Search

US-12620264-B2 - Actuating a system via live person image recognition

US12620264B2US 12620264 B2US12620264 B2US 12620264B2US-12620264-B2

Abstract

A computer includes a processor and a memory, the memory stores instructions executable by the processor to generate a first cropped image by applying a first cropping operation to a captured image, in which the first cropping operation is selected based on a first component to be actuated. The instructions can additionally be to apply a live person image recognition algorithm to the first cropped image to determine whether a recognized person is represented in the first cropped image and to output a signal to initiate actuation of the first component in response to the live person image recognition algorithm responsive to determining that the recognized person is in the first cropped image and is permitted to actuate the first component.

Inventors

  • Jake Schwartz
  • John Robert Van Wiemeersch
  • Medha Karkare
  • Ryan Edwin Hanson

Assignees

  • FORD GLOBAL TECHNOLOGIES, LLC

Dates

Publication Date
20260505
Application Date
20230419

Claims (20)

  1. 1 . A system comprising: a computer including a processor coupled to a memory, the memory storing instructions including instructions executable by the processor to: generate a first cropped image by applying a first cropping operation to a captured image, wherein the first cropping operation is selected based on a selection of a first component to be actuated from a plurality of components; apply a live person image recognition algorithm to the first cropped image to determine whether a recognized person is represented in the first cropped image; and output a signal to initiate actuation of the first component in response to the live person image recognition algorithm responsive to determining that the recognized person is in the first cropped image and is permitted to actuate the first component.
  2. 2 . The system of claim 1 , wherein the first cropping operation includes removing a portion of the captured image outside of an area that includes a largest area of facial features of the recognized person.
  3. 3 . The system of claim 1 , wherein the first cropping operation includes removing a portion of the captured image outside of a central portion of the captured image that includes facial features of the recognized person.
  4. 4 . The system of claim 1 , wherein the first cropping operation includes removing a portion of the captured image outside of facial features and least-obstructed shoulders of the recognized person.
  5. 5 . The system of claim 1 , wherein the live person image recognition algorithm includes comparing a person represented in the first cropped image with image parameters of a most recent user of the system.
  6. 6 . The system of claim 1 , wherein the first component is a component in a vehicle.
  7. 7 . The system of claim 1 , wherein the instructions further include instructions to: independently apply the live person image recognition algorithm to the first cropped image and to a second cropped image generated by applying a second cropping operation to the captured image; and determine a sum of weighted scores for the first cropping operation and the second cropping operation.
  8. 8 . The system of claim 1 , wherein the first component is a door lock.
  9. 9 . The system of claim 1 , wherein the first component is a seat.
  10. 10 . The system of claim 1 , wherein the first component is a climate control.
  11. 11 . The system of claim 1 , wherein the first component is an infotainment component.
  12. 12 . The system of claim 1 , wherein the first component is a propulsion component of a vehicle.
  13. 13 . The system of claim 1 , wherein the instructions further include instructions to: increment a lockout counter in response to applying multiple occurrences of the live person image recognition algorithm.
  14. 14 . The system of claim 1 , wherein the instructions further include instructions to: select a second cropping operation for application to a second captured image to control a second component of the system; apply the live person image recognition algorithm to determine whether the recognized person is represented in the second captured image cropped by the second cropping operation; and actuate the second component in response to the live person image recognition algorithm determining that the recognized person is permitted to actuate the second component.
  15. 15 . A method comprising: generating a first cropped image by applying a first cropping operation to a captured image, wherein the first cropping operation is selected based on a selection of a first component to be actuated from a plurality of components; applying a live person image recognition algorithm to the first cropped image to determine whether a recognized person is represented in the first cropped image; and outputting a signal to initiate actuation of the first component in response to the live person image recognition algorithm responsive to determining that the recognized person is in the first cropped image and is permitted to actuate the first component.
  16. 16 . The method of claim 15 , further comprising: incrementing a lockout counter in response to applying multiple occurrences of the live person image recognition algorithm.
  17. 17 . The method of claim 15 , wherein the first component is a door lock.
  18. 18 . The method of claim 15 , wherein the first component is a seat.
  19. 19 . The method of claim 15 , wherein the first component is a propulsion component.
  20. 20 . The method of claim 15 , wherein the first component is a component in a vehicle.

Description

BACKGROUND Systems are often equipped with cameras, which can facilitate observation of an environment external to the system as well as areas internal to the system. For example, in a system that includes a vehicle, a facial recognition system, operating under the control of a vehicle computer, can compare parameters of a captured image of the facial and/or other features of a person, who may be attempting to enter the vehicle, operate the vehicle, etc., with a set of stored parameters to determine whether the person is authorized to do so. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an example vehicle. FIG. 2 depicts a top view of a group of people attempting to access a vehicle. FIG. 3 depicts people positioned within a field-of-view of a vehicle internal camera. FIG. 4 is a process flow diagram of an example process for actuating a system via live person image recognition. DETAILED DESCRIPTION This disclosure provides techniques for accessing components of a system, such as a vehicle, via live person image recognition prior to permitting a person to actuate or control the component. In example implementations, images of a live person's facial and/or other features, e.g., the live person's shoulders, may be utilized in an image recognition process to biometrically identify the person, thereby determining whether the person is authorized to access and/or exercise control over the vehicle component. Operation of a vehicle component may include actuation of a door lock, accessing and/or controlling ignition of the vehicle, controlling propulsion, controlling steering of the vehicle, controlling an adjusting component of a seat, accessing a climate-control component, modifying an infotainment setting, and/or accessing a human-machine interface (HMI). In this context, a “live person image recognition process” means a process by which features of a person's face and/or other portions of the person are compared with a set of stored live person image parameters, and/or parameters relating to other portions of the person, e.g., the person's shoulders, to permit a computer of the vehicle to identify the person. In some instances, prior to obtaining access to the vehicle based on recognition of an image of a person, the person may participate in an enrollment process. That is, persons may be authenticated to access a system or components thereof based on whether they can be verified to have been enrolled. In an enrollment process, a camera of the vehicle may capture and store images of a person's facial and/or other features from various angles, which allows an onboard vehicle computer to recognize the person's facial and/or other features at various orientations of the person's features with respect to the camera. However, in an example situation, such as may be encountered when a person is attempting to unlock a vehicle door utilizing facial recognition, multiple individuals positioned within the field-of-view of a camera mounted near the vehicle door may result in the vehicle facial recognition system attempting to execute recognition utilizing facial and/or other features of a person who has not been enrolled. For example, a person holding a child may initiate a facial recognition process in which the vehicle computer attempts to execute facial recognition utilizing an image of the child's facial features rather than the facial features of the person holding the child. In another example, an authorized operator, i.e., a person authorized to actuate or control ignition of the vehicle, may be seated in a driver's position of the vehicle along with another person seated in the vehicle's rear seat. In such an instance, a vehicle computer receiving images from a dashboard-mounted camera may deny access to the vehicle's ignition component based on facial recognition of the person seated in the rear seat, who may not be enrolled as a person authorized to actuate or control the vehicle's ignition. In some instances, in response to multiple occurrences of the unsuccessful attempts by the vehicle computer to identify an enrolled person, the vehicle may implement a “lockout counter,” which prohibits any person from actuating a vehicle component, such as, e.g., controlling an unlocking mechanism of a vehicle door, controlling ignition of the vehicle, etc., for a predetermined duration. Thus, as disclosed herein, the vehicle computer may advantageously select a cropping operation to be applied to images of a live person's facial and/or other features, e.g., the live person's shoulders, present in a scene captured by a vehicle camera. In this context, a cropping operation means removal or adjustment of an image of a scene that excludes one or more areas of the scene, so as to direct a process of identifying a live person within the adjusted image via the person's facial and/or other features. Any suitable cropping operation or technique can be applied to an image of a captured scene. Such operations can in