Longest line method and apparatus for fingerprint alignment

• US 5,717,777 A
• Filed: 01/11/1996
• Issued: 02/10/1998
• Est. Priority Date: 01/11/1996
• Status: Expired due to Term

• Current Assignee: Activcard Ireland, Limited (Assa Abloy AB)
• Original Assignee: DEW Engineering and Development Limited (Trelleborg AB)
• Inventors: Hamid, Laurence, Wong, John Douglas Evan, Borza, Michael Andrew
• Primary Examiner(s): Boudreau, Leo
• Assistant Examiner(s): Werner, Brian P.
• Application Number:
  US08/585,349
• Time in Patent Office: 761 Days
• Field of Search: 382/116, 382/124, 382/125, 382/126, 382/127, 283/68, 283/69, 356/71, 340/825.34
• US Class Current: 382/124
• CPC Class Codes: G06V 40/1347 Preprocessing; Feature extraction
• CPC Class Codes: G06V 40/1365 Matching; Classification

Abstract

A non-minutia method and apparatus for comparing a reference fingerprint image to a candidate fingerprint image are disclosed. During enrollment of a reference image, a unique reference set of data is generated in a reference coordinate system by placing a group of line segments on the reference image. A collection of constraints is used for defining the subspace for each line of the reference group. A similar set of candidate line segments is constructed on the candidate fingerprint image in a candidate coordinate system. A candidate set of data is determined and then compared with the reference set of data using translation and rotation of the coordinate systems. Tolerances of the comparison are also provided for.

Claims

1. A method for storing a set of reference metrics for a fingerprint in a reference storage, comprising the steps of:

scanning said fingerprint to obtain a reference image in a reference coordinate system;

selecting a first and a second constraint and defining a space on said fingerprint image;

defining a first subspace on said space and generating a first line segment within said first subspace in accordance with a third constraint;

measuring a first reference set of metrics for said first line segment, determining a fourth constraint using said first constraint and said first reference set of metrics, and defining a second subspace on said space according to said fourth constraint;

generating a second line segment within said second subspace according to said third constraint;

measuring a second reference set of metrics for said second line segment, determining a fifth constraint from said second reference sets of metrics and said first constraint, and defining a third and a fourth subspace on said space according to said fourth and fifth constraints;

generating a third line segment within said third subspace according to said second, fourth, and fifth constraints, and said first set of reference metrics;

generating a fourth line segment within said fourth subspace according to said second, fourth, and fifth constraints, and said second set of reference metrics; and

storing said first to fourth reference sets of metrics in said reference storage.

2. A method as claimed in claim 1, wherein said third constraint comprises a maximum detecting routine for selecting a line segment with a maximum length for said respective subspace.

3. A method as claimed in claim 1, wherein said fourth constraint comprises a subspace exterior to a circular area around the median point of said first line segment, having a radius equal to said first constraint.

4. A method as claimed in claim 1, wherein said fifth constraint comprises a subspace exterior to a circular area around the median point of said second line segment having a radius equal to said first constraint.

5. A method as claimed in claim 1, wherein said first constraint is 10λ

, where λ

is the distance between two ridges of a fingerprint.

6. A method as claimed in claim 1, wherein said second constraint is a 30°

angle.

7. A method as claimed in claim 1, wherein said subspace comprises the totality of grooves around a core zone of said reference image.

8. A method as claimed in claim 1, wherein each of said first to fourth reference set of metrics comprises the length of said respective first to fourth line segment, the angle between said respective line segment and x-axis of said reference coordinate system and the (x, y)-coordinates of the median point of said respective first to fourth line segment.

9. A method as claimed in claim 1, further comprising the step of scanning and processing a candidate fingerprint and preparing a candidate image in a candidate coordinate system, and reading a candidate ID associated with said candidate fingerprint.