Fingerprint Recognition Using Directional Information In Wavelet Transform Domain

ABSTRACT
In this paper, a fingerprint recognition algorithm is suggested. The algorithm is developed based on the wavelet transform, and the dominant local orientation which is derived from the coherence and the gradient of Gaussian. By using the wavelet transform, the algorithm does not require conventional preprocessing procedures such as smoothing, binarization, thinning and restoration. Computer simulation results show that when the rate of incorrect recognition of two different fingerprints as identical fingerprints - is held at O.O%, the rate of incorrect recognition of two identical fingerprints as different ones - turns out as 2.5% in real time.


Introduction
The objective of this paper is to develop a real-time fingerprint recognition algorithm. For the recognition, proposed algorithm uses information achieved based on the directions along the ridge curves. In order to find the ridge curves, the algorithm employs the wavelet transform, and the dominant local orientation which is derived from the gradient of Gaussian and coherence. By using the wavelet transform, the algorithm could avoid many of conventional preprocessing procedures such as smoothing, binarization, thinning and restoration. For the recognition, fingerprint images are compared in three different wavelet transformed bands; one that represents the low pass-filtered original image sub sampled to half (LL band), another band that shows characteristics contained in vertical direction (LH band), and third one that contains horizontal directional characteristics (HL band). The proposed algorithm is implemented on aMATLAB 6.5environment. In first
Section, the wavelet transform and its directional characteristics that are contained in each of wavelet bands are illustrated. Section second discusses the procedures adopted to extract the feature vector contained in fingerprint, namely, dominant local orientation. Third Section explains the way of determining the center point which is used as a reference position to match input image with sample image. This step is crucial for successful fingerprint recognition. Fourth Section reports our simulation results. Finally, conclusions of this paper are made in last Section.

Download this Seminar Report

Password: be

Share on Google Plus

About Unknown

This is a short description in the author block about the author. You edit it by entering text in the "Biographical Info" field in the user admin panel.