Fingerprint image segmentation based on quadric surface model

Authors:
Yilong Yin;Yanrong Wang;Xiukun Yang
Affiliations:
Computer Department, Shandong University, Jinan, China;Computer Department, Shandong University, Jinan, China;Identix Incorporated, Jersey City, NJ
Venue:
AVBPA'05 Proceedings of the 5th international conference on Audio- and Video-Based Biometric Person Authentication
Year:
2005

Citing 4
Cited 2

Segmentation of fingerprint images using the directional image

Pattern Recognition
Segmentation of fingerprint images—a composite method

Pattern Recognition
Systematic Methods for the Computation of the Directional Fields and Singular Points of Fingerprints

IEEE Transactions on Pattern Analysis and Machine Intelligence
Segmentation of fingerprint images using linear classifier

EURASIP Journal on Applied Signal Processing

Two steps for fingerprint segmentation

Image and Vision Computing
K-means based fingerprint segmentation with sensor interoperability

EURASIP Journal on Advances in Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

It is essential to segment fingerprint image from background effectively, which could improve image processing speed and fingerprint recognition accuracy. This paper proposes a novel fingerprint segmentation method at pixel level based on quadric surface model. Three parameters, Coherence, Mean and Variance of each pixel are extracted and spatial distribution model of fingerprint pixels is acquired and analyzed. Our study indicates that the performance of fingerprint image segmentation with a linear classifier is very limited. To deal with this problem, we develop a quadric surface formula for fingerprint image segmentation and acquire coefficients of the quadric surface formula using BP neural network trained on sample images. In order to evaluate the performance of our proposed method in comparison to linear classifiers, experiments are performed on public database “FVC2000 DB2”. Experimental result indicates that the proposed model can reduce pixel misclassification rate to 0.53%, which is significantly better than the linear classifier's misclassification rate of 6.8%.