Method for generating an image of vehicle surroundings, and apparatus for generating an image of vehicle surroundings

Patent metadata

Field	Value
Publication number	US-12080075-B2
Application number	US-202017595681-A
Country	US
Kind code	B2
Filing date	May 12, 2020
Priority date	May 21, 2019
Publication date	Sep 3, 2024
Grant date	Sep 3, 2024

Abstract

A method and apparatus for generating an image of vehicle surroundings are disclosed, including: capturing vehicle surroundings by vehicle cameras arranged on a vehicle body of a vehicle, and generating camera images by the cameras. The camera images of adjacent cameras have overlapping image regions), generating a virtual representation of the surroundings in a virtual three-dimensional space, during which the camera images are projected onto a virtual projection surface in the space. A non-stationary virtual camera in the virtual space determines a position and/or orientation thereof. A first selection region is placed on the surface in a first overlapping image region depending on a virtual camera field of vision, at least one image parameter of a first vehicle camera in the first selection region is calculated, and at least one second vehicle image parameter is adjusted to the at least one first vehicle image parameter in the first selection region.

First claim

The invention claimed is:

1. A method for generating an image of vehicle surroundings, comprising:
capturing vehicle surroundings by a plurality of vehicle cameras which, in particular, are arranged on a vehicle body of a vehicle,
generating camera images by the plurality of vehicle cameras, wherein the camera images of adjacent vehicle cameras have overlapping image regions,
generating a virtual representation of the surroundings in a three-dimensional virtual space, wherein, during the generation, the camera images are projected onto a virtual projection surface in the three-dimensional virtual space,
providing a non-stationary virtual camera in the three-dimensional virtual space and determining at least one of a virtual camera position or a virtual camera orientation,
placing a first selection region on the virtual projection surface in a first overlapping image region of the overlapping image regions depending on a field of vision of the virtual camera,
calculating at least one image parameter of a first vehicle camera of the plurality of vehicle cameras in the first selection region, and
adjusting at least one image parameter of a second vehicle camera of the plurality of vehicle cameras to the at least one image parameter of the first vehicle camera in the first selection region.

2. The method according to claim 1 , further comprising:
placing a second selection region on the virtual projection surface within a second overlapping image region of the overlapping image regions depending on the field of vision of the virtual camera,
calculating at least one image parameter of a further vehicle camera of the plurality of vehicle cameras, the camera image of which has the second overlapping image region, in the second selection region,
adjusting at least one image parameter of another vehicle camera of the plurality of vehicle cameras, the camera image of which likewise has the second overlapping image region, to the at least one image parameter of the further vehicle camera in the second selection region, wherein the further vehicle camera comprises a first one of the second vehicle camera or a third vehicle camera of the plurality of vehicle cameras, and the another vehicle camera comprises a second one of the second vehicle camera or the third vehicle camera.

3. The method according to claim 2 , wherein the another vehicle camera comprises the second vehicle camera and the further vehicle camera comprises the third vehicle camera, and wherein the at least one image parameter of the second vehicle camera, which has the first and the second overlapping image regions, is adjusted between the first and the second selection regions by interpolation.

4. The method according to claim 3 , wherein at least one of a position of the first selection region, a position of the second selection region, or X, Y and Z coordinate values of a currently rendered point is/are considered during the interpolation.

5. The method according to claim 1 , wherein the placing of the first selection region comprises:
placing the first selection region at a standard position within the first overlapping image regions,
verifying whether the first selection region is visible to the virtual camera at the standard position, and
displacing the first selection region to a first position, spaced apart from the standard position, on the virtual projection surface within the first overlapping image region if the selection region at the standard position is not visible to the virtual camera, wherein the first position is visible to the virtual camera, and wherein the at least one image parameter of the first vehicle camera is calculated in the first selection region at the first position.

6. The method according to claim 5 , wherein the standard position is used as an evasive position if no selection region visible to the virtual camera can be ascertained.

7. The method according to claim 1 , further comprising providing parameters of a vehicle model, and integrating the parameters of the vehicle model in the three-dimensional virtual space.

8. The method according to claim 1 , wherein the at least one image parameters of the first and second vehicle cameras comprise at least one of an image brightness, an image contrast, an image color, an image sharpness, a color saturation, or a texture frequency.

9. The method according to claim 1 , further comprising:
placing a second selection region on the virtual projection surface within a second overlapping image region of the overlapping image regions depending on the field of vision of the virtual camera,
calculating at least one image parameter of a further vehicle camera of the plurality of vehicle cameras, the camera image of which has the second overlapping image region, in the second selection region, and
adjusting at least one image parameter of another vehicle camera, the camera image of which likewise has the second overlapping image region, to the at least one image parameter of the further vehicle camera of the plurality of vehicle cameras in the second selection region,
wherein the first vehicle camera is a front vehicle camera and/or the further vehicle camera is a rear vehicle camera, or the first vehicle camera is a rear vehicle camera and/or the further vehicle camera is a front vehicle camera.

10. The method according to claim 1 , wherein the second vehicle camera is a left side or a right side vehicle camera.

11. An apparatus for generating an image of vehicle surroundings, comprising:
a plurality of vehicle cameras for capturing vehicle surroundings and producing camera images, wherein the camera images of adjacent vehicle cameras have overlapping image regions, and wherein the vehicle cameras are in particular mounted on a vehicle bodywork of a vehicle, and
a computing unit having one or more inputs coupled to outputs of the vehicle cameras for receiving the camera images, the computing unit being is configured to generate a virtual representation of the surroundings in a three-dimensional virtual space, wherein, during the generation, the camera images are projected onto a virtual projection surface in the three-dimensional virtual space, and to provide a non-stationary virtual camera in the three-dimensional virtual space,
wherein the computing unit is furthermore configured to place a first selection region on the virtual projection surface in a first overlapping image region depending on a field of vision of the virtual camera, to calculate at least one image parameter of a first vehicle camera in the first selection region, and to adjust at least one image parameter of a second vehicle camera to the at least one image parameter of the first vehicle camera in the first selection region.

12. The apparatus according to claim 11 , wherein the computing unit is furthermore configured to place a second selection region on the virtual projection surface within a second overlapping image region depending on the field of vision of the virtual camera, to calculate at least one image parameter of a further vehicle camera, the camera image of which has the second overlapping image region, in the second selection region, and to adjust at least one image parameter of another vehicle camera, the camera image of which likewise has the second overlapping image region, to the at least one image parameter of the further vehicle camera in the second selection region.

13. An apparatus for generating an image of vehicle surroundings, comprising:
a computing unit having a plurality of input terminals, the computing unit configured for
receiving, at the plurality of input terminals of the computing unit, camera images from a pluralit

Assignees

• Conti Temic Microelectronic Gmbh

Inventors

• Kutschera Thomas
• Wiest Joachim
• Friebe Markus

Classifications

• B60W2420/403

  Image sensing, e.g. optical camera · CPC title

• B60R1/27

  providing all-round vision, e.g. using omnidirectional cameras · CPC title

• B60W40/12

  related to parameters of the vehicle itself {, e.g. tyre models} · CPC title

• G06T7/90

  Determination of colour characteristics · CPC title

• G06T7/97

  Determining parameters from multiple pictures (depth or shape recovery from multiple images G06T7/55; stereo camera calibration G06T7/85) · CPC title