Method for detecting an optical structure

• US 8,237,794 B2
• Filed: 08/17/2005
• Issued: 08/07/2012
• Est. Priority Date: 10/01/2004
• Status: Active Grant

• Current Assignee: Robert Bosch GmbH
• Original Assignee: Robert Bosch GmbH
• Inventors: Moritz, Rainer, Maass, Alexander, Schick, Jens, Wuerz-Wessel, Alexander
• Primary Examiner(s): Tiv, Backhean
• Assistant Examiner(s): Chou, Alan S
• Application Number:
  US11/664,347
• Publication Number:
  US 20070299596A1
• Time in Patent Office: 2,547 Days
• Field of Search: 348/148, 348/208.2, 348/208.99
• US Class Current: 348/148
• CPC Class Codes: G06T 7/215 Motion-based segmentation
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration

Abstract

A method for detecting an optical structure from image sequences of an image recording system. An optical flow field is derived from the image sequences of the image recording system. The vectors of the optical flow field are broken down into components by projection onto a coordinate system of the image recorded by the image recording system. At least one component of the flow vectors is analyzed for a change of direction. The image elements containing a change of direction of a vector component are connected to form a curve.

Claims

1. A method for detecting an optical structure from image sequences of an image recording system, comprising:

deriving an optical flow field from the image sequences of the image recording system;

breaking down flow vectors of the optical flow field into axial components by projection onto axes of a coordinate system of an image recorded by the image recording system;

for each line of image extending parallel to a selected axis of the coordinate system, examining axial components of the flow vectors extending parallel to the selected axis of the coordinate system for a change of direction; and

connecting image elements containing the change of direction of the axial components of the flow vectors extending parallel to the selected axis of the coordinate system, to form a curve.

2. The method as recited in claim 1, wherein the flow vectors of the optical flow field are projected onto an orthogonal coordinate system, a horizontal axis of the coordinate system is aligned parallel to lines of the image and vertical axis of the coordinate system is parallel to columns of the image.

3. The method as recited in claim 2, wherein vector components of the flow vectors running parallel to the vertical axis of the coordinate system are examined for the change of direction.

4. The method as recited in claim 2, wherein the method is used for calibrating an image recording system using the curve.

5. The method as recited in claim 2, wherein the method is used for detecting a roll angle or a roll rate of a vehicle using the curve.

6. A method for detecting an optical structure from image sequences of an image recording system, comprising:

deriving a first flow field from the image sequences of the image recording system;

deriving a second optical flow field from the image sequence reflected on a reflective surface and detected by the image recording system;

breaking down flow vectors of the first and second optical flow fields into axial components by projection onto axes of a coordinate system of an image recorded by the image recording system;

for each line of image extending parallel to a selected axis of the coordinate system, examining axial components of the flow vectors of the first and second optical flow fields extending parallel to the selected axis of the coordinate system for a change of direction; and

connecting image elements containing the change of direction of the axial components of the flow vectors extending parallel to the selected axis of the coordinate system, to form a curve.

7. The method as recited in claim 6, further comprising:

determining an edge or contour of a reflective surface by comparing the change of direction of vector components between flow vectors obtained from direct image sequences and reflected image sequences.

8. The method as recited in claim 7, wherein an edge or contour of the hood of a vehicle is detected.

9. The method as recited in claim 6, wherein each column of the image is examined for a presence of a vector component folded in its direction.

10. The method as recited in claim 6, wherein groups of columns of the image are examined for a presence of a vector component folded in its direction.

11. A method for calibrating an onboard image recording system, comprising:

recording image sequences from a detection range of the image recording system via the onboard image recording system, the detection range of the image recording system including a part of a vehicle structure of the vehicle;

deriving optical flow fields from the image sequences recorded by the image recording system;

breaking down flow vectors of the optical flow field into axial components by projection onto axes of a coordinate system of an image recorded by the image recording system;

for each line of image extending parallel to a selected axis of the coordinate system, examining axial components of the flow vectors of the optical flow fields extending parallel to the selected axis of the coordinate system for a change of direction;

connecting image elements containing the change of direction of the axial components of the flow vectors extending parallel to the selected axis of the coordinate system, to form a curve; and

comparing a position of the curve with a setpoint position.

12. The method as recited in claim 11, wherein the vehicle structure is a hood of the vehicle.