Fingerprint pre-segmentation method based on Mean Shift Algorithm

Author

Liu Xinmei ; Hou Wen ; Yin Junling ; Han Yan

Author_Institution

Nat. Key Lab. of Electron. Test Technol., North Univ. of China, Taiyuan, China

Volume

3

fYear

2011

fDate

15-17 Oct. 2011

Firstpage

1241

Lastpage

1245

Abstract

In the research or application of fingerprint images, people focus on the region where the ridges are seen. The background region of the whole image contains various degrees of noise caused by unevenly pressing of the fingerprint and the contaminated camera lens. This phenomenon calls for the considerations of noise elimination, image enhancement, and pre-segmentation. Mean Shift algorithm is a nonparametric statistics method that searches for the most approximate mode to the sample distribution. The fingerprint pre-segmentation method based on mean shift algorithm was studied and applied in a number of images stored in FVC2004 image Database. The results show that the Mean Shift method is applicable to most images. There are very few results are out of the optimal ones.

Keywords

fingerprint identification; image denoising; image enhancement; image segmentation; nonparametric statistics; visual databases; FVC2004 image database; contaminated camera lens; fingerprint images; fingerprint presegmentation method; image background region; image enhancement; mean shift algorithm; noise elimination; nonparametric statistics method; Estimation; Fingerprint recognition; Image matching; Image segmentation; Kernel; Noise; Vectors; Fingerprint Segmentation; Mean Shift; kernel function;

fLanguage

English

Publisher

ieee

Conference_Titel

Image and Signal Processing (CISP), 2011 4th International Congress on

Conference_Location

Shanghai

Print_ISBN

978-1-4244-9304-3

Type

conf

DOI

10.1109/CISP.2011.6100446

Filename

6100446