15-12-2012, 02:19 PM
FINGERPRINT RECOGNITION
[attachment=43919]
INTRODUCTION
Objective and Scope
Study literature and methodology used
Implement a fingerprint recognition system
(verification)
(given 2 fingerprints identify whether they belong
to same person or not)
Give experimental results
What is Fingerprint Recognition ?
o It refers to the automated method of verifying a match between two
human fingerprints.
o One of the forms of biometrics used to identify an individual and verify
their identity.
o Fingerprints are highly reliable for of identification because of their
uniqueness and consistency over time.
IMAGE ENHANCEMENT
Enhancement by Fast Fourier Transformation
divide the image into small processing blocks (32 by 32
pixels) and perform the Fourier transform
enhance a specific block by its dominant frequencies
Extract Region of Interest (ROI)
Morphological operations ‘OPEN’ and ‘CLOSE’ used
OPEN’ expand images and remove peaks introduced by
background noise
‘CLOSE’ shrink images and eliminate small cavities
Final ROI = subtraction of CLOSE from OPEN area
LIMITATIONS & FUTURE WORK
Matching not implemented completely
Matching algorithm has Large computation complexity
Relatively Low verification rate ~ 75 %
Algorithm is vulnerable to scaling, translation.
Need better methods like - Fingerprint Matching Using
Minutiae Triangulation
[attachment=43919]
INTRODUCTION
Objective and Scope
Study literature and methodology used
Implement a fingerprint recognition system
(verification)
(given 2 fingerprints identify whether they belong
to same person or not)
Give experimental results
What is Fingerprint Recognition ?
o It refers to the automated method of verifying a match between two
human fingerprints.
o One of the forms of biometrics used to identify an individual and verify
their identity.
o Fingerprints are highly reliable for of identification because of their
uniqueness and consistency over time.
IMAGE ENHANCEMENT
Enhancement by Fast Fourier Transformation
divide the image into small processing blocks (32 by 32
pixels) and perform the Fourier transform
enhance a specific block by its dominant frequencies
Extract Region of Interest (ROI)
Morphological operations ‘OPEN’ and ‘CLOSE’ used
OPEN’ expand images and remove peaks introduced by
background noise
‘CLOSE’ shrink images and eliminate small cavities
Final ROI = subtraction of CLOSE from OPEN area
LIMITATIONS & FUTURE WORK
Matching not implemented completely
Matching algorithm has Large computation complexity
Relatively Low verification rate ~ 75 %
Algorithm is vulnerable to scaling, translation.
Need better methods like - Fingerprint Matching Using
Minutiae Triangulation