Fingerprint image enhancement based on nonlinear anisotropic reverse-diffusion equations

Author

Hao, Yufeng ; Yuan, Chunwei

Author_Institution

Dept. of Biomedical Eng., Southeast Univ., Nanjing, China

Volume

1

fYear

2004

fDate

1-5 Sept. 2004

Firstpage

1601

Lastpage

1604

Abstract

We present an approach based on nonlinear anisotropic reverse diffusion equations (NARDE) to enhance noisy fingerprint images. After studying the nonlinear diffusion equations to denoised images and nonlinear reverse diffusion process in nature, a new scheme is advanced with new selection criteria of optimal stopping time. Experiments were carried out and showed the high efficiency of the noise removing sharpening scheme and the role of selection criteria on noisy fingerprint images.

Keywords

image denoising; image enhancement; medical image processing; denoised images; image enhancement; noisy fingerprint images; nonlinear anisotropic reverse-diffusion equations; optimal stopping time; Anisotropic magnetoresistance; Diffusion processes; Filtering; Fingerprint recognition; Image enhancement; Image matching; Image processing; Nonlinear equations; Partial differential equations; Wiener filter; Nonlinear anisotropic reverse-diffusion equations; fingerprint images enhancement;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2004. IEMBS '04. 26th Annual International Conference of the IEEE

Conference_Location

San Francisco, CA

Print_ISBN

0-7803-8439-3

Type

conf

DOI

10.1109/IEMBS.2004.1403486

Filename

1403486