US 4,752,966 A · Fingerprint identification system

Abstract

An automatic fingerprint enrollment and identification technique is provided for automatically processing an input image represented by pixels having either a one or zero value. The automatic processing involves a series of steps in which the initial processing is a rough screening technique for eliminating a reasonable portion of the noise while passing through almost all of the minutia. As the steps proceed, the processing becomes more complex to provide a tighter screening out of noise while passing through as much of the minutia as possible. The later steps by operating on relatively fewer minutia or anomaly points that the earlier steps can be economical though more complex so as to be more discriminate than the earlier steps. The processing provides an optimum trade off between enrolling as large a proportion of the population as possible while minimizing the number of false identification errors.

Claims24

Dependent claims

In the machine method of identifying fingerprint minutia from a binary field of pixels having first and second binary signal values representing the fingerprint, the improvement comprising the steps of:

providing a set of binary value signals representing pixels of a fingerprint image generated from an optical scan of a fingerprint,
scanning each of a predetermined set of anomaly templates across the field of said pixels represented by said set of signals,
identifying as an anomaly each match of any one of said anomaly templates against a subfield of said pixels to provide a set of anomalies, each of said anomaly templates being of sufficient size to provide an indication of anomaly type, anomaly curvature and anomaly thickness,
testing the zone of pixels of contrasting value around each of said identified anomalies for thickness uniformity,
identifying as a minutia candidate those anomalies where the surrounding zone exhibits a uniformity of thickness in accordance with a predetermined criteria to provide a set of minutia candidates,
erecting a predetermined window around each of said set of minutia candidates,
identifying each of said anomalies in said window,
comparing the relationship between the set of said anomalies in said window against a predetermined fingerprint noise characteristic to provide a determination of the presence of fingerprint noise, and
rejecting the minutia candidate of said window where there has been a determination of fingerprint noise, to thereby provide a reduced set of minutia candidates.

6 more elements

The machine method of enrolling a set of fingerprint minutia from a binary field of pixels having first and second binary signal values representing the fingerprint comprising the steps of:

generating an initial set of fingerprint minutia from an initial optical scan of a fingerprint to provide a field of minutia,
identifying the angle within said field of each of said minutia by comparing each of a set of angle identification minutia templates against each of said minutia,
repeating said steps of generating and identifying on repeated applications of the fingerprint that is subject to the initial scan, to provide a separate set of minutia on each of said subsequent applications of the fingerprint,
erecting a predetermined subfield zone centered upon each of said minutia from said initial scan,
correlating the position of said initial scan minutia against each of said subsequent scan minutia within the corresponding subfield zone,
designating as a minutia for the subsequent scan involved only those minutia whose position corresponds within predetermined limits with the position of the initial scan minutia, and
retaining in a final enrollment file only those minutia which are identified in a predetermined number of said scans to provide a stable, limited enrollment file.

4 more elements

The machine method of enrolling a set of fingerprint minutia from a binary field of pixels having first and second binary signal values representing the fingerprint and of verifying correspondence to a binary field of pixels generated from a subsequent application of the fingerprint, comprising the steps of:

generating an initial set of fingerprint minutia from an initial optical scan of a fingerprint to provide a field of minutia,
repeatedly generating sets of fingerprint minutia from subsequent optical scans of repeated applications of the fingerprint that is subject to the initial scan,
retaining in a final enrollment file only those minutia which are identified in a predetermined number of said scan to provide a stable, limited enrollment file,
verifying the correspondence between minutia from a binary field of pixels representing a subsequent application of an input fingerprint and said minutia from said final enrollment file, said step of verifying employing analytical techniques and criteria substantially identical to those employed in said steps of repeatedly generating sets of minutia from repeated applications of the fingerprint during enrollment,
whereby the biases in verification are matched by and at least partially compensated by the comparable biases in the repeated scans and repeated steps of generating minutia during enrollment.

2 more elements

Dependencies are inferred from the claim text.

Fingerprint identification system

Life of this patent

Fingerprint identification system

Life of this patent

Abstract

Claims24