Scribd
Upload
75 views

Fingerprint Matching Algorithm

This document summarizes the steps in a fingerprint matching algorithm. It involves feature enhancement of the fingerprint image through segmentation, normalization, and orientation estimation. Ridges are enhanced through Gabor filtering and the image is binarized and thinned. Minutiae points are then extracted using a crossing number concept and classified. For matching, each minutiae point's location, orientation, and type are saved in a matrix. Template and input minutiae are used as reference points to convert remaining points to polar coordinates for comparison.

Uploaded by

Jitendra Naik
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
75 views

Fingerprint Matching Algorithm

This document summarizes the steps in a fingerprint matching algorithm. It involves feature enhancement of the fingerprint image through segmentation, normalization, and orientation estimation. Ridges are enhanced through Gabor filtering and the image is binarized and thinned. Minutiae points are then extracted using a crossing number concept and classified. For matching, each minutiae point's location, orientation, and type are saved in a matrix. Template and input minutiae are used as reference points to convert remaining points to polar coordinates for comparison.

Uploaded by

Jitendra Naik
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Fingerprint Matching Algorithm

Step One - Feature Enhancement:

Segmentation:

Segmentation is the process of separating the foreground regions in the image from
the background regions. The foreground regions correspond to the clear fingerprint
area containing the ridges and valleys and have a very high variance. The
background corresponds to the regions outside the
borders of the fingerprint area, which causes extraction of noisy and false
minutiae and generally exhibit a very low grey-scale variance value.

Hence, a method based on variance thresholding can be used i.e. if the


grayscalevariance of block is less than the globalthreshold, then the block is
assigned to be a background region. The grey-level variance for a block of sizeW
X W is de_ned as: whereV (k) is the variance for block k, I(i; j) is the grey-level
value at pixel (i; j), and M(k) is the mean grey-level value for the block k.

Normalisation:

Normalisation is used to standardise the intensity values in an image by


adjusting the range of grey-level values so that it lies within a desired range of
values.
Let I(i;j) represent the grey-level value at pixel (i; j), and N(i; j) represent the
normalised grey-level value at pixel (i; j). The normalised image is read as where
M and V are the estimated meanand variance of I(i; j),
respectively, and M0 and V0 are the desired mean and variance values,
respectively.

Orientation estimation :
1
The orientation field of a fingerprint image defines the local orientation of the
ridges contained in the fingerprint. Pixel-wise scheme is used to produce a finer
and accurate estimation of the orientation field.
Steps are:
1. A block of size W XW is centred at pixel (i; j) in the normalised _ngerprint
image.
2.For each pixel in the block, compute the gradients dx(i; j) and dy(i; j), which are
the gradient magnitudes in the x and y directions, respectively. The local
orientation at pixel (i; j) can then be estimated using the following equations:

where Ө(I,j)=is the least square estimate of the local orientation at the block centre
of that pixel(i; j).

3.Smooth the orientation field in a local neighbourhood using a Gaussian filter.


4.The final smoothed orientation field O at pixel (i; j) is defined as:

Ridge frequency estimation

2
Firstly divide the image into blocks of size W x W. The next step is to project the
grey level values of all the pixels located inside each block along a direction
orthogonal to the local ridge orientation. This projection forms an almost
sinusoidal-shape wave with the local minimum
points corresponding to the ridges in the fingerprint.

Gabor filtering

The ridge orientation and ridge frequency parameters are used to construct the
even-symmetric Gabor filterhaving frequency-selective and orientation-selective
properties. These properties allow the filter to be tuned to give maximal response
to ridges at a specific orientation and frequency in the fingerprint image.The even-
symmetric Gabor filter is the real part of the Gabor function, which is given by a
cosine wave modulated by a Gaussian filter. The enhanced image E is obtained as
follows:

Binarisation

Binarisation is the process that converts a greylevelimage into a binary image


improvingthe contrast between the ridges and valleys in a _ngerprint image, and
consequently facilitates the extraction of minutiae.The binarisation process
involves examining the grey-level value of each
pixel in the enhanced image, and, if the value is greater than the global threshold,
then the pixel value is set to a binary value one; otherwise, it is set to zero.

3
Thinning

Thinning is a morphological operation that successively erodes away the


foreground pixels until they are one pixel wide.Thinning algorithm 'parallel
thinning with two-subiteration'is accessible in MATLAB via the `thin' operation
under the bwmorphfunction.Each subiteration begins by
examining the neighbourhood of each pixel in the binary image, and based on a
particular set of pixel-deletion criteria, it checks whether the pixel can be deleted
or not. These subiterations continue until no more pixels can be deleted.This
skeleton image is then used in the subsequent extraction of minutiae

Step two- Minutiae extraction:

After a fingerprint image has been enhanced, the next step is to extract the
minutiae from the enhanced image. Following the extraction of minutiae, a final
image postprocessing stage is performed to eliminate false minutiae. Crossing
Number (CN) concept is used for minutiae extraction. A window of 3 x 3 is used to
to scan across the thinned image to compute a crossing
no(CN).

Where Pi is the pixel value in the neighbourhood of P. For a pixel P, its eight
neighbouring pixels are scanned in an anti-clockwise direction as follows:
Pixel can then be classified according to the property of its CN value as follows:

Step three- Minutiae Matching:


To compare the input _ngerprint data with the template data minutiae matching is
used.For each extracted minutiae point, the following information is recorded in
matrix form.
No. of rows : Number of minutiae points
4
No. of columns: 4
Column 1: row index of each minutiae point.
Column 2: column index of each minutiae point.
Column 3: orientation angle of each minutiaepoint.
Column 4: type of minutia(1=ridge ending,2=continuing ridge point etc.) Each
minutiae point is compared with template minutiae point. Template and input
minutiae are selected as ref. points for their respective data sets.The ref points are
used to convert the remaining data points to polar co-ordinates

5
Rotate values (k,m) represents the di_erences between the orientation angle of Tk
and Im. Tk and Imrepresents the extracted data in all the columns of row k row m
in the template and input matrices, respectively.

You might also like