Fingerprint Recognition Using Matlab 254b6e

MAJOR PROJECT PRESENTATION 1

ON

PRESENTED BY :-

ABHISHEK DE & ADITYA GOENKA B.TECH 8th SEMESTER DEPARTMENT OF ELECTRONICS & COMMUNICATION INSTITUTE OF ENGINEERING &TECHNOLOGY, ALWAR

GUIDED BY :-

MR.RAKESH KUMAR SHARMA HEAD OF DEPARTMENT ELECTRONICS & COMMUNICATION ENGINEERING

Fingerprint Recognition 2

Outline: 

Introduction • •



Project Scope Fingerprint Research Background

Algorithm • • •

Overview of Approach Detailed Design Conclusion

WHAT IS FINGERPRINT??? 3



 

Feature pattern of one finger Composed of many ridges and furrows Fingerprint recognition is done using abnormal points on these ridges, called as ‘minutia’

FINGERPRINT RECOGNITION 4

Right Loop

Left Loop

Delta Pore

Whorl

Arch

Tented Arch

WHY FINGERPRINT MATCHING???

5



 

Unique Scientifically Reliable Permanent

Termination

Bifurcation

Algorithm Level Design 6

Preprocessing Minutia extraction Post-processing

•Image Enhancement •Image Binarization •Image Segmentation •Thinning

•Pixel-to-pixel matching

STEPS INVOLVED IN FINGERPRINT MATCHING 7

8

LOADING ORIGINAL FINGERPRINT

HISTOGRAM EQUALIZATION 9

ORIGINAL IMAGE

HISTOGRAM EQUALIZED IMAGE

HISTOGRAM PLOTS OF IMAGES 10

HISTOGRAM OF ORIGINAL IMAGE

HISTOGRAM OF EQUALIZED IMAGE

ENHANCEMENT USING FFT 11

•

The enhanced image after FFT has the improvements to connect some falsely broken points on ridges and to remove some spurious connections between ridges

•

Formula for the calculation using Fast Fourier Transform(FFT) is :-

BINARIZATION 12

•

Binarization is to transform the 8-bit Gray fingerprint image to a 1-bit image with 0-value for ridges and 1value for furrows.

•

After the operation, ridges in the fingerprint are highlighted with black color while furrows are white. BINARIZED IMAGE

IMAGE SEGMENTATION 13

•

The main purpose behind image segmentation is to obtain the Region of Interest (ROI)

•

To extract the ROI, a two-step method is used :STAGE 1 - Block direction estimation STAGE 2 - ROI extraction by Morphological operations

RIDGE DIRECTION 14

•

•

Gradient values along xdirection (gx) and y-direction (gy) for each pixel of the block is calculated using ‘Sobel Filter’ Tangent value of the block direction is estimated using the following formula :tg2 = 2sin cos /(cos2 -sin2 ) DIRECTION MAP

ROI EXTRACTION USING MORPHOLOGICAL OPERATION 15

Original image

After close operation

After open operation

ROI + Bound

SEGMENTED IMAGE 16

Binarized Image

ROI Extracted Image

17

THINNING OF ROI EXTRACTED IMAGE •

Ridge Thinning is to eliminate the redundant pixels of ridges till the ridges are just one pixel wide.

•

The thinned ridge map is then filtered by other three Morphological operations to remove some H breaks, isolated points and spikes.

Thinned image

PIXEL TO PIXEL MATCHING 18

•

The images after thinning of minutia of two different fingerprints are matched by pixel-to-pixel comparison of the two images.

•

If minimum 80 % of the pixels match, then fingerprints are approximately matched.

References 19

[1] Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing”,ISBN978-81-203-3640-7, by Prentice Hall Publication, Third Edition, 2008. [2] Rafael C. Gonzalez, Richard E. Woods,Steven L. Eddins, “Digital Image Processing Using Matlab”,ISBN-978-81-7758-898-9, by Prentice Hall Publication, Fourth Edition, 2008. [3] Jain, A.K., Hong, L., and Bolle, R.(1997), “On-Line Fingerprint Verification,” IEEE Trans. On Pattern Anal and Machine Intell, 19(4), pp. 302-314.

Conclusion 20



All the objectives of the software have been accomplished and the main aim of fingerprint recognition with image processing techniques have been achieved.



Application of fingerprint recognition techniques are widely used in forensic and biometric diagnosis.



Plethora of research related activities have also been focussed.

21

QUESTIONS?????

22

THANK YOU !!!!!!