IP Library Granted Patent US 12,475,720
Granted Patent B2
US 12,475,720 · App. 18/315,701 · Granted Nov 18, 2025

Methods and systems for parking detection

Inventors: David W. Wilson (Milford, MI); Pulkit Monga (Westland, MI); Kalash Jain (Bloomfield Hills, MI); Olivier Barree (Elancourt, FR); Dena Memari (North Bergen, NJ)
Assignee: ZF Friedrichshafen AG
G06V20/586G06V10/25G06V20/582G06V20/588G08G1/141G08G1/146
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,475,720
App. No.
18/315,701
Granted
Nov 18, 2025
Kind
B2
Abstract

A parking detection method includes: accessing data corresponding to an image from a camera on a vehicle; computing a confidence score for a possible vacant parking location from the data corresponding to the image from the camera on the vehicle; and when the confidence score is greater than a threshold confidence, identifying the possible vacant parking location as an actual vacant parking location from the data corresponding to the image from the camera on the vehicle.

Claims (47)

1 . A parking detection method, comprising:

accessing, with a computing device, data corresponding to an image from a front camera on the vehicle;

computing, with the computing device, a confidence score for a possible parking location from the data corresponding to the image from the front camera on the vehicle, wherein computing the confidence score comprises detecting a possible parking sign located on or above a roadway from the data corresponding to the image from the front camera on the vehicle;

when the confidence score is greater than a threshold confidence, identifying, with the computing device, the possible parking location as an actual parking location; and

when the confidence score is less than the threshold confidence, confirming, with the computing device, the possible parking location as the actual parking location based at least in part on data corresponding to an image from a rear camera on the vehicle, the rear camera having a wider field of view than the front camera.

2 . The parking detection method of claim 1 , further comprising accessing, with the computing device, data corresponding to a location of a vehicle, wherein accessing the data corresponding to the image from the front camera comprises accessing the data corresponding to the image from the front camera while the location of the vehicle corresponds to a region of interest.

3 . The parking detection method of claim 2 , wherein the region of interest comprises an area near a destination of the vehicle.

4 . The parking detection method of claim 1 , wherein computing the confidence score further comprises one or more of:

detecting a possible parking lane from the data corresponding to the image from the front camera on the vehicle;

detecting a possible road marking from the data corresponding to the image from the front camera on the vehicle; and

detecting a possible parked vehicle from the data corresponding to the image from the front camera on the vehicle.

5 . The parking detection method of claim 1 , further comprising transmitting data corresponding to a location of the actual parking location to a remote computing device located offboard the vehicle.

6 . A control unit, programmed to implement the method of claim 1 .

7 . A parking detection method, comprising:

accessing, with a computing device, data corresponding to an image from a rear camera on the vehicle;

computing, with the computing device, a confidence score for a possible parking location from the data corresponding to the image from the rear camera on the vehicle, wherein computing the confidence score comprises detecting a possible parking sign located on a roadway from the data corresponding to the image from the rear camera on the vehicle;

when the confidence score is greater than a threshold confidence, identifying, with the computing device, the possible parking location as an actual parking location; and

when the confidence score is less than the threshold confidence, confirming, with the computing device, the possible parking location as the actual parking location based at least in part on data corresponding to a meshed image formed from the image from the rear camera and a reversed image from a front camera on the vehicle.

8 . The parking detection method of claim 7 , further comprising accessing, with the computing device, data corresponding to a location of a vehicle, wherein accessing the data corresponding to the image from the rear camera comprises accessing the data corresponding to the image from the rear camera while the location of the vehicle corresponds to a region of interest.

9 . The parking detection method of claim 8 , wherein the region of interest comprises an area near a destination of the vehicle.

10 . The parking detection method of claim 7 , wherein computing the confidence score further comprises one or more of:

detecting a possible parking lane from the data corresponding to the image from the rear camera on the vehicle;

detecting a possible road marking from the data corresponding to the image from the rear camera on the vehicle; and

detecting a possible parked vehicle from the data corresponding to the image from the rear camera on the vehicle.

11 . The parking detection method of claim 7 , further comprising transmitting data corresponding to a location of the actual parking location to a remote computing device located offboard the vehicle.

12 . A control unit, programmed to implement the method of claim 7 .

13 . A parking detection method, comprising:

accessing, with a computing device, data corresponding to an image from a front camera on a vehicle;

accessing, with the computing device, data corresponding to an image from a rear camera on the vehicle;

identifying, with the computing device, a possible vacant parking location from one of both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle;

computing, with the computing device, a confidence score for the possible vacant parking location from one or both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle;

when the confidence score is greater than a threshold confidence, identifying, with the computing device, the possible vacant parking location as an actual vacant parking location; and

when the confidence score is less than the threshold confidence, confirming, with the computing device, the possible vacant parking location as the actual vacant parking location based at least in part the other or both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle,

wherein the front camera is a monochrome camera, the rear camera is a color camera, and the rear camera has a wider field of view than the front camera.

14 . The parking detection method of claim 13 , wherein the front camera comprises a camera configured as a component of an advanced driver assistance system of the vehicle.

15 . The parking detection method of claim 13 , wherein the rear camera comprises a camera configured as a backup camera of the vehicle.

16 . The parking detection method of claim 13 , wherein computing the confidence score comprises one or more of:

detecting a possible parking sign from one or both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle;

detecting a possible parking lane from one or both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle;

detecting a possible road marking from one or both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle; and

detecting a possible parked vehicle from one or both of the data corresponding to the image from the front camera on the vehicle and the data corresponding to the image from the rear camera on the vehicle.

17 . The parking detection method of claim 13 , further comprising:

determining, with the computing device, one or more parameters of the actual vacant parking location, wherein the one or more parameters comprise an orientation of the actual parking spot and a size of the parking spot; and

transmitting data corresponding to a location of the actual vacant parking location and the one or more parameters of the actual vacant parking location to another computing device located offboard the vehicle.

18 . The parking detection method of claim 13 , wherein the computing device is located onboard the vehicle, and the method further comprises transmitting, with the computing device, data corresponding to the actual parking location to another computing device located offboard the vehicle.

19 . The parking detection method of claim 18 , wherein transmitting the data corresponding to actual parking location comprises transmitting the data corresponding to actual parking location at a configurable frequency.

20 . A control unit, programmed to implement the method of claim 13 .

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 27, 2023
From: WILSON, DAVID W.; MONGA, PULKIT; JAIN, KALASH; BARREE, OLIVIER; MEMARI, DENA
To: ZF FRIEDRICHSHAFEN AG
Reel/Frame 065960/0617 →
Continuity (2)
Provisional Application 63341185 · May 12, 2022
Related Publication 20230368546A1 · Nov 16, 2023
References Cited (9)
US 11137754B2 · Van Wiemeersch · 2021 [cited by examiner]
US 12067878B1 · Campbell · 2024 [cited by examiner]
US 20140055601A1 · Yamada · 2014 [cited by examiner]
US 20190180621A1 · Matsuda · 2019 [cited by examiner]
US 20200307554A1 · Lai et al. · 2020 [cited by applicant]
DE 102015211514A1 · 2016 [cited by examiner]
JP 2017076275A · 2017 [cited by examiner]
Burleigh, Nicholas et al. “Deep Learning for Autonomous Driving” 2019 Digital Image Computing: Techniques and applications(Dicta), IEEE, Dec. 2, 2019, pp. 1-8, XP033683049. [cited by applicant]
European Search Report for Application No. EP23172470 Dated Sep. 7, 2023. [cited by applicant]