# US 2018/0349679 A1 · Apparatus, Systems and Methods for Improved Facial Detection and Recognition in Vehicle Inspection Security Systems

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2018/0349679 A1
- Document ID: US20180349679A1
- Status: Published application
- Kind: A1
- Application number: 15/924,880
- Filed: 2018-03-19
- Published: 2018-12-06
- Earliest claimed: 2017-05-30
- Assignees: Gatekeeper Ltd.
- Applicants: Gatekeeper, Inc.
- Inventors: Robert Rozploch; Diego Garrido; Marc Hansen
- Web page: https://patentgrep.com/patent/US20180349679A1

## CPC

- G06V 40/161: {Detection; Localisation; Normalisation}
- G06V 20/593: {Recognising seat occupancy}
- G06V 40/165: {using facial parts and geometric relationships}
- G06V 40/172: {Classification, e.g. identification}
- G07C 9/253: {visually}
- G06V 20/625: {License plates}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G07C: TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE

## Abstract

Devices, systems, and methods provide for improved facial detection and facial recognition in vehicle interior inspections. In various embodiments, one or more high-resolution cameras having one or more lens and filter attachments operate in conjunction with one or more auxiliary illumination devices and a related computer system. According to various embodiments, an auxiliary illumination device can be synchronized to one or more cameras, and configured to supply auxiliary illumination to facilitate capture of accurate and usable images. The filter(s) and advanced image processing solutions assist with facial detection and identifying individuals inside a vehicle, removing light glare and undesired reflections from a window surface, and capturing an image through a tinted window, among other things. Further, embodiments can compare a captured image to an authenticated image from a database, in order to confirm the identity of a vehicle occupant.

## Claims

### Claim 1

A method for detecting a vehicle occupant, comprising the steps of:
providing a camera comprising a camera lens having a field of view;
detecting the presence of an oncoming vehicle, wherein the vehicle includes at least one vehicle window separating a vehicle interior enclosure from an external environment, wherein the at least one vehicle window has a frame surrounding the at least one vehicle window;
triggering the automatic capture of a plurality of current facial images of at least one vehicle occupant present within the vehicle interior enclosure when the at least one vehicle window frame is maximized in the field of view, wherein each of the plurality of current facial images is captured automatically via the camera through the camera lens, and wherein each of the plurality of current facial images is further captured from a different distance to the oncoming vehicle;
obtaining, by a processing unit, at least one archived image of at least one individual;
comparing, by the processing unit, at least one of the current facial images with the at least one archived image; and
determining whether the vehicle occupant in the at least one of the current facial images is the same individual as the at least one individual from the at least one archived image.

### Claim 2 (depends on 1)

The method of claim 1, wherein the step of comparing at least one of the current facial images with the at least one archived image is performed by the processing unit using facial recognition programming.

### Claim 3 (depends on 1)

The method of claim 1, wherein the at least one window frame is a trapezoidal frame of a front windshield of the oncoming vehicle.

### Claim 4 (depends on 1)

The method of claim 1, wherein the step of triggering the automatic capture is further performed when at least a portion of the at least one vehicle occupant is in the field of view.

### Claim 5 (depends on 1)

The method of claim 1, wherein the camera is located in the external environment.

### Claim 6 (depends on 1)

The method of claim 1, wherein the step of triggering the automatic capture is performed through optical flow algorithmic image analysis.

### Claim 7 (depends on 1)

The method of claim 1, including the step of authenticating the at least one vehicle occupant for access to an access-controlled asset based upon the step of comparing, and further based upon one or more of: a detected license plate number, an under-vehicle image, and a readable tag on the vehicle.

### Claim 8 (depends on 1)

The method of claim 1, including the step of determining, by the processing unit, a total number of vehicle occupants in the vehicle interior enclosure based upon the captured plurality of current facial images.

### Claim 9

A system for detecting a vehicle occupant, comprising:
a camera comprising a camera lens facing a field of view;
at least one processor; and
at least one memory device storing a plurality of instructions which, when executed by the at least one processor, cause the at least one processor to:
store one or more images of at least one individual person;
automatically trigger the activation of the camera when a frame of a front windshield of a moving vehicle is maximized within the field of view;
compare at least one image captured by the camera of at least one vehicle occupant present within a vehicle interior enclosure of a vehicle, with the one or more stored images of the at least one individual person, wherein the at least one image is captured by the camera through the window of the moving vehicle; and
determine whether the at least one vehicle occupant and the at least one individual person are the same.

### Claim 10 (depends on 9)

The system of claim 9, wherein the frame of the front windshield is a trapezoidal-shaped frame.

### Claim 11 (depends on 9)

The system of claim 9, wherein the at least one image captured by the camera is taken from a plurality of current facial images of the at least one vehicle occupant, and wherein each of the plurality of current facial images is captured from a different distance to the vehicle.

### Claim 12 (depends on 11)

The system of claim 11, wherein the plurality of instructions, when executed by the at least one processor, further cause the at least one processor to determine a number of vehicle occupants in the vehicle interior enclosure based upon the captured plurality of current facial images.

### Claim 13 (depends on 9)

The system of claim 9, wherein the plurality of instructions, when executed by the at least one processor, further cause the at least one processor to authenticate the at least one vehicle occupant for access to an access-controlled asset based upon the step of comparing, and further based upon one or more of: a detected license plate number, an under-vehicle image, and a readable tag on the vehicle.

### Claim 14 (depends on 9)

The system of claim 9, wherein the camera is located external to the vehicle.

### Claim 15 (depends on 9)

The system of claim 9, wherein automatically triggering the activation of the camera is performed when at least a portion of the at least one vehicle occupant is in the field of view.

### Claim 16 (depends on 9)

The system of claim 9, wherein automatically triggering the activation of the camera is performed through optical flow algorithmic image analysis.
