Iris Recognition PDF

IRIS ECOGNITION AUTOMATED BIOMETRIC RECOGNITION QUALITY SPEED ACCURAC Y INCREASED POPULARITY AS BIOMETRIC IRIS A POPULAR ICON IN MOVIES 3 BIOMETRICS WITHIN EYE Conjuntival Vasculature Iris IRIS AS BIOMETRIC? RELATIVELY ACCESSIBLE WHY USE THE IRIS ? ANATOM...

IRIS ECOGNITION AUTOMATED BIOMETRIC RECOGNITION QUALITY SPEED ACCURAC Y INCREASED POPULARITY AS BIOMETRIC IRIS A POPULAR ICON IN MOVIES 3 BIOMETRICS WITHIN EYE Conjuntival Vasculature Iris IRIS AS BIOMETRIC? RELATIVELY ACCESSIBLE WHY USE THE IRIS ? ANATOMY OF THE EYE MULTI-LAYERED STRUCTURE ANATOMY OF THE EYE LIKE CAMERA APERTURE BUT, ON THE EYE  LIGHT REGULATION CONSTRICTI DILATION ON LIVENESS DETECTION (COUNTER CIRCUMVENTIO ANATOMY OF THE IRIS IRIS UP CLOSE COLLARETTE CRYPT PATTERN OF ZIG ZAGS  Diverse Rich Texture, Coarse Unique (Even in Twins) IRIS MOVEMENT: RADIAL + CONTRACTION FURROWS COLOUR Pigmentation controlled by melanin granules (genetically determined) Cellular density of the stroma can affect colour. Colour plays little role. Texture detail of anterior portion is useful.  IRIS RECOGNITION SYSTEM – PATTERN MATCHING Acquisition, Segmentation Normalization Encoding/matching modules.NEAR INFRA RED - (NIR) SEVERAL STAGES ACQUISITION 2D image. Monochrome camera sensitive to Near-Infrared (NIR) range of the electromagnetic spectrum. External source of NIR light, often co-located with acquisition system, to illuminate iris. CRITICAL COMPONENT - SEGMENTATION NORMALISATION Transform iris texture from Cartesian coordinates via rubber sheet model. “Unwrapping of the iris”. Results in a rectangular entity Rows correspond to the angular direction Columns correspond to its radial direction.  NORMALISATION (a) Variations in pupil size (Dilation) (b) Variations across the population. (c) Normalization ensures common mapping - Matching stage, two normalized irises registered by translation operation that accounts for tilting of head at acquisition ENCODING AND MATCHING Multi-resolution analysis of iris. Highly Detailed 2D Gabor Wavelets to extract the local phasor information of the iris texture. Resulting 2D binary code (iris code) Two such iris codes can be compared using the Hamming distance, which computes the number of corresponding bits that are different across them HARDWARE PROBLEMS Subject’s eye must be stable, in close proximity to the camera. Research has explored the possibility of acquiring iris scans from moving subjects who are at a considerable distance. IRIS AT A DISTANCE EYE DETECTION >> IRIS ACQUISITION SITA IRIS ON THE MOVE SUMMARY Iris is highly Unique and Difficult to Circumvent Pattern around pupillary zone can be extracted using NIR imaging camera Pattern is normalised to unique 2D Iris Code Acquisition, Iris at a Distance and On the Move becoming a reality.