Positive patient identification

• US 7,436,986 B2
• Filed: 03/25/2003
• Issued: 10/14/2008
• Est. Priority Date: 03/25/2003
• Status: Active Grant

• Current Assignee: Bausch & Lomb Incorporated (Bausch Health Cos., Inc.)
• Original Assignee: Bausch & Lomb Incorporated (Bausch Health Cos., Inc.)
• Inventors: Caldwell, Lloyd M.
• Primary Examiner(s): Bella; Matthew C
• Assistant Examiner(s): LIEW, ALEX KOK SOON
• Application Number:
  US10/397,100
• Publication Number:
  US 20040190759A1
• Time in Patent Office: 2,030 Days
• Field of Search: None
• US Class Current: 382/117
• CPC Class Codes: A61B 5/1171 based on the shapes or appearances of their bodies or parts thereof
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris

Abstract

A process and an enabling system and associated devices are directed to automatically and positively identifying a person based upon a matching comparison between a stored, historical iris recognition-based record and a contemporaneously obtained record. The algorithm involves identifying anchor features in selected regions of an iris image and mapping an iris pattern associated with the anchor features into a topologically consistent flat analysis space. Analysis according to the invention generates historical (reference) and contemporaneous Iris Feature Vectors for individuals upon which matching comparisons can be made to positively identify an individual from the reference base.

Claims

1. A system for positively identifying a patient, comprising:

an eye illumination source;

a sensor configured to capture an image of the patient'"'"'s eye containing an iris;

an optical system for transferring the image to the sensor;

means for analyzing the image;

an information storage/retrieval component coupled to the analyzing means and having stored therein a plurality of historical Iris Feature Vectors (IFVs) representing identification indicia of a plurality of patients;

wherein the analyzing means is adapted to generate from the image a contemporaneous IFV representing an identification indicia of the patient;

wherein the analyzing means is suitably programmed for comparing the contemporaneous IFV to the plurality of stored, historical IFVs, and determining a similarity relationship between the contemporaneous IFV and the historical IFVs, and

wherein the analyzing means is adapted to identify in the image an inner region and an outer region, an inner region anchor feature located on an outer border of the inner region, and an outer region anchor feature located on an outer border of the outer region, the contemporaneous IFV being located in an analysis space bounded by the outer border of the inner region, and the outer border of the outer region and a third border extending between the inner region anchor feature and the outer region anchor feature, wherein said anchor features are landmarks of the patient'"'"'s eye.

the outer border of the inner region being located at a pupil/iris boundary, and the outer border of the outer region being located at a limbus boundary.

12. The system of claim 1, wherein the analyzing means is adapted to identify the contemporaneous IFVs within a sample regions inside the analysis space, the total area of the sample regions being less than the area of the analysis space.

13. The system of claim 12, wherein the sample region comprises a plurality of sample region boxes.

14. The system of claim 1, wherein the border extending between the inner region anchor feature and the outer region anchor feature is a straight line.

2. A method for facilitating positively identifying a patient, comprising the following steps:

a) acquire an image of the patient'"'"'s eye containing an iris;

b) determine an inner region and an outer region of the eye;

c) locate an inner region anchor feature within a search space on an outer border of the inner region and located at a pupil/iris boundary;

d) locate an outer region anchor feature within a search space on an outer border of the outer region and located at a limbus boundary;

e) identify a third border extending between the inner region anchor feature and the outer region anchor feature, wherein said anchor features are landmarks of the patient'"'"'s eye;

f) determine a plurality of iris image intensity values in a region of the image bounded by the third border and the outer border of the inner region and the outer border of the outer region;

g) map the intensity values into a normalized analysis space;

h) create a pattern of sample regions, choosing a size and shape for each sample region;

i) sample the normalized analysis space with the sample region pattern;

j) create from 1 to n Iris Feature Vectors (IFVs) for each sample region;

k) create an array of IFVs for the sample regions, and;

l) store the array in a storage/retrieval medium.

3. The method of claim 2, wherein the array of stored IFVs comprise an historical IFV representing a particular patient'"'"'s identification data set.

4. The method of claim 2 further comprising sorting the array of IFVs.

5. The method of claim 3, further comprising performing steps a) through k) at a subsequent time on a particular patient, wherein the subsequently obtained array comprises a contemporaneous IFV for the particular patient.

6. The method of claim 5, comprising comparing the contemporaneous IFV with a stored plurality of historical IFVs, and generating a similarity index, x, for each comparison, so as to match the contemporaneous IFV with an historical IFV.

7. The method of claim 6, wherein 0≦

x≦

1.

8. The method of claim 6, comprising perturbing a sample region location in the sample region pattern when the similarity index is not sufficient to positively match the contemporaneous IFV with a single historical IFV.

9. A storage/retrieval medium having stored therein an historical IFV obtained in accordance with the method of claim 2.

10. A device readable medium having stored therein a contemporaneous IFV obtained in accordance with the method of claim 4.

11. An ophthalmic diagnostic or therapeutic system being sufficient to perform an algorithm, and being programmed in such a manner to execute the algorithm for a positive patient identification, said algorithm comprising:

a) acquire an iris image of a patient'"'"'s eye containing an iris;

b) determine an inner region and an outer region of the eye;

c) locate an inner region anchor feature within a search space on an outer border of the inner region and located at a pupil/iris boundary;

d) locate an outer region anchor feature within a search space on an outer border of the outer region and located at a limbus boundary;

e) identify a third border extending between the inner region anchor feature and the outer region anchor feature, wherein said anchor features are landmarks of the patient'"'"'s eye;

f) determine a plurality of iris image intensity values in a region of the image bounded by the third border and the outer border of the inner region and the outer border of the outer region into a normalized analysis space;

g) map the intensity values into a normalized analysis space;

h) create a pattern of sample regions, choosing a size and shape for each sample region;

i) sample the normalized analysis space with the sample region pattern;

j) create from 1 to n Iris Feature Vectors (IFVs) for each sample region;

k) create an array of the IFVs, and;

l) store the array in a storage/retrieval medium.