Abstract
Segmentation of fingerprint image is necessary to reduce the size of the input data, eliminating undesired background, which is the noisy and smudged area in favor of the central part of the fingerprint. In this paper, an algorithm for the segmentation which uses two stages coarse to fine approach is presented. The coarse segmentation will be performed at first using the orientation certainty values that derived from the blockwise directional field of the fingerprint image. The coarse segmented image will be carry on to the second stage which consist Fourier based enhancement and adaptive thresholding. Orientation certainty values of the resultant binarized image are calculated once again to perform the fine segmentation. Finally, binary image processing is applied as postprocessing to further reduce the segmentation error. Visual inspection shows that the proposed method produce accurate segmentations result. The algorithm is also evaluated by counting the number of false and missed detected center points and compare with the fingerprint image which have no segmentation and with the proposed method without postprocessing. Experiments show that the proposed segmentation method perform well than others.
| Original language | English |
|---|---|
| Pages (from-to) | 624-632 |
| Number of pages | 9 |
| Journal | Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science) |
| Volume | 2903 |
| Publication status | Published - 2003 Dec 1 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Hardware and Architecture
- Engineering(all)
- Computer Science(all)
- Theoretical Computer Science
Cite this
}
Fingerprint images segmentation using two stages coarse to fine discrimination technique. / Ong, T. S.; Andrew, T. B J; David, N. C L; Sek, Y. W.
In: Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science), Vol. 2903, 01.12.2003, p. 624-632.Research output: Contribution to journal › Conference article
TY - JOUR
T1 - Fingerprint images segmentation using two stages coarse to fine discrimination technique
AU - Ong, T. S.
AU - Andrew, T. B J
AU - David, N. C L
AU - Sek, Y. W.
PY - 2003/12/1
Y1 - 2003/12/1
N2 - Segmentation of fingerprint image is necessary to reduce the size of the input data, eliminating undesired background, which is the noisy and smudged area in favor of the central part of the fingerprint. In this paper, an algorithm for the segmentation which uses two stages coarse to fine approach is presented. The coarse segmentation will be performed at first using the orientation certainty values that derived from the blockwise directional field of the fingerprint image. The coarse segmented image will be carry on to the second stage which consist Fourier based enhancement and adaptive thresholding. Orientation certainty values of the resultant binarized image are calculated once again to perform the fine segmentation. Finally, binary image processing is applied as postprocessing to further reduce the segmentation error. Visual inspection shows that the proposed method produce accurate segmentations result. The algorithm is also evaluated by counting the number of false and missed detected center points and compare with the fingerprint image which have no segmentation and with the proposed method without postprocessing. Experiments show that the proposed segmentation method perform well than others.
AB - Segmentation of fingerprint image is necessary to reduce the size of the input data, eliminating undesired background, which is the noisy and smudged area in favor of the central part of the fingerprint. In this paper, an algorithm for the segmentation which uses two stages coarse to fine approach is presented. The coarse segmentation will be performed at first using the orientation certainty values that derived from the blockwise directional field of the fingerprint image. The coarse segmented image will be carry on to the second stage which consist Fourier based enhancement and adaptive thresholding. Orientation certainty values of the resultant binarized image are calculated once again to perform the fine segmentation. Finally, binary image processing is applied as postprocessing to further reduce the segmentation error. Visual inspection shows that the proposed method produce accurate segmentations result. The algorithm is also evaluated by counting the number of false and missed detected center points and compare with the fingerprint image which have no segmentation and with the proposed method without postprocessing. Experiments show that the proposed segmentation method perform well than others.
UR - http://www.scopus.com/inward/record.url?scp=0347586940&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0347586940&partnerID=8YFLogxK
M3 - Conference article
VL - 2903
SP - 624
EP - 632
JO - Lecture Notes in Computer Science
JF - Lecture Notes in Computer Science
SN - 0302-9743
ER -