Algorithm CSPN

Organization: Centre for Signal Processing, Nanyang Technological University (Singapore)

CSPN is a minutiae-based fingerprint verification algorithm that extracts minutiae by adaptive tracing the gray level ridges and matches minutia sets based on the minutia local and global structures. In addition, CSPN employs some unique techniques such as image segmentation adaptive to the image contrast and area, selective image processing, higher order statistics and ridge associated post-processing. More detailed information can be found in the following publications:

[1] X. Jiang et al., "Minutiae Extraction by Adaptive Tracing the Gray Level Ridge of the Fingerprint Image", IEEE ICIP'99, Japan, 1999.
[2] X. Jiang et al., "Fingerprint Image Processing for Automatic Verification", ICICS'99, Singapore, 1999.
[3] X. Jiang, "Fundamental Frequency Estimation By Higher Order Spectrum", IEEE ICASSP'00, Turkey, 2000.
[4] X. Jiang et al., "Fingerprint Minutiae Matching Based on the Local And Global Structures", ICPR'00, Spain, 2000.
[5] X. Jiang, "Fingerprint Image Ridge Frequency Estimation by Higher Order Spectrum", IEEE ICIP'00, Canada, 2000.
[6] X. Jiang et al., "Detecting the Fingerprint Minutiae by Adaptive Tracing the Gray Level Ridge" Pattern recognition, to be published.

CSPN was developed mainly for portable stand-alone system where the algorithm speed and capability of recognizing partial fingerprints are emphasized for low cost processors and small fingerprint sensors. In addition, the algorithm is rotation and translation invariant. For more information please contact Dr. Xudong Jiang, email:

Detailed results on DB1
Detailed results on DB2
Detailed results on DB3
Detailed results on DB4

Comparison chart

© 2000-2001 BioLab - University of Bologna For information or suggestions: