Optical skin detection for face unlock

Patent metadata

Field	Value
Publication number	US-12530925-B2
Application number	US-202519088110-A
Country	US
Kind code	B2
Filing date	Mar 24, 2025
Priority date	Feb 18, 2021
Publication date	Jan 20, 2026
Grant date	Jan 20, 2026

Abstract

Proposed herein are systems for face authentication. The systems are configured to classify an object based on a distribution of light intensity.

First claim

The invention claimed is:

1 . A system comprising:
a near infrared (NIR) camera configured to capture a scene, the NIR camera comprising a CMOS sensor; an illumination unit comprising at least one of an LED and a laser, the illumination unit configured to project NIR light in a light pattern toward the scene; at least one optical element, wherein the NIR light passes through the at least one optical element; a processor configured to:
output a first image comprising light reflected from the scene and detect a face in the first image;
output a second image comprising a reflective light pattern;
analyze the reflective light pattern to determine whether the detected face includes depth information consistent with a three-dimensional object;
determine a material classification of the detected face based on a comparison of light intensity distributions in the second image and in a reference image;
authenticate the detected face based on the depth information and the material classification.

2 . The system of claim 1 , wherein the NIR camera is sensitive to light in a wavelength range of approximately 700 nm to 1100 nm.

3 . The system of claim 1 , wherein the illumination unit is configured to project the NIR light in a light pattern, the light pattern comprising a grid, a dot matrix, a pseudo-random, a random, or a stripe configuration.

4 . The system of claim 1 , wherein the reference image used for material classification is stored in at least one data storage device of the processor or retrieved from a remote server.

5 . The system of claim 1 , wherein the material classification is performed using a machine learning model trained to distinguish human skin from nonbiological materials.

6 . The system of claim 1 , wherein the processor is configured to capture multiple reflective light patterns over time and determine depth information.

7 . The system of claim 1 , wherein the processor is configured to determine unsuccessful authentication by detecting depth information inconsistent with the three-dimensional object.

8 . The system of claim 5 , wherein the processor is configured to determine unsuccessful authentication by determining a material classification consistent with a nonbiological material.

9 . The system of claim 1 , wherein the NIR camera and the illumination unit are integrated into a mobile device housing.

10 . The system of claim 1 , wherein the optical element comprises at least one of a diffractive optical element (DOE) and a lens.

11 . The system of claim 1 , wherein the optical element comprises a plurality of lenses and a diffractive optical element (DOE).

12 . The system of claim 1 , wherein the optical element is configured to modify the light from the illumination unit into a random light pattern, a periodic light pattern, or a combination thereof.

13 . The system of claim 1 , wherein the illumination unit further comprises a second LED or a second laser configured to output an NIR light beam toward the scene.

14 . The system of claim 1 , wherein the illumination unit further comprises a vertical cavity surface-emitting laser (VCSEL) configured to project an NIR light beam toward the scene.

15 . The system of claim 13 , wherein the first image comprises light reflected from the scene in response to the projection of the NIR light beam toward the scene.

16 . The system of claim 1 , wherein the second image comprises the reflective light pattern reflected in response to the projection of the illumination unit.

17 . A system comprising:
a near infrared (NIR) camera configured to capture a scene, the NIR camera comprising a CMOS sensor; an illumination unit comprising:
at least one of a first LED and a first laser configured to project NIR light in a light pattern toward the scene; and
at least one of a second LED and a second laser configured to project an NIR light beam toward the scene;
at least one optical element, wherein the NIR light passes through the at least one optical element; a processor configured to:
output a first image comprising light reflected from the scene in response to the projection of the NIR light beam;
detect a face in the first image;
output a second image comprising a reflective light pattern reflected from the scene in response to the projection of the NIR light in the light pattern;
analyze the reflective light pattern to determine whether the detected face includes depth information consistent with a three-dimensional object;
determine a material classification of the detected face based on a comparison of light intensity distributions in the second image and in a reference image;
authenticate the detected face based on the depth information and the material classification.

18 . The system of claim 17 , wherein the NIR camera is sensitive to light in a wavelength range of approximately 700 nm to 1100 nm.

19 . The system of claim 17 , wherein the light pattern comprises a grid, a dot matrix, a pseudo-random, a random, or a stripe configuration.

20 . The system of claim 17 , wherein the reference image used for material classification is stored in at least one data storage device of the processor or retrieved from a remote server.

21 . The system of claim 17 , wherein the material classification is performed using a machine learning model trained to distinguish human skin from nonbiological materials.

22 . The system of claim 17 , wherein the processor is configured to capture multiple reflective light patterns over time and determine depth information.

23 . The system of claim 17 , wherein the processor is configured to determine unsuccessful authentication by detecting depth information inconsistent with the three-dimensional object.

24 . The system of claim 21 , wherein the processor is configured to determine unsuccessful authentication by determining a material classification consistent with a nonbiological material.

25 . The system of claim 17 , wherein the NIR camera and the illumination unit are integrated into a mobile device housing.

26 . The system of claim 17 , wherein the optical element comprises at least one of a diffractive optical element (DOE) and a lens.

27 . The system of claim 17 , wherein the optical element comprises a plurality of lenses and a diffractive optical element (DOE).

28 . The system of claim 17 , wherein the optical element is configured to modify the light from the illumination unit into a random light pattern, a periodic light pattern, or a combination thereof.

29 . A mobile device comprising:
a near infrared (NIR) camera configured to capture a scene, the NIR camera comprising a CMOS sensor, and wherein the NIR camera is sensitive to light in a wavelength range of approximately 700 nm to 1100 nm; an illumination unit comprising at least one of an LED and a laser, the illumination unit configured to project NIR light in a light pattern toward the scene; at least one optical element, wherein the NIR light passes through the at least one optical element; a processor configured to:
output a first image comprising light reflected from the scene and detect a face in the first image;
output a second image comprising a reflective light pattern;
analyze the reflective light pattern to determine whether the detected face includes depth information consistent with a three-

Assignees

• Trinamix Gmbh

Inventors

• Schillen Peter
• Guthier Benjamin
• Schick Friedrich
• Guenther Manuel
• Diesselberg Lars
• Lennartz Christian

Classifications

• G06F21/32 Primary

  using biometric data, e.g. fingerprints, iris scans or voiceprints · CPC title

• G06V40/168

  Feature extraction; Face representation · CPC title

• G06V10/141

  Control of illumination · CPC title

• G01S17/894

  Three-dimensional [3D] imaging with simultaneous measurement of time-of-flight at a two-dimensional [2D] array of receiver pixels, e.g. time-of-flight cameras or flash lidar · CPC title

• G06T2207/30201

  Face · CPC title