Camera height calculation method and image processing apparatus

Patent metadata

Field	Value
Publication number	US-11580663-B2
Application number	US-202117191263-A
Country	US
Kind code	B2
Filing date	Mar 3, 2021
Priority date	Apr 27, 2020
Publication date	Feb 14, 2023
Grant date	Feb 14, 2023

Abstract

A camera height calculation method that causes a computer to execute a process, the process includes obtaining one or more images captured by an in-vehicle camera, extracting one or more feature points from the one or more images, identifying first feature points that exist over a road surface from the one or more feature points, and calculating a height of the in-vehicle camera from the road surface, based on positions of the identified first feature points.

First claim

What is claimed is:

1. A camera height calculation method that causes a computer to execute a process, the process comprising:
obtaining one or more images captured by an in-vehicle camera;
extracting one or more feature points from the one or more images;
identifying first feature points that exist over a road surface from the one or more feature points;
calculating a height of the in-vehicle camera from the road surface, based on positions of the identified first feature points; and
calculating a distance between an object and the in-vehicle camera based on the calculated height of the in-vehicle camera.

2. The camera height calculation method according to claim 1 ,
wherein the first feature points are included in a predetermined first range with reference to the in-vehicle camera, based on three-dimensional coordinates of the one or more feature points and the in-vehicle camera.

3. The camera height calculation method according to claim 2 ,
wherein the first feature points are included in a second range in a horizontal direction about the in-vehicle camera and included in a third range in a lower direction from a height of the in-vehicle camera in a three-dimensional coordinate system with reference to the in-vehicle camera.

4. The camera height calculation method according to claim 1 , wherein:
the one or more images are acquired by one or more image-capturing operations in response to a movement of the in-vehicle camera,
the one or more feature points are extracted from the one or more images, and
the identifying of the first feature points includes:
identifying one or more first candidates of the first feature points from the first feature points included in a fourth range with reference to each of the one or more image-capturing positions, based on three-dimensional coordinates indicating the one or more image-capturing positions and three-dimensional coordinates of the one or more feature points in a three-dimensional coordinate system with reference to the in-vehicle camera;
extracting, for each of the one or more image-capturing positions, one or more second candidates of the first feature points included in a fifth range about the one image-capturing position from the first candidates,
calculating, for each of the second candidates, a first distance from the one or more image-capturing positions in a vertical direction in the three-dimensional coordinate system,
calculating an average value of first distances calculated for each the second candidates, and
narrowing down, as the first feature points, the first candidates having the first distances included in a sixth range from the average value, from the second candidates.

5. The camera height calculation method according to claim 4 ,
wherein the identifying of the first feature points further includes:
extracting, for each of the one or more image-capturing positions, second feature points included in the fifth range about the image-capturing position from the first feature points by the narrowing down of the first candidates,
identifying a minimum value of a first coordinate in the vertical direction of the second feature point, and
narrowing down, from the second feature points, the first feature points having the first coordinates included in a seventh range from the minimum value.

6. The camera height calculation method according to claim 1 ,
wherein the identifying of the first feature points:
identifies an area of the road surface by image recognition of the obtained one or more images, and
identifies the first feature points from the one or more feature points included in the identified area of the road surface.

7. The camera height calculation method according to claim 1 ,
wherein the calculating of the height calculates the height based on relative positions in a three-dimensional coordinate system of the first feature point and the in-vehicle camera.

8. The camera height calculation method according to claim 1 , wherein:
the one or more images are acquired by one or more image-capturing operations in response to a movement of the in-vehicle camera, and
the calculating of the height
for each of image-capturing positions of the one or more images, smooths coordinates in a vertical direction of third feature points included in an eighth range about the image-capturing position among the first feature points in a three-dimensional coordinate system with reference to the in-vehicle camera, and
calculates the height from a difference between the smoothed coordinate and the coordinate in the vertical direction of the image-capturing position.

9. An image processing apparatus comprising:
a memory; and
a processor coupled to the memory and configured to:
obtain one or more images captured by an in-vehicle camera;
extract one or more feature points from the one or more images;
identify first feature points that exist over a road surface from the one or more feature points;
calculate a height of the in-vehicle camera from the road surface, based on positions of the identified first feature points; and
calculate a distance between an object and the in-vehicle camera based on the calculated height of the in-vehicle camera.

Assignees

• Fujitsu Ltd

Inventors

• Kanetake Jun
• Hasada Rie
• Ishida Haruyuki
• Fujita Takushi

Classifications

• G06V10/44

  Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components · CPC title

• G06V20/56 Primary

  exterior to a vehicle by using sensors mounted on the vehicle · CPC title

• G06T7/73 Primary

  using feature-based methods · CPC title

• G06T2207/30252

  Vehicle exterior; Vicinity of vehicle · CPC title