Recent Fingerprint Image Enhancement Using Wavelet Transforms -- from Wolfram Library Archive

Products

Consulting & Solutions

Learning & Support

Company

Wolfram|Alpha

Enable JavaScript to interact with content and submit forms on Wolfram websites. Learn how

Title

Recent Fingerprint Image Enhancement Using Wavelet Transforms

Authors

S. Y. Lee

Organization:

Inje University

Department:

Mathematica Technical and Training Center and School of Computer Aided Science

J. W. Hwang

Organization:

Inje University

Department:

Mathematica Technical and Training Center and School of Computer Aided Science

M. G. Seo

Organization:

Inje University

Department:

Mathematica Technical and Training Center and School of Computer Aided Science

Y. S. Myung

Organization:

Inje University

Department:

Mathematica Technical and Training Center and School of Computer Aided Science

Conference

2002 Applications of Computer Algebra Conference (ACA '2002)

Conference location

Volos, Greece

Description

It is very important to extract distinct and local characteristics (minutiae) from fingerprint images to construct an automatic fingerprint identification and recognition system. The distinct characteristics are usually classified as arch, right-delta, left-delta, whorl, and transient. On the other hand, the minutiae in the fingerprint pattern are local discontinuities such as terminations and bifurcation. First we review a binarization-based approach composed of four steps: local histogram equalization ,median filtering, binarization and thinning, and post processing to obtain correct ridgeline flows. In this work, we use the wavelet transforms and wavelet packet transforms to obtain those of distinct and local characteristics in the fingerprint images. The wavelet scalar quantization (WSQ) which was originally adopted by the U.S. government FBI is focused on the compression ratio. As it stands, the wavelettransform including the wavelet packet transform gives us powerful techniques for edge detection, edge enhancement and image identification and recognition. We use the MS Visual C++ in a binarization-based approach. Finally we emphasize that two Mathematica packages of Wavelet Explorer and Digital Image Processing are mainly used for our purpose.

Subjects

	Wolfram Technology > Application Packages > Wolfram Research Applications > Digital Image Processing
	Wolfram Technology > Application Packages > Wolfram Research Applications > Wavelet Explorer

Keywords

automatic fingerprint identification, recognition system, arch, right-delta, left-delta, whorl, transient, wavelet packet transform, Wavelet Explorer, Digital Image Processing

WOLFRAM