Facial recognition system

Patent metadata

Field	Value
Publication number	US-10657362-B2
Application number	US-201615739924-A
Country	US
Kind code	B2
Filing date	Jun 30, 2016
Priority date	Jun 30, 2015
Publication date	May 19, 2020
Grant date	May 19, 2020

Abstract

Various embodiments of a facial recognition system are provided. In one embodiment, a processor determines a value for a lighting parameter associated with a captured facial image, determines whether any previously obtained images in a biometric database includes a similar value for the lighting parameter and, if not, stores the newly captured image in the database along with the lighting parameter value. In another embodiment, the processor calculates a score indicative of the likelihood that the face in the captured facial image is identical to the face of a previously obtained image in the database, determines whether the score exceeds a threshold value and, if so, generates a signal indicating a match. The processor adjusts the threshold based on one or more parameter values.

First claim

What is claimed is:

1. A facial recognition system, comprising:
a camera;
a processor configured to:
receive image data from the camera, the image data including a first image of a face of an individual;
determine a first value for a lighting parameter associated with the first image;
determine whether a collection of previously obtained images of the face of the individual in a biometric database includes a second image having a second value for the lighting parameter that meets a first predetermined condition relative to the first value for the lighting parameter; and,
store the first image in the biometric database along with the first value for the lighting parameter when the collection of previously obtained images does not include a second image having a second value for the lighting parameter that meets the first predetermined condition relative to the first value for the lighting parameter.

2. The facial recognition system of claim 1 wherein the processor is further configured to:
identify quantitative features of the face of the individual in the first image that facilitate biometric validation;
create a biometric template of the first image using the quantitative features if the quantitative features meet predetermined requirements; and,
store the biometric template in the biometric database and associate the biometric template with the first image and the first value of the lighting parameter in the biometric database.

3. The facial recognition system of claim 1 wherein the processor is further configured to:
determine, when the collection of previously obtained images does include a second image having a second value for the lighting parameter that meets the first predetermined condition relative to the first value for the lighting parameter, a number of images in the collection of previously obtained images having a value for the lighting parameter that meets the first predetermined condition relative to the first value for the lighting parameter; and,
storing the first image in the biometric database along with the first value for the lighting parameter when the number does not exceed a predetermined number.

4. The facial recognition system of claim 3 wherein the processor is further configured to:
identify, when the number does exceed the predetermined number, a low scoring image having the lowest qualitative score among the collection of previously obtained images that have a value for the lighting parameter that meets the first predetermined condition relative to the first value for the lighting parameter; and,
replace the low scoring image with the first image in the biometric database if a qualitative score associated with the first image exceeds the lowest qualitative score associated with the low scoring image.

5. The facial recognition system of claim 1 wherein the lighting parameter is a degree of luminance in the first image.

6. The facial recognition system of claim 1 wherein the lighting parameter is a degree of contrast in the first image.

7. The facial recognition system of claim 1 , further comprising a light sensor configured to generate a signal indicative of the first value of the lighting parameter.

8. The facial recognition system of claim 1 wherein the processor is further configured, in determining the first value for the lighting parameter associated with the first image, to evaluate a lighting level in the first image.

9. The facial recognition system of claim 1 , further comprising an entitlement scanner configured to obtain entitlement data from an entitlement presented by the individual and wherein the processor is further configured to receive the entitlement data from the entitlement scanner and to identify the collection of previously obtained images of the face of the individual based on the entitlement data.

10. The facial recognition system of claim 1 wherein the processor is further configured to:
determine, when the collection or previously obtained images includes a second image having a second value for the lighting parameter that meets the first predetermined condition relative to the first value for the lighting parameter, whether the first image meets a second predetermined condition relative to the second image; and,
generate a signal indicating the faces of the individual in the first and second image are identical if the first image meets the second predetermined condition relative to the second image.

11. A facial recognition system, comprising:
a camera;
a processor configured to:
receive image data from the camera, the image data including a first image of a face of an individual;
calculate a first score indicative of the likelihood that the face of the individual in the first image is identical to the face of the individual in a second image from a biometric database;
determine whether the first score exceeds a dynamic match threshold;
generate a signal indicating that the faces of the individual in the first and second images are identical if the first score exceeds the dynamic match threshold; and,
adjust the dynamic match threshold based on values of one or more parameters
calculate a second score indicative of the likelihood that the face of the individual in the first image is identical to the face of the individual in a third image from the biometric database;
determine whether the second score exceeds the dynamic match threshold;
generate a signal indicating that the faces of the individual in the first and third images are identical if the second score exceeds the dynamic match threshold

wherein the processor is further configured to:
calculate a third score indicative of the likelihood that the face of the individual in the first image is identical to the face of the individual in a fourth image from the biometric database;
determine a number of the first, second and third scores that exceed an average match threshold lower than the dynamic match threshold;
generate a signal indicating that the faces of the individual in the first and one of the second, third and fourth images are identical if the number exceeds a predetermined number.

12. The facial recognition system of claim 11 wherein the processor is configured to adjust the average match threshold based on values of one or more parameters.

13. The facial recognition system of claim 11 , further comprising an entitlement scanner configured to obtain entitlement data from an entitlement presented by the individual and wherein the processor is further configured to receive the entitlement data from the entitlement scanner and to obtain the second image from biometric database based on the entitlement data.

14. The facial recognition system of claim 11 wherein the one or more parameters include a calendar date.

15. The facial recognition system of claim 11 wherein the one or more parameters include a time of day.

16. The facial recognition system of claim 11 wherein the one or more parameters include a frequency of use of the facial recognition system.

17. The facial recognition system of claim 11 wherein the one or more parameters include a performance speed of the facial recognition system.

18. The facial recognition system of claim 11 wherein the one or more parameters include a number of times that the dynamic match threshold is not exceeded.

Assignees

• Nec Corp America

Inventors

• Ranganath Krishna
• Ganz Allen
• Yap Alvin Leong Ying
• Phillips Ashlee
• Molinari Michael
• Callier Algernon

Classifications

• G06K9/6214

  Physics · mapped topic

• G06K9/00288 Primary

  Physics · mapped topic

• G06K9/00241

  Physics · mapped topic

• G06K9/6215

  Physics · mapped topic

• G06K9/00926

  Physics · mapped topic