Machine Vision Vehicle Wheel Alignment Image Processing Methods

• US 20060274302A1
• Filed: 05/31/2006
• Published: 12/07/2006
• Est. Priority Date: 06/01/2005
• Status: Active Grant

• Current Assignee: Hunter Engineering Company
• Original Assignee: Hunter Engineering Company
• Inventors: Shylanski, Mark S., Golab, Thomas J., Dorrance, Daniel R., Burns, Leigh R. Jr.
• Granted Patent: US 7,583,372 B2
• Time in Patent Office: Days
• US Class Current: 356/139.90
• CPC Class Codes: G01B 11/2755 using photoelectric detection means
• CPC Class Codes: G01B 2210/146 Two or more cameras imaging the same area
• CPC Class Codes: G06T 7/0006 using a design-rule based approach
• CPC Class Codes: G06T 7/593 from stereo images

Abstract

A method of the present invention provides methods for processing images of vehicle wheel assemblies acquired during vehicle service procedures to determine vehicle wheel assembly pose components, from which vehicle wheel alignment angles may be calculated.

Claims

1. A method for determining at least one pose component associated with a vehicle wheel assembly, comprising:

acquiring at least two images of at least a portion of the vehicle wheel assembly;

identifying a corresponding area of interest in each of said images;

generating multi-dimensional data representative of each of said identified corresponding areas of interest;

correlating said multi-dimensional data to a mathematical model; and

wherein said mathematical model represents at least one pose component of a surface having symmetry about an axis.

2. The method of claim 1 wherein an axis of symmetry of said mathematical model represents an estimate of a vehicle wheel assembly axis of rotation.

3. The method of claim 1 wherein the pose of a vehicle wheel assembly is represented as a mathematical transform between at least one measurement device and the vehicle wheel assembly.

4. The method of claim 1 further including the step of projecting said multi-dimensional data onto a plane to generate a representation of at least a portion of a vehicle wheel assembly cross section.

5. The method of claim 4 further including the step of utilizing an iterative optimization process to identify a vehicle wheel assembly axis of symmetry from said generated representation.

6. The method of claim 1 wherein the said mathematical model represents a planar surface;

and

wherein a normal vector to the said planar surface represents an estimate of a vehicle wheel assembly axis of rotation.

7. A method for determining at least one pose component associated with a vehicle wheel assembly, comprising:

acquiring at least two images of at least a portion of the vehicle wheel assembly;

identifying a corresponding area of interest in each of said images;

generating multi-dimensional data representative of said identified corresponding areas of interest; and

correlating said multi-dimensional data to a mathematical model in the form of at least one parametrically constructed equation, said mathematical model representing at least one component of pose associated with a portion of the vehicle wheel assembly.

8. The method of claim 7 wherein the said mathematical model is in the form of at least one hyperquadric equation.

9. The method of claim 7 wherein the said mathematical model is in the form of at least one superquadric equation.

10. A method for determining at least one pose component associated with a vehicle wheel assembly, comprising:

acquiring at least two images of at least a portion of the vehicle wheel assembly;

identifying a corresponding area of interest in each of said images, said corresponding areas of interest represented as a collection of pixels in each of said acquired images;

generating multi-dimensional data representative of said identified corresponding areas of interest; and

utilizing said generated multi-dimensional data to determine a pose component associated with the vehicle wheel assembly.

11. The method of claim 10 for determining at least one pose component associated with a vehicle wheel assembly wherein said multi-dimensional data includes a planar surface representation of each of said identified corresponding areas of interest.

12. A method for determining a pose component associated with a vehicle wheel assembly for use in a vehicle service procedure, comprising:

acquiring a plurality of images of at least a portion of the vehicle wheel assembly;

identifying at least one area of interest in each of at least two of said acquired images;

determining a representation of correspondence between an area of interest in a first image and an area of interest in a second image; and

determining at least one pose component of the vehicle wheel assembly utilizing said representation of correspondence.

13. A method for calculating at least one alignment angle of a vehicle wheel assembly, comprising:

acquiring first and second images of at least a portion of the vehicle wheel assembly;

generating a multi-dimensional depth map of surface features of the vehicle wheel assembly from said first image; and

determining a change in at least one pose component associated with the vehicle wheel assembly between said first image and said second image utilizing said generated multi-dimensional depth map.

14. The method of claim 13 further including the steps of iteratively generating a representation of the vehicle wheel assembly utilizing said multi-dimensional depth map and comparing said representation of the vehicle wheel assembly with said second image of the vehicle wheel assembly to identify at least one pose component associated with the vehicle wheel assembly in said second image.

15. The method of claim 13 wherein the said multi-dimensional depth map includes at least one representation of a range and a brightness associated with an area of interest.

16. The method of claim 13 further including the step of mathematically transforming said multi-dimensional depth map.

17. The method of claim 16 wherein the said transformation of said multi-dimensional depth map produces an estimated image of the said area of interest.

18. The method of claim 16 wherein the said mathematical transformation of said multi-dimensional depth map is based on an estimate of a pose of the vehicle wheel assembly in three dimensional space.

19. The method of claim 18 further including the step of iteratively optimizing said pose estimate using the said mathematically transformed multi-dimensional depth map to calculate best-fit parameters.

20. A method for determining during a vehicle service procedure, at least one pose component associated with a vehicle wheel assembly, comprising:

acquiring at least one image of at least a portion of the vehicle wheel assembly;

generating a plurality of multi-dimensional data points representative of at least a portion of said vehicle wheel assembly in said acquired image; and

correlating said plurality of multi-dimensional data points to a mathematical model;

wherein a pose component of said mathematical model represents a pose component associated with said vehicle wheel assembly.

21. The method of claim 20 for determining at least one pose component associated with a vehicle wheel assembly wherein said step of mapping includes iteratively fitting said plurality of multi-dimensional data points to said mathematical model at a plurality of poses to identify a best-fit pose of said mathematical model with said plurality of multi-dimensional data points.

22. The method of claim 20 for determining at least one pose component associated with a vehicle wheel assembly wherein said plurality of multi-dimensional data points includes a point cloud.