FilterbankBased Fingerprint Matching

1
Filterbank-Based Fingerprint Matching

• Kitiwat Limmongkol
• Cheng-Yu Yang
• Department of Electrical Engineering
• Columbia University
• Spring 2006

E6886 Topics in Signal Processing Multimedia
Security System
Final Project Proposal Presentation
April 5, 2006
2
Presentation Outline
E6886 Topics in Signal Processing Multimedia
Security System

• Introduction
• Why should we use fingerprint?
• Project objective
• Methodology
• Why we should this method?
• Feature extraction algorithm
• Matching
• Evaluation and Improvement
• Classmates fingerprints identification
• Improvement of verification accuracy

I
II
III
3
Project Objective
E6886 Topics in Signal Processing Multimedia
Security System

• To learn and implement filterbank-based
  fingerprint matching 1 method using Gabor
  filters and calculate Euclidean distance between
  two FingerCodes.
• To experience on how to extract features,
  classifications and filtering process.

I
II
III
4
Why Filterbank-based method?
E6886 Topics in Signal Processing Multimedia
Security System

• The verification accuracy is only marginally
  inferior to the best results of minutiae-based
  algorithms 2 but this method is faster.
• This system performs better than a
  state-of-the-art minutiae-based system when the
  performance requirement of the application system
  does not demand a very low false acceptance rate.

I
II
III
5
Presentation Outline
E6886 Topics in Signal Processing Multimedia
Security System

• Introduction
• Why should we use fingerprint?
• Project objective
• Methodology
• Why we should this method?
• Feature extraction algorithm
• Matching
• Evaluation and Improvement
• Classmates fingerprints identification
• Improvement of verification accuracy

I
II
III
6
Four Main Steps in feature extraction algorithm
E6886 Topics in Signal Processing Multimedia
Security System

• 1.Determine a reference point and region of
  interest for the fingerprint image.
• 2.Tessellate the region of interest around the
  reference point.

I
II
III
7
Four Main Steps in feature extraction algorithm
(Continue)
E6886 Topics in Signal Processing Multimedia
Security System

• 3.Filter the region of in eight different
  directions using a bank of Gabor filters.
• 4.Compute the average absolute deviation from the
  mean (AAD) of gray values in individual sectors
  in filtered images to define the feature vector
  or the FingerCode.

I
II
III
8
E6886 Topics in Signal Processing Multimedia
Security System
I
II
Fig1 System diagram of the fingerprint
authentication system
III
9
Reference point definition
E6886 Topics in Signal Processing Multimedia
Security System

• Fingerprints have many conspicuous landmark
  structures and a combination of them could be
  used for establishing a reference point.
• We define the reference point of a fingerprint as
  the point of maximum curvature of the concave
  ridges in the fingerprint image

I
II
III
10
Reference point (Continue)
E6886 Topics in Signal Processing Multimedia
Security System
I
Fig2 Concave and convex ridges in a fingerprint
image when the finger is positioned upright
II
III
11
Why filtering?
E6886 Topics in Signal Processing Multimedia
Security System

• We use Gabor filter to remove noise, preserve the
  true ridge and valley structures and provide
  information contained in a particular orientation
  in the image.

I
II
III
12
Feature vector
E6886 Topics in Signal Processing Multimedia
Security System

I
II
III
13
Matching
E6886 Topics in Signal Processing Multimedia
Security System
I
II
III
14
E6886 Topics in Signal Processing Multimedia
Security System
Presentation Outline

• Introduction
• Why should we use fingerprint?
• Project objective
• Methodology
• Why we should this method?
• Feature extraction algorithm
• Matching
• Evaluation and Improvement
• Classmates fingerprints identification
• Improvement of verification accuracy

I
II
III
15
Evaluation and Improvement
E6886 Topics in Signal Processing Multimedia
Security System

• We will implement this method using our
  classmates fingerprint for identification and
  calculate the accuracy of the result.
• We also will improve the verification accuracy by
  increasing the number of sectors and filters of
  fingerprint image.
• Extra Matching improvement by using
  Neyman-Pearson 3 rule to combine scores
  obtained from the proposed filtered-based and
  minutiae-based 2 matchers.

I
II
III
16
Time Line

• Now May, 19 (Two weeks)
• Complete Filterbank-based method in the paper
  1.
• May, 20 June, 3 (Two weeks)
• Improve the verification accuracy of the system.
• June, 4 June, 10 (One week)
• Prepare final project presentation and report.

17
References

• 1 A. K. Jain, S. Prabhakar, L. Hong, and S.
  Pankanti, Filterbank-Based Fingerprint
  Matching, IEEE Trans. Image Processing, vol. 9,
  no. 5, pp. 846-859, 2000.
• 2 A. K. Jain, L. Hong, S. Pankanti, and R.
  Bolle, An identity authentication system using
  fingerprints, Proc. IEEE, vol. 85, pp.
  13651388, Sept. 1997.
• 3 R. O. Duda and P. E. Hart, Pattern
  Classification and Scene Analysis, New York
  Wiley, 1973

I
II
III