BIOMETRIC IMAGE OPTIMIZATION USING LIGHT FIELDS

• US 20170024603A1
• Filed: 07/18/2016
• Published: 01/26/2017
• Est. Priority Date: 07/22/2015
• Status: Abandoned Application

• Current Assignee: Anthony Ray Misslin
• Original Assignee: Anthony Ray Misslin
• Inventors: Misslin, Anthony Ray
• Application Number:
  US15/212,444
• Publication Number:
  US 20170024603A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06V 10/145 Illumination specially adapted for pattern recognition, e.g. using gratings
• CPC Class Codes: G06V 40/1312 direct reading, e.g. contactless acquisition
• CPC Class Codes: G06V 40/19 Sensors therefor

Abstract

The application of light field imaging in the capture of biometric images enables image processing to use continuous focus adjustment of a single image to construct an image of a biometric feature that is in focus across an expanded depth of field. Because a single image is the source of all information, the final image is formed without requiring combination of multiple images that may have physically moved between capture of the images. Light field imaging can be accomplished through multiple methods including plenoptic cameras or focus stacking cameras. The use of near infrared wavelengths in images captured using light fields optimizes the deployment of biometric systems by eliminating intense visible light currently implemented to capture some biometrics. The use of near infrared wavelengths for light field imaging enhances iris capture as the iris display more useful characterization under near infrared illumination.

Claims

1. A method of optimizing a single image of a biometric parameter, comprising:

capturing an image of the biometric parameter having at least one of image portions that are in different focal planes and image portions that are viewable from different viewing angles;

forming a single image includes forming multiple views of the biometric parameter wherein each one of the multiple views comprises a different combination of at least one of a focal plane and a viewing angle; and

processing the multiple views to create a single improved image of the biometric parameter with the image portions all in focus and including information from multiple viewing angles.

2. The method of claim 1, wherein forming each one of the multiple views include a different combination of a focal plane and a viewing angle.

3. The method of claim 1, further including processing the image on a device forming the single image.

4. The method of claims 1, further including processing the image on one of a stand-alone device, computer, and workstation.

5. The method of claim 1, further including processing of the biometric image captured to determine the areas of the image which are in focus for each of the views;

storing the portion of the image which is in focus for each view on the device; and

processing the stored images to form a single image of the biometric with the optimal in focus information available.

6. The method of claim 1, further including processing of the biometric image captured to determine the areas available of the image which provide a different viewing angle of the biometric;

storing the portions of the image available from each view with different viewing angles;

processing the portions of each stored view that is in focus; and

processing the portions of each view that is in focus from each of the viewing angles to provide an image of the biometric from viewing angles that extend beyond the perpendicular view of the biometric.

7. The method of claim 1, wherein the method includes using a plenoptic camera for capturing the single image.

8. The method of claim 1, wherein the method includes using a focus stacking camera for capturing the single image.

9. The method of claim 1, wherein a single fingerprint is the biometric image captured and optimized using at least one of image portions that are in different focal planes and image portions that are viewable from different viewing angles.

10. The method of claim 1, wherein multiple fingerprints are the biometric image captured and optimized using at least one of image portions that are in different focal planes and image portions that are viewable from different viewing angles.

11. The method of claim 10, wherein an image is captured of multiple fingerprints further including processing of the biometric image to determine the location of each of the fingers in the image;

processing each fingerprint to determine the view with the fingerprint in focus;

storing the view with the fingerprint in focus on the device; and

processing all stored fingerprint images to combine the saved fingerprints into one biometric image.

12. The method of claim 1, wherein palm prints are the biometric image captured and optimized using at least one of image portions that are in different focal planes and image portions that are viewable from different viewing angles.

13. The method of claim 1, wherein iris images are the biometric image captured and optimized using at least one of image portions that are in different focal planes and image portions that are viewable from different viewing angles.

14. The method of claim 1, wherein at least one of scars, marks, tattoos, and skin texture are the biometric image captured and optimized using at least one of image portions that are in different focal planes and image portions that are viewable from different viewing angles.

15. The method of claim 1, wherein capturing the biometric image is performed in the near infrared spectrum.