System and method for registering patterns transformed in six degrees of freedom using machine vision

• US 6,771,808 B1
• Filed: 12/15/2000
• Issued: 08/03/2004
• Est. Priority Date: 12/15/2000
• Status: Expired due to Fees

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Wallack, Aaron S.
• Primary Examiner(s): WERNER, BRIAN P
• Application Number:
  US09/737,836
• Time in Patent Office: 1,327 Days
• Field of Search: 382/151, 382/199, 382/215, 382/216, 382/289, 382/290, 382/294, 348/87, 348/95, 348/126
• US Class Current: 382/151
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 10/245 by locating a pattern; Special marks for positioning
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors

Abstract

A system and method for utilizing a search tool that registers transformation of a trained pattern by at least four degrees of freedom to register the instance of a pattern in an arbitrary six-degree-of-freedom pose is provided. The search tool is first trained to recognize a plurality of versions of a trained pattern/fiducial that are incrementally transposed to induce differing levels of known aspect and shear. An object having several instances of the trained pattern located at known spacings and orientations therebetween is imaged by the trained search tool, and the located instances of the trained pattern are compared to expected instances of the trained pattern to measure relative six-degree-of-freedom orientation for the underlying object.

Claims

1. A system for registering an object in six degrees of freedom using a machine vision system comprising:

a search tool of the machine vision system adapted to recognize a plurality of instances of a trained pattern, the plurality of instances each being transformed to exhibit different amounts of aspect and shear and to provide a plurality of search results corresponding, respectively to the plurality of instances of the trained pattern; and

a combiner that combines the plurality of search results, wherein found relative positions of the instances of the trained pattern are compared with known relative positions of the instances of the trained pattern so as to provide a location of the object in the six degrees of freedom.

2. The system as set forth in claim 1 wherein the combiner includes a linear transform between the expected relative position of each of the instances of the trained pattern and a normalized measured position of the instances of the trained pattern.

3. The system as set forth in claim 1 wherein the plurality of instances of the trained pattern comprise a plurality of transposed, synthetically generated image data, and the different amounts of aspect and shear are based upon predetermined known increments.

4. The system as set forth in claim 1 wherein the plurality of instances of the trained pattern comprise a plurality of different user-specified values for aspect and shear provided at runtime to the search tool so as to change an orientation of the trained pattern.

5. The system as set forth in claim 1 wherein the plurality of instances of the trained pattern each comprise portions of an overall pattern.

6. A method for registering an object in six degrees of freedom using a machine vision system comprising:

recognizing, with a search tool of the machine vision system, a plurality of instances of a trained pattern, the plurality of instances each being transformed to exhibit different amounts of aspect and shear;

providing, from the search tool, a plurality of search results corresponding, respectively, to the plurality of instances of the trained pattern; and

combining the plurality of search results, including comparing found relative positions of the instances of the trained pattern with known relative positions of the instances of the trained pattern so as to provide a location of the object in the six degrees of freedom.

7. The method as set forth in claim 6 wherein the step of combining includes applying a linear transform between the expected relative position of each of the instances of the trained pattern and a normalized measured position of the instances of the trained pattern.

8. The method as set forth in claim 6 further comprising scoring each of the plurality of search results and selecting a best scoring of the search results for combining by the step of combining.

9. The method as set forth in claim 6 wherein the plurality of instances of the trained pattern comprise a plurality of transposed, synthetically generated image data, and the different amounts of aspect and shear are based upon predetermined known increments.

10. The method as set forth in claim 6 wherein the plurality of instances of the trained pattern comprise a plurality of different user-specified values for aspect and shear provided at runtime to the search tool so as to change an orientation of the trained pattern.

11. The method as set forth in claim 6 wherein the plurality of instances of the trained pattern each comprise portions of an overall pattern.

12. A computer-readable medium including program instructions executed on a computer for registering an object in six degrees of freedom using a machine vision system, the computer-readable medium including program instructions for performing the steps of:

recognizing, with a search tool of the machine vision system, a plurality of instances of a trained pattern, the plurality of instances each being transformed to exhibit different amounts of aspect and shear;

providing, from the search tool, a plurality of search results corresponding, respectively, to the plurality of instances of the trained pattern; and

combining the plurality of search results, including comparing found relative positions of the instances of the trained pattern with known relative positions of the instances of the trained pattern so as to provide a location of the object in the six degrees of freedom.

13. The computer-readable medium as set forth in claim 12 wherein the step of combining includes applying a linear transform between the expected relative position of each of the instances of the trained pattern and a normalized measured position of the instances of the trained pattern.

14. The computer-readable medium as set forth in claim 12 further comprising scoring each of the plurality of search results and selecting best scoring of the search results for combining by the step of combining.

15. The computer-readable medium as set forth in claim 12 wherein the plurality of instances of the trained pattern comprise a plurality of transposed, synthetically generated image data, and the different amounts of aspect and shear are based upon predetermined known increments.

16. The computer-readable medium as set forth in claim 12 wherein the plurality of instances of the trained pattern comprise a plurality of different user-specified values for aspect and shear provided at runtime to the search tool so as to change an orientation of the trained pattern.

17. The computer-readable medium as set forth in claim 12 wherein the plurality of instances of the trained pattern each comprise portions of an overall pattern.