Share Email Print
cover

Proceedings Paper

Fingerprint image segmentation based on linear classifier
Author(s): Chunxiao Ren; Yilong Yin; Jun Ma
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Fingerprint segmentation is an important step in automatic fingerprint identification fields. This paper discusses a new method, which is based on a linear classifier, to enhance the performance of fingerprint image segmentation. The novel linear classifier is a label box that is employed to establish a model and deal with fingerprint image quickly and accurately. In order to evaluate the performance of the new method in comparison to the methods based on other linear and nonlinear classifiers, experiments are performed on FVC2000 DB2. The experimental results show the proposed method is able to provide more accurate high-resolution segmentation results than those of previously known ones because only 0.80% of the pixels are misclassified by the method, while the nonlinear classifier, quadric surface classifier, misclassifies 0.97% of the pixels.

Paper Details

Date Published: 14 November 2007
PDF: 6 pages
Proc. SPIE 6790, MIPPR 2007: Remote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications, 67905J (14 November 2007); doi: 10.1117/12.774816
Show Author Affiliations
Chunxiao Ren, Shandong Univ. (China)
Yilong Yin, Shandong Univ. (China)
Jun Ma, Shandong Univ. (China)


Published in SPIE Proceedings Vol. 6790:
MIPPR 2007: Remote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications

© SPIE. Terms of Use
Back to Top