Method of pattern location using color image data

• US 7,965,887 B2
• Filed: 12/01/2005
• Issued: 06/21/2011
• Est. Priority Date: 12/01/2005
• Status: Active Grant

• Current Assignee: Cognex Technology & Investment Corporation (Cognex Corporation)
• Original Assignee: Cognex Technology & Investment Corporation (Cognex Corporation)
• Inventors: Michael, David, Wallack, Aaron S.
• Primary Examiner(s): Ahmed; Samir A
• Assistant Examiner(s): Liu; Li
• Application Number:
  US11/292,170
• Publication Number:
  US 20090141973A1
• Time in Patent Office: 2,028 Days
• Field of Search: 382/165, 382/190, 382/170, 382/181
• US Class Current: 382/165
• CPC Class Codes: G06V 10/28 Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors

Abstract

Machine vision tools are applied to color images using methods that utilize an optimized spectrum of the color information. Such methods include full color normalized correlation techniques and methods to convert color images to greyscale using weighting factors that maximize color contrast in a corresponding greyscale image.

Claims

1. An apparatus comprising:

a machine vision processor programmed to perform the steps of;

obtaining a color model image of an expected pattern;

obtaining a color run-time image including at least one instance of the pattern;

computing a dynamic weight for each of a plurality of color channels in the color model image by calculating a weighting factor for each color channel and subsequently evaluating the weighting factors by computing a variance in contrast;

computing a greyscale image of the color model image using the dynamic weights, and a greyscale image of the color run-time image using the dynamic weights; and

using the greyscale image of the color model image to determine a pose having at least one non-translational degree of freedom of the expected pattern in the greyscale run-time image of the pattern.

2. The apparatus claim 1 wherein the machine vision processor is further programmed to perform the step of using color information from the color model image to determine a pose in the color run time image having at least one non-translational degree of freedom of the expected pattern, the step of using the greyscale image including analyzing the greyscale run-time image using the pose in the color run time image.

3. The apparatus claim 2 wherein the color model image and the color run-time image each comprises a plurality of color channels, and the step of determining a pose comprises computing a vector representation of each of the color model image and the color run-time image.

4. The apparatus claim 2 wherein the color model image and the color run-time image each comprises a plurality of color channels, and the step of determining a pose comprises computing a vector representation of each of the color model image and the color run-time image.

5. The apparatus claim 2 wherein the step of using color information from the color model image to determine the pose further comprises using vector-based normalized correlation.

6. The apparatus claim 1 wherein the machine vision processor is further programmed to perform the step of analyzing the color run-time image using the pose that was determined using the greyscale image of the color model image.

7. The apparatus of claim 6 wherein the step of analyzing the color run-time image using the pose further comprises:

performing an affine transformation of the color model image using the pose; and

using the color model image to locate an instance of the pattern of the object in the color run-time image.

8. The apparatus of claim 1 wherein the machine vision processor is programmed to perform step of computing a dynamic weight by, for each channel, selecting a weighting factor as a function of the magnitudes of several variance in contrast values associated with different weighting factors.

9. The apparatus of claim 8 wherein the step of computing a dynamic weight includes selecting weighting factors that result in a highest variance in contrast value associated.

10. An apparatus comprising:

a machine vision processor programmed to perform the steps of;

obtaining a first color image that includes an instance of an expected pattern;

generating a plurality of greyscale images using a plurality of different channel weighting factor sets that represent a search space;

evaluating the weighting factor sets by computing a variance in contrast for each of the weighting factor sets;

selecting the weighting factor set associated with the largest variance in contrast as a set of dynamic weights.

11. The apparatus of claim 10 wherein the processor is further programmed to perform the steps of receiving a color run time image and using the set of dynamic weights to convert a color run time image to a greyscale run time image.

12. The apparatus of claim 11 wherein the first color image is a color model image and the greyscale image corresponding to the color model image and the set of dynamic weights is a greyscale model image, the processor further programmed to perform the steps of using the greyscale model image to identify at least one pose of the expected pattern in the greyscale run time image.