information is depends on extraneous factors

A IRISCODE Sample

Iris-code Generation Gabor Filtering : Complex planar wave restricted to a 2-D Gaussian envelope Aside from scale and orientation, the only thing that can make two Gabor wavelets differ is the ratio between wavelength and the width of the Gaussian envelope. Every Gabor wavelet has a certain wavelength and orientation, and can be convolved with an image to estimate the magnitude of local frequencies of that approximate wavelength and orientation in the image. A jet of Gabor wavelets consists of numerous wavelets of different wavelengths and orientations. By convolving an image with a full jet of filters it is possible to quantify the magnitude and phase of the local frequency information within the bandwidths of the jet. Gabor wavelets are formed through 2 components - a complex sinusoidal carrier and a Gaussian envelope. g(x,y)=s(x,y)w(x,y) The complex carrier takes the form : s(x,y)=e^j(2π(u0x+v0y)+P)

The real and imaginary parts can be visualized as given below Real part Imaginary Part

The real part of the function is given by : Re(s(x,y))=cos(2π(u0x+v0y)+P) The imaginary part is given by Im(s(x,y))=sin(2π(u0x+v0y)+P) The parameters u0 and v0 represent the frequency of the horizontal and vertical sinusoids respectively. P represents an arbitrary phase shift.

The second component of a Gabor wavelet is its envelope. The resulting wavelet is the product of the sinusoidal carrier and this envelope. The envelope has a Gaussian profile and is described by the following equation: g(x,y)= Ke−π(a2(x−x0)2r+b2(y−y0)2r) where : (x−x0)r=(x−x0)cos(θ)+(y−y0)sin(θ) (y−y0)r=−(x−x0)sin(θ)+(y−y0)cos(θ) The parameters used above are K - a scaling constant (a,b) - envelope axis scaling constants, θ - envelope rotation constant, (x0,y0) - Gaussian envelope peak.

To put it all together, we multiply s(x,y) by w(x,y). This produces a wavelet like this one:

Generating the Iriscode Image of an eye Unrolled image of an eye – mapped to Cartesian co-ordinates

We extract a set of unique features from this iris and then store them. When presented with an unknown iris, we can compare the stored features to the features in the unknown iris to see if they are the same. These set of features are called the “Iriscode”. Any given iris has a unique texture that is generated through a random process before birth. Filters based on Gabor wavelets turn out to be very good at detecting patterns in images.

We'll use a fixed frequency 1D Gabor filter to look for patterns in our unrolled image. First, we'll take a one pixel wide column from our unrolled image and convolve it with a 1D Gabor wavelet. Because the Gabor filter is complex, the result will have a real and imaginary part which are treated separately. We only store a small number of bits for each iris code, so the real and imaginary parts are each quantized. If a given value in the result vector is greater than zero, a one is stored; otherwise zero is stored. Once all the columns of the image have been filtered and quantized, we can form a new black and white image by putting all of the columns side by side.

Real part of Iris-code Imaginary part of Iris-code

Comparing Iris-codes for Authentication The user's eye is photographed and the iris code produced from the image. We measure the Hamming distance between two iris codes. The Hamming distance between any two equal length binary vectors is simply the number of bit positions in which they differ divided by the length of the vectors. This way, two identical vectors have distance 0 while two completely different vectors have distance 1. Its worth noting that on average two random vectors will differ in half their bits giving a Hamming distance of 0.5. The Hamming distance is mathematically defined in this equation: D=A ⊕ B/length(A)

In theory, two iris codes independently generated from the same iris will be exactly the same. In reality though, this doesn't happen vary often for reasons such as imperfect cameras, lighting or small rotational errors. To account for these slight inconsistencies, two iris codes are compared and if the distance between them is below a certain threshold we'll call them a match. This is based on the idea of statistical independence. The iris is random enough such that iris codes from different eyes will be statistically independent (ie. have a distance larger than the threshold) and therefore only iris codes of the same eye will fail the test of statistical independence. Empirical studies with millions of images have supported this assertion.

Iris scanning enabled Atm Authenticam used for iris scanning

 ADVANTAGES OF IRIS SCANNING  Iris pattern posses high degree of randomness, variability & uniqueness  Patterns are apparently stable throughout life  Encoding & decision making are tractable  Image analysis & encoding time : 1sec  Search period : iris codes/sec

 DISADVANTAGES OF IRIS SCANNING  Small target (1cm) to acquire from a distance (1m)  Located behind a curved, wet, reflecting surface  Obscured by eyelashes,lenses,reflections  Partially occluded by eyelids, often dropping  Illumination should not be visible or bright

 CHARACTERICTICS FOR SELECTING A BIOMETRIC TECHNOLOGY  Accuracy  Ease of use  Error incidence  Cost  User acceptance  Required security level  Long term stability

Character istic Finger Print Hand Geometr y RetinaIrisFaceSignatureVoice Ease of use High LowMedium High Error Incidence Dryness, Dirt,Age Hand Injury,Ag e GlassesPoor Lighting Lighting, Age,Glas ses Changing Signature s Noise,Co lds,Weath er AccuracyHigh Very High High Cost******* User Acceptan ce Medium High Required Security HighMediumHighVery High Medium High LTSHighMediumHigh Medium

 METHODS TO RATE BIOMETRICS ACCURACY  FALSE- ACCEPTANCE RATE (FAR)  FALSE- REJECTION RATE (FRR)

The plot of FAR Vs FRR representing crossover error rate

 COST ANALYSYS FOR BIOMETRICS  Hardware  Back-end processing power to maintain database  Research & testing of the system  User education  Exception processing  System maintenance

 BIOMETRICS APPLICATIONS  Indian Initiatives  Bioenable Technology, Pune  Siemens Information System Ltd., Banglore  Axis Software, Pune  Jaypeetex,Mumbai  Global Developments  Internet Security : Litronix, Usa  Windows Biometrics : Microsoft  Net Nanny Software International  Biometric Smart Cards : Polarid & Atmel

 FUTURE OF BIOMETRICS  ATM Machine Use  Travel & Tourism  Public Identity Card - Aadhar

 CONCLUSION  Your fingerprints, iris pattern, and voice will verify your identity  You can unlock your house or withdraw money from your bank with just a blink of eye, a tap of your finger, or by just showing your face

 BIBLIOGRAPHY  Biometrics : Journal of International Biometric Society  The Biometrics Consortium ……………May 2002  Electronics For You ……………………June 2002    

Thank You