AUTOMATIC FEATURE POINT DETECTION FOR CALIBRATION OF MULTI-CAMERA SYSTEMS

• US 20170124710A1
• Filed: 10/14/2016
• Published: 05/04/2017
• Est. Priority Date: 10/30/2015
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: KWON, Do-Kyoung, BATUR, Aziz Umit, APPIA, Vikram
• Granted Patent: US 10,438,081 B2
• Time in Patent Office: Days
• CPC Class Codes: G06T 2207/30204 Marker
• CPC Class Codes: G06T 2207/30248 Vehicle exterior or interior
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
• CPC Class Codes: G06V 10/245 by locating a pattern; Special marks for positioning
• CPC Class Codes: G06V 10/462 Salient features, e.g. scale invariant feature transforms [SIFT]
• CPC Class Codes: H04N 23/698 for achieving an enlarged field of view, e.g. panoramic image capture

Abstract

In some embodiments, a computer-readable medium stores executable code, which, when executed by a processor, causes the processor to: capture an image of a finder pattern using a camera; locate a predetermined point within the finder pattern; use the predetermined point to identify multiple boundary points on a perimeter associated with the finder pattern; identify fitted boundary lines based on the multiple boundary points; and locate feature points using the fitted boundary lines.

Claims

1. A computer-readable medium storing executable code, which, when executed by a processor, causes the processor to:

capture an image of a finder pattern using a camera;

locate a predetermined point in the finder pattern;

use the predetermined point to identify multiple boundary points on a perimeter associated with the finder pattern;

identify fitted boundary lines based on the multiple boundary points; and

locate feature points using the fitted boundary lines.

2. The computer-readable medium of claim 1, wherein the finder pattern comprises a first square positioned within a second square.

3. The computer-readable medium of claim 1, wherein the finder pattern comprises a first square positioned within a second square, and wherein the first and second squares are positioned within a third square.

4. The computer-readable medium of claim 3, wherein the processor binarizes the image, and, after binarization, the first and third squares share a common color that is different from a color of the second square.

5. The computer-readable medium of claim 1, wherein said image is a fish-eye lens image.

6. The computer-readable medium of claim 1, wherein, to identify said multiple boundary points, the processor identifies multiple, equally-spaced lines passing through the predetermined point and determines the points at which said multiple lines intersect said perimeter associated with the finder pattern.

7. The computer-readable medium of claim 6, wherein said perimeter is a perimeter of a square in the finder pattern.

8. The computer-readable medium of claim 6, wherein, to determine said points at which the multiple lines intersect the perimeter, the processor identifies at least one color transition.

9. A method, comprising:

capturing an image of a finder pattern using a first camera in a multi-camera system;

locating a predetermined point in the finder pattern;

locating multiple boundary points along a perimeter of the finder pattern using said predetermined point;

fitting boundary lines along the perimeter of the finder pattern using said multiple boundary points;

locating feature points along said perimeter using the fitted boundary lines; and

generating a perspective transformation matrix based on said feature points.

10. The method of claim 9, wherein said predetermined point of the finder pattern is the center of the finder pattern.

11. The method of claim 9, wherein the finder pattern comprises a first square centered within a second square.

12. The method of claim 11, further comprising binarizing the image, wherein, after said binarization is complete, one of the first and second squares is white and the other of the first and second squares is black.

13. The method of claim 12, wherein said binarized image includes said finder pattern positioned on an object, said object has a color matching that of the first square.

14. The method of claim 9, further comprising using the perspective transformation matrix to generate a bird'"'"'s-eye view of an object on which said multi-camera system is mounted.

15. The method of claim 9, wherein locating the predetermined point of the finder pattern comprises identifying a point in said finder pattern that lies along a line that intersects each perimeter of the finder pattern twice, wherein said line has multiple segments, each segment located between a different pair of consecutive perimeter crossings, and wherein a ratio of the lengths of said segments falls within a predetermined ratio range.

16. The method of claim 9, wherein locating said multiple boundary points comprises identifying points at which a line passing through the predetermined point intersects said perimeter of the finder pattern.

17. The method of claim 9, wherein said feature points are points at which said fitted boundary lines intersect.

18. A system, comprising:

a camera configured to capture an image of a finder pattern; and

a processor coupled to the camera and configured to identify a center of the finder pattern in the image, to locate one or more feature points of the finder pattern based on said center, and to calibrate the system using said one or more feature points.

19. The system of claim 18, wherein the processor is further configured to locate multiple boundary points along a perimeter associated with the finder pattern based on the center, wherein the processor locates said one or more feature points based on the boundary points.

20. The system of claim 19, wherein the process is further configured to identify multiple fitted boundary lines based on the multiple boundary points, and wherein the processor locates said one or more feature points based on the multiple fitted boundary lines.