24-04-2012, 04:56 PM
Implementation of Gabor Filter for Fingerprint Recognition
[attachment=20736]
Introduction
Fingerprint-based identification is one of the most important biometric technologies which
have drawn a substantial amount of attention recently. Humans have used fingerprints for
personal identification for centuries and the validity of fingerprint identification has been well
established. In fact, fingerprint technology is so common in personal identification that it has
almost become the synonym of biometrics. Fingerprints are believed to be unique across
individuals and across fingers of same individual. Even identical twins having similar DNA,
are believed to have different fingerprints. These observations have led to the increased use
of automatic fingerprint based identification in both civilian and law-enforcement
applications.
Gabor Filter
A Gabor filter is linear filter whose impulse response is defined by a harmonic function multiplied by Gaussian function. The Fourier transform of a Gabor filter‟s impulse response is the convolution of Fourier transform of harmonic function and the Fourier function of Gaussian function. This is the formula of the complex Gabor function:
Filter Design
After the coefficient and the method of filter design has been determine then the designing
process started. This filter used field programmable gate array (FPGA). To design the logic
gate of the filter, it used Verilog high device language (HDL). In designing process, this
project used top-down method of design.
From the method of design defined above, the designing process started with defining the
block diagram of the top level. Then the design went down to the lower level to get the full
architecture of the filter. The design then was transform into the verilog code using Xilinx
design tool.
Convolution between image pixel and coefficient kernel
It is discusses that on the method of convolution between coefficient kernels with the image
data in the memory. It discusses on the flow of which image data will be convolute with the
kernels and the next data to be convolute. This methodology was very important as it will
determine on how the control logic unit need to control the flow of the data and the arithmetic
process. Figure 2 may better explain on the flow of the convolution between image pixel and
coefficient kernels.
Conclusion
In this Report, a novel fingerprint representation technique that uses minutiae extraction and core detection has been presented. Experiments indicate that the computation of the orientation field performs much better than a purely minutiae based matching scheme. Currently, core information is being used to align the image and then trace vectors to the bifurcations; the magnitude of the vectors are use also as characteristics from the image along with the extracted minutiae. The following areas of improvement are also being studied
[attachment=20736]
Introduction
Fingerprint-based identification is one of the most important biometric technologies which
have drawn a substantial amount of attention recently. Humans have used fingerprints for
personal identification for centuries and the validity of fingerprint identification has been well
established. In fact, fingerprint technology is so common in personal identification that it has
almost become the synonym of biometrics. Fingerprints are believed to be unique across
individuals and across fingers of same individual. Even identical twins having similar DNA,
are believed to have different fingerprints. These observations have led to the increased use
of automatic fingerprint based identification in both civilian and law-enforcement
applications.
Gabor Filter
A Gabor filter is linear filter whose impulse response is defined by a harmonic function multiplied by Gaussian function. The Fourier transform of a Gabor filter‟s impulse response is the convolution of Fourier transform of harmonic function and the Fourier function of Gaussian function. This is the formula of the complex Gabor function:
Filter Design
After the coefficient and the method of filter design has been determine then the designing
process started. This filter used field programmable gate array (FPGA). To design the logic
gate of the filter, it used Verilog high device language (HDL). In designing process, this
project used top-down method of design.
From the method of design defined above, the designing process started with defining the
block diagram of the top level. Then the design went down to the lower level to get the full
architecture of the filter. The design then was transform into the verilog code using Xilinx
design tool.
Convolution between image pixel and coefficient kernel
It is discusses that on the method of convolution between coefficient kernels with the image
data in the memory. It discusses on the flow of which image data will be convolute with the
kernels and the next data to be convolute. This methodology was very important as it will
determine on how the control logic unit need to control the flow of the data and the arithmetic
process. Figure 2 may better explain on the flow of the convolution between image pixel and
coefficient kernels.
Conclusion
In this Report, a novel fingerprint representation technique that uses minutiae extraction and core detection has been presented. Experiments indicate that the computation of the orientation field performs much better than a purely minutiae based matching scheme. Currently, core information is being used to align the image and then trace vectors to the bifurcations; the magnitude of the vectors are use also as characteristics from the image along with the extracted minutiae. The following areas of improvement are also being studied