doi:10.1016/S0031-3203(02)00322-9 
Copyright © 2003 Pattern Recognition Society. Published by Elsevier Science B.V.
Learning fingerprint minutiae location and type*1
a Digital Persona Inc., 805 Veterans Blvd., Suite 301, Redwood City, CA 94063, USA
b Department of Computer Science and Engineering, Michigan State University, East Lansing, MI 48824, USA
c IBM T. J. Watson Research Center, Yorktown Heights, NY 10598, USA
Received 8 March 2002;
revised 2 October 2002;
accepted 2 October 2002. ;
Available online 15 February 2003.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
For simplicity of pattern recognition system design, a sequential approach consisting of sensing, feature extraction and classification/matching is conventionally adopted, where each stage transforms its input relatively independently. In practice, the interaction between these modules is limited. Some of the errors in this end-to-end sequential processing can be eliminated, especially for the feature extraction stage, by revisiting the input pattern. We propose a feedforward of the original grayscale image data to a feature (minutiae) verification stage in the context of a minutiae-based fingerprint verification system. This minutiae verification stage is based on reexamining the grayscale profile in a detected minutia's spatial neighborhood in the sensed image. We also show that a feature refinement (minutiae classification) stage that assigns one of two class labels to each detected minutia (ridge ending and ridge bifurcation) can improve the matching accuracy by 
1% and when combined with the proposed minutiae verification stage, the matching accuracy can be improved by 3.2% on our fingerprint database.
Author Keywords: Fingerprint matching; Feature extraction; Feedforward; Minutia verification; Minutia classification; Gabor filters; Learning vector quantization
Fig. 1. A general pattern recognition system with proposed feature refinement stage and a feedforward of original image data for feature verification.
Fig. 2. Examples of fingerprint minutiae: ridge endings (□) and bifurcations (○).
Fig. 3. Various stages in a typical minutiae extraction algorithm [1].
Fig. 4. Examples of a ridge bifurcation and a ridge ending in a thinned fingerprint image. In (a) and (b), all the pixels that reside on the ridge have two 8-connected neighbors. In (a), the pixel with three neighbors is a ridge bifurcation and in (b), the pixel with only one neighbor is a ridge ending.
Fig. 5. Sample images from our database with varying quality index (QI). No false minutiae were detected in (a), 7 in (b), and 27 in (c) by the automatic minutiae detection algorithm [1].
Fig. 6. Gabor filter (orientation=0°, mask size=33×33, f=0.1, δx=4.0, δy=4.0).
Fig. 7. Stages in feature extraction for minutiae verification. A true minutia location and the associated direction is marked in (a); the 64×64 area centered at the minutia location and oriented along the minutia direction is also shown. In (b), the grayscale values in the 64×64 neighborhood are shown. The output of 0°-oriented Gabor filter applied to (b) is shown in (c); note the problems at the boundary due to convolution. The central 32×32 region is extracted and shown in (d). (e) shows the same 32×32 region as in (d) but the grayscale range has been scaled to integers between 0 and 7.
Fig. 8. Two examples of images in the GT database. The ground truth minutiae provided by an expert are marked on the image.
Fig. 9. Distribution of quality of fingerprints in the GT database.
Fig. 10. Examples of grayscale profiles in the neighborhood of (a) minutiae and (b) nonminutiae. These 32×32 subimages that are scaled to 8 grayscales, are used for training a LVQ classifier.
Fig. 11. ROC for fingerprint matching when both minutiae classification and verification are used.
Fig. 12. Minutiae detection and classification: (a) minutiae detection using the algorithm in [1] without pruning; (b) result of classifying minutiae, minutia bifurcations are marked with black and endings are marked with white; (c) result of minutiae verification; (d) the results of minutiae pruning and no minutiae classification are shown for a comparison. Note that visually, the results of minutiae verification proposed in this paper are better than the rather ad-hoc minutiae pruning used in [1].
Abstract
Purchase PDF (2701 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (6)
Purchase PDF (896 K)
