YOU ARE DOWNLOADING DOCUMENT

Please tick the box to continue:

Transcript
Page 1: Fingerprint Analysis (part 1) Pavel Mr ázek

Fingerprint Analysis(part 1)

Pavel Mrázek

Page 2: Fingerprint Analysis (part 1) Pavel Mr ázek

What is fingerprint• Ridges, valleys

• Singular points– Core– Delta

• Orientation field• Ridge frequency

Page 3: Fingerprint Analysis (part 1) Pavel Mr ázek

Fingerprint classes

Page 4: Fingerprint Analysis (part 1) Pavel Mr ázek

Small scale: Minutia• 150 types in theory• 7 used by human experts• 2 types for the machine:

– Ending– Bifurcation

Page 5: Fingerprint Analysis (part 1) Pavel Mr ázek

Minutia examples

Page 6: Fingerprint Analysis (part 1) Pavel Mr ázek

SensingTraditional (off line): rolled ink impression

+ paper scan

• Plus: big area

• Minuses: – Inconvenient– Distortion– Too much/little ink

Page 7: Fingerprint Analysis (part 1) Pavel Mr ázek

SensingOptical sensors

Page 8: Fingerprint Analysis (part 1) Pavel Mr ázek

SensingOptical sensors

• Good: large area possible, good image quality, contactless scanning available

• Bad: size

Page 9: Fingerprint Analysis (part 1) Pavel Mr ázek

SensingSilicon sensors

• Capacitive

• Electric field

• Thermal

Page 10: Fingerprint Analysis (part 1) Pavel Mr ázek

SensingSilicon sensors

• Good image quality, small form factor • Price proportional to size

Page 11: Fingerprint Analysis (part 1) Pavel Mr ázek

SensingSilicon sensors

• Area

• Swipe

Page 12: Fingerprint Analysis (part 1) Pavel Mr ázek

Fingerprint types

Page 13: Fingerprint Analysis (part 1) Pavel Mr ázek

Minutia detection overview

Page 14: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation fieldOrientation field (or ridge flow)

estimation: • Crucial step before

image enhancement

• Various methods:– Gradient-based– Gabor filters– FFT

Page 15: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation estimation• Gradient direction

– local characteristics– same ridge orientation,

opposite gradients– more global view needed

• Classical solution: Structure tensor(second moment matrix, interest operator)– start from a 2x2 matrix

(positive semidefinite)– safe to average information

Page 16: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation estimationStructure tensor• Local:• Larger scale: average componentwise

(Gaussian window, linear/nonlinear smoothing)

• 2 nonnegative eigenvalues– both small: backgroung / low contrast– one big, one small: regular ridge area– both big: multiple orientations

(core, delta, scar)

Page 17: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation estimationStructure tensor• system of 2 orthogonal eigenvectors• shows dominant direction

Page 18: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation estimation

Page 19: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation estimation

Page 20: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation estimation• Problematic images

• Solution– Enforce smoothness– Use prior knowledge

Page 21: Fingerprint Analysis (part 1) Pavel Mr ázek

Orientation model

Page 22: Fingerprint Analysis (part 1) Pavel Mr ázek

References• Maltoni et al.: Handbook of Fingerprint Recognition. Springer

2003.• Maltoni. A tutorial on fingerprint recognition. In LNCS 3161,

Springer 2005.• Hong, Wan, Jain. Fingerprint image enhancement: algorithm

and performance evaluation. IEEE PAMI 1998.• Zhou, Gu. A model-based method for the computation of

fingerprints’ orientation field. IEEE TIP 2004.• Weickert. Coherence enhancing shock filters. DAGM 2003.

• Contact: mrazekp -at- cmp.felk.cvut.cz


Related Documents