Automatic camera calibration

• US 8,619,144 B1
• Filed: 03/14/2012
• Issued: 12/31/2013
• Est. Priority Date: 03/14/2012
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Rawles, LLC (Amazon.com, Inc.)
• Inventors: Crump, Edward Dietz, Chang, Samuel Henry, Puri, Manika, Gonzalez, Steve G.
• Primary Examiner(s): Hsia, Sherrie
• Application Number:
  US13/419,857
• Time in Patent Office: 657 Days
• Field of Search: 348/722, 348/187, 348/188, 348/180
• US Class Current: 348/187
• CPC Class Codes: G06T 7/80 Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
• CPC Class Codes: H04N 17/002 for television cameras

Abstract

Devices and techniques are described for automatically calibrating a camera system. The camera system undergoing calibration is coupled to a computing device and an automated positioning platform coupled to a target structure. The computing device acquires images from the camera of the target structure in a plurality of repeatable poses. From these acquired images, intrinsic camera parameters may be determined. Once determined, the parameters may be used to correct images acquired by the camera system.

Claims

1. A system comprising:

a target structure comprising one or more fiducial features;

an automated positioning platform coupled to the target structure and under control of one or more computing systems, the positioning platform configured to orient the target structure to a plurality of poses by moving in one or more degrees of freedom at least partly in response to receiving corresponding instructions from the one or computing systems;

a camera configured to acquire one or more acquired images of the target structure in each of the plurality of poses;

a calibration module configured to;

determine an observed location of the one or more fiducial features within at least a portion of the one or more acquired images of the target structure in the plurality of poses; and

generate calibration data associated with the camera based at least in part on the observed location of the one or more fiducial features in the acquired images.

2. The system of claim 1, wherein the one or more fiducial features comprise a checkerboard pattern.

3. The system of claim 2, wherein the checkerboard pattern comprises squares of two or more sizes.

4. The system of claim 1, wherein the positioning platform is configured to vary a linear target distance between the target structure and the camera before acquisition of the image of the target structure and at least partly in response to receiving a corresponding instruction from the one or more computing systems.

5. The system of claim 4, wherein the linear target distance is adjusted such that the target structure substantially fills a field-of-view of the camera.

6. The system of claim 1, wherein the positioning platform comprises a Stewart platform.

7. The system of claim 1, wherein the positioning platform comprises two or more rotary stages coupled to one another.

8. The system of claim 1, the calibration data comprising one or more intrinsic camera parameters of the camera.

9. The system of claim 8, the one or more intrinsic camera parameters comprising a focal point, a principal point, or both a focal point and a principal point.

10. The system of claim 1, the calibration data comprising one or more extrinsic camera parameters of the camera.

11. One or more computer-readable non-transitory storage media storing instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:

setting, to a pose, a positioning platform coupled to a target presenting one or more fiducial features;

acquiring, with a camera, a set of one or more images of the target while the positioning platform is in the pose;

selecting an image from the set of one or more images meeting one or more criteria;

determining an observed location of one or more fiducial features in the selected image; and

generating calibration data associated with the camera based at least in part on the observed location of the one or more fiducial features.

12. The one or more computer-readable non-transitory storage media of claim 11, further comprising:

until a pre-determined level of accuracy in the calibration data has been reached;

setting the positioning platform to an nth pose,

acquiring, with the camera a second set of one or more images of the target while the positioning platform is in the nth pose;

selecting a second selected image from the nth set of one or more images meeting one or more criteria;

determining an observed location of one or more fiducial features in the nth selected image; and

generating calibration data associated with the camera based at least in part on the observed location of the one or more fiducial features in the nth selected image.

13. The one or more computer-readable non-transitory storage media of claim 11, the determining of the location of the one or more fiducial features in the selected image comprising identifying a plurality of corners of the one or more fiducial features.

14. The one or more computer-readable non-transitory storage media of claim 11, the one or more criteria comprising one or more of:

fit of lines to a pre-determined value through a pattern on the target,

an amount of image distortion, or

an amount of image clarity.

15. The one or more computer-readable storage media of claim 11, further comprising estimating one or more intrinsic camera parameters for the camera based at least in part on the calibration data.

16. One or more computer-readable non-transitory storage media storing instructions that, when executed by one or more processors, cause the one or more processors to perform acts comprising:

directing a camera target coupled to a positioning platform between a plurality of poses by adjusting the positioning platform in a linear motion;

acquiring with a camera one or more images of the camera target in the plurality of poses; and

generating camera calibration data based at least in part on features present in the camera target.

17. The one or more computer-readable non-transitory storage media of claim 16, wherein the plurality of poses further comprise different relative angles between the target and the camera.

18. The one or more computer-readable non-transitory storage media of claim 16, wherein the features comprise one or more fiducial features.

19. The one or more computer-readable non-transitory storage media of claim 16, wherein the generating the camera calibration data comprises determining an observed location of one or more of the features.

20. The one or more computer-readable non-transitory storage media of claim 16, wherein the camera calibration data comprises one or more of the following intrinsic camera parameters:

a focal point,

a principal point,

an optical skew coefficient, or

an optical distortion coefficient.