Image analysis of degraded laser-luminescent fingerprints

The present work reports an efficient three-stage matching algorithm for fingerprint classification that is presently done by computationally intensive, heuristic techniques. The matching algorithm includes preprocessing by a new transform domain filter, computation of moments as invariant features, and finally, use of a nearest neighbor clustering analysis for fingerprint matching. The transform domain filter involves selective amplification of the spectral band containing the highest energy, and subsequent use of a band-pass filter. The resulting enhanced image is almost noise-free, and shows prominent features in the fingerprints that cannot be extracted by other conventional enhancement techniques. The moments of the spectrally enhanced fingerprint image provide invariant features. The final classification of the fingerprint is performed by a nearest neighbor clustering analysis. The first four moment invariants are employed for matching, since higher order moment invariants are very sensitive to noise.