System and method for determining clutter in an acquired image

• US 10,192,283 B2
• Filed: 12/22/2014
• Issued: 01/29/2019
• Est. Priority Date: 12/22/2014
• Status: Active Grant

• Current Assignee: Cognex Corporation
• Original Assignee: Cognex Corporation
• Inventors: Davis, Jason, Michael, David J., Bogan, Nathaniel R.
• Primary Examiner(s): Kholdebarin, Iman K
• Application Number:
  US14/580,153
• Publication Number:
  US 20160180198A1
• Time in Patent Office: 1,499 Days
• Field of Search: None
• CPC Class Codes: G06T 3/00 Geometric image transformations in the plane of the image
• CPC Class Codes: G06T 7/001 using an image reference approach
• CPC Class Codes: G06T 7/40 Analysis of texture depth or shape recovery from texture G06T7/529
• CPC Class Codes: G06V 10/225 based on a marking or identifier characterising the area
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 10/993 Evaluation of the quality of the acquired pattern
• CPC Class Codes: G06V 2201/01 Solutions for problems related to non-uniform document background

Abstract

This invention provides a system and method for determining the level of clutter in an image in a manner that is rapid, and that allows a scoring process to quickly determine whether an image is above or below an acceptable level of clutter—for example to determine if the underlying imaged runtime object surface is defective without need to perform a more in-depth analysis of the features of the image. The system and method employs clutter test points that are associated with regions on the image that should contain a low gradient magnitude, indicative of emptiness. This enables the runtime image to be analyzed quickly by mapping trained clutter test points at locations in the coordinate space in which lack of emptiness indicates clutter, and if detected, can rapidly indicate differences and/or defects that allow for the subject of the image to be accepted or rejected without further image analysis.

Claims

1. A method for determining clutter in an imaged scene with a vision system comprising the steps of:

providing a runtime image of a scene;

providing a trained pattern with a set of clutter test points based upon sub-pixel locations that represent a level of emptiness in the trained pattern;

establishing a runtime pose with a coordinate space for the image with respect to the trained pattern;

mapping the clutter test points on the coordinate space for the image;

determining a level of emptiness respectively at the mapped clutter test points; and

based upon the level of emptiness, determining a level of clutter in at least a portion of the image.

2. The method as set forth in claim 1 wherein the step of determining the level of emptiness comprises generating respective clutter score values for the mapped clutter test points.

3. The method as set forth in claim 2 further comprising summing the clutter score values and generating an overall clutter score value for at least the portion of the image.

4. The method as set forth in claim 1 wherein the step of determining the level of emptiness comprises determining a gradient magnitude respectively at the mapped clutter test points and comparing the magnitude to a predetermined threshold.

5. The method as set forth in claim 1 wherein the step of establishing the pose comprises at least one of (a) employing alignment tools in a vision system to automatically align the runtime image to the trained pattern and (b) obtaining information indicating the pose from a user input.

6. A method for extracting a training pattern for determining a level of clutter in at least a portion of an imaged scene with a vision system comprising the steps of:

providing a training image with a feature of interest;

providing a predetermined clutter threshold value relative to a level of emptiness that indicates clutter-free regions in the image; and

establishing clutter test points based upon sub-pixel locations with respect to a coordinate space of the training image based upon locations in the coordinate space that represent the level of emptiness that is free-of clutter.

7. The method as set forth in claim 6 wherein the training image includes a mask that indicates which areas of the training image should be evaluated for emptiness.

8. The method as set forth in claim 6 where the training image includes features specified by description free of reliance on pixel values.

9. The method as set forth in claim 6 wherein the step of establishing the clutter test points includes determining a gradient magnitude in regions of the training image less than a predetermined threshold value.

10. The method as set forth in claim 6 further comprising storing the clutter test points in the coordinate space of the trained image for use in clutter determination in a runtime image.

11. The method as set forth in claim 6 wherein the clutter threshold value is computed based upon a histogram of gradient values within the training image.

12. The method as set forth in claim 6 wherein the clutter threshold is provided as an input parameter.

13. A system for determining a level of clutter in a runtime candidate image based on a trained pattern using a vision system comprising:

a training time clutter point generator that generates an array of clutter test points based upon sub-pixel locations relative to a coordinate space in a training image having information based upon locations in the training image that have a level of emptiness below a clutter threshold, the clutter threshold being established based upon predetermined parameters and the information in the training image; and

a runtime clutter determiner that maps the locations of the clutter test points to the runtime candidate image based upon a runtime pose, and that computes a level of emptiness at each of the locations to determine the level of clutter in at least a portion of the acquired image.

14. The system as set forth in claim 13 wherein the level of emptiness in the runtime candidate image is based upon a determined gradient magnitude at the locations, respectively.

15. The system as set forth in claim 13 further comprising a clutter utilizer that utilizes determined clutter level to perform a predetermined action with respect to a surface associated with the runtime candidate image based upon the determined level of the clutter.

16. The system as set forth in claim 15 wherein the predetermined action includes at least one of part reject, issuing an alert, transmitting quality data and stopping a moving line.

17. The system as set forth in claim 13 wherein at least one of the predetermined parameters is input by a user.

18. The system as set forth in claim 13 wherein the clutter determiner creates a clutter score at each of the locations.

19. The system as set forth in claim 18 wherein the clutter determiner sums the clutter score for each of the locations to generate an overall clutter score for the runtime candidate image.

20. The system as set forth in claim 13 wherein the information includes a feature of interest and further comprising an alignment tool that determines the runtime pose based upon a location of at least a portion of the feature of interest in each of the training image and the runtime candidate image.