Abstract
This paper presents a novel approach of fingerprint image enhancement that relies on detecting the fingerprint ridges as image regions where the second directional derivative of the digital image is positive. A facet model is used in order to approximate the derivatives at each image pixel based on the intensity values of pixels located in a certain neighborhood. We note that the size of this neighborhood has a critical role in achieving accurate enhancement results. Using neighborhoods of various sizes, the proposed algorithm determines several candidate binary representations of the input fingerprint pattern. Subsequently, an output binary ridge-map image is created by selecting image zones, from the available binary image candidates, according to a MAP selection rule. Two public domain collections of fingerprint images are used in order to objectively assess the performance of the proposed fingerprint image enhancement approach.
Highlights
Fingerprints are graphical ridge patterns present on human fingers, which, due to their uniqueness and permanence, are among the most reliable human characteristics that can be used for people identification [1, 2]
We introduced a novel approach to fingerprint image enhancement
Our enhancement algorithm calculates a binary representation of the fingerprint pattern based on the sign of second directional derivative of the digital image
Summary
Fingerprints are graphical ridge patterns present on human fingers, which, due to their uniqueness and permanence, are among the most reliable human characteristics that can be used for people identification [1, 2]. An attempt to enhance the image would be to classify image pixels as ridge or valley pixels by comparing their intensities with a certain threshold intensity value Such a simple point operation approach fails to reduce the noise present in the image because it does not take into consideration the strong correlation that exists between neighborhood intensity values in the fingerprint image pattern. Instead of designing filters tuned on corresponding spatial frequency of each image region, Willis and Myers proposed in [7] to use as a filter directly the magnitude of the Fourier transform of the local image region This magnitude already exhibits most of the qualities required from a properly designed enhancement filter since it has a dominant component at the corresponding ridge orientation and frequency, and on the other hand, due to the noise irregularity it exhibits small other components
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have