System and a method for detecting a damaged or missing machine part

• US 8,411,930 B2
• Filed: 05/04/2007
• Issued: 04/02/2013
• Est. Priority Date: 05/12/2006
• Status: Active Grant

• Current Assignee: Motion Metrics International Corp (WEIR Group PLC)
• Original Assignee: Alberta Research Council Inc. (Alberta Science & Research Authority)
• Inventors: Ridley, Rodney K., Kazmierczak, Tadeusz, Cai, Linsong, Johnston, Paul Charles, Pinto, Helen, Sun, Jun, Provencher, Marc, Woolley, John Harry, Cardinal, Garry
• Primary Examiner(s): Chowdhury, Tarifur
• Assistant Examiner(s): HANSEN, JONATHAN M
• Application Number:
  US12/300,257
• Publication Number:
  US 20100142759A1
• Time in Patent Office: 2,160 Days
• Field of Search: 382/209, 382/217, 382/218, 382/141, 382/152
• US Class Current: 382/152
• CPC Class Codes: E02F 9/267 Diagnosing or detecting failure of vehicles
• CPC Class Codes: E02F 9/2808 Teeth
• CPC Class Codes: G06T 7/0008 checking presence/absence

Abstract

A system and a method for detecting a damaged or missing machine part. The system includes an image capturing device for capturing images of the machine and a processor for processing the captured images. The system may further include a sensible output for providing an indication of a damaged or missing machine part. The method includes capturing images of the machine against a background which moves relative to the machine over time, selecting a pair of time-separated images from the captured images, generating a displacement image from the pair of images, comparing the machine from the displacement image with a machine model, and identifying a damaged or missing machine part from the comparison of the displacement image with the machine model. The method may further include providing a sensible output which indicates a damaged or missing machine part.

Claims

1. A method for detecting a damaged or missing machine part, the method comprising:

(a) capturing images of a machine and a background, wherein the background moves relative to the machine over time

(b) selecting time-separated images which reflect movement of the background relative to the machine;

(c) generating a displacement image from the selected time-separated images, wherein the displacement image provides an indication of movement of each pixel represented in the displacement image between the selected time-separated images;

(d) comparing the machine from the displacement image with a model machine; and

(e) identifying a damaged or missing machine part from the comparison of the machine from the displacement image and the model machine.

2. The method as claimed in claim 1, further comprising providing a sensible output which indicates the damaged or missing machine part.

3. A system for detecting whether a machine part of a machine is damaged or missing, the system comprising:

(a) an image capturing device for capturing time-separated images of the machine and a background, wherein the background moves relative to the machine over time, and wherein the captured time-separated images are comprised of pixels; and

(b) a processor configured to generate a displacement image from time-separated images selected from the captured time-separated images, wherein the displacement image provides an indication of movement of the pixels between the selected time-separated images, and configured to identify from the displacement image whether the machine part is damaged or missing.

4. The system as claimed in claim 3, further comprising a sensible output fbr providing an indication if the machine part is damaged or missing.

5. The system as claimed in claim 3 wherein the processor is further configured to select the time-separated images such that the selected time-separated images reflect movement of the background relative to the machine.

6. The system as claimed in claim 5 wherein the processor is configured to select the time-separated images by initial processing one or more pairs of the captured time-separated images.

7. The system as claimed in claim 6 wherein the initial processing is comprised of determining whether the machine can be adequately identified in the pairs of captured time-separated images.

8. The system as claimed in claim 7 wherein the machine can be adequately identified in the pairs of captured time-separated images if the machine is within a defined position zone of the captured time-separated images and if the machine is within a defined size range in the captured time-separated images.

9. The system as claimed in claim 6 wherein the initial processing is comprised of determining whether the background is moving relative to the machine in the pairs of captured time-separated images.

10. The system as claimed in claim 6 wherein the initial processing is comprised of determining whether a magnitude of movement of the background relative to the machine in the pairs of captured time-separated images is within a desired range.

11. The system as claimed in claim 3 wherein the processor is configured to identify from the displacement image whether the machine part is damaged or missing by comparing the machine from the displacement image with a model machine to provide a comparison, and identifying from the comparison whether the machine part is damaged or missing.

12. A method for detecting whether a machine part of a machine damaged or missing, the method comprising:

(a) capturing time-separated images of the machine and a background, wherein the background moves relative to the machine over time, and wherein the captured time-separated images are comprised of pixels;

(b) selecting time-separated images which reflect movement of the background relative to the machine;

(c) generating a displacement image from the selected time-separated images, wherein the displacement image provides an indication of movement of the pixels between the selected time-separated images;

(d) identifying from the displacement image whether the machine part is damaged or missing.

13. The method as claimed in claim 12, further comprising providing a sensible output which indicates the damaged or missing machine part.

14. The method as claimed in claim 12 wherein selecting the time-separated images is comprised of initial processing one or more pairs of the captured time-separated images.

15. The method as claimed in claim 14 wherein the initial processing is comprised of determining whether the machine can be adequately identified in the pairs of captured time-separated images.

16. The method as claimed in claim 15 wherein the machine can be adequately identified in the pairs of captured time-separated images if the machine is within a defined position zone of the captured time-separated images and if the machine is within a defined size range in the captured time-separated images.

17. The method as claimed in claim 14 wherein the initial processing is comprised of determining whether the background is moving relative to the machine in the pairs of captured time-separated images.

18. The method as claimed in claim 14 wherein the initial processing is comprised of determining whether a magnitude of movement of the background relative to the machine in the pairs of captured time-separated images is within a desired range.

19. The method as claimed in claim 12 wherein the machine is a bucket and wherein the machine part is a tooth on the bucket.

20. The method as claimed in claim 12 wherein identifying whether the machine part is damaged or missing is comprised of comparing the machine from the displacement image with a model machine to provide a comparison, and identifying, from the comparison whether the machine pan is damaged or missing.

21. A method for detecting whether a machine part of a machine is damaged or missing, the method comprising:

(a) capturing time-separated images of the machine and a background, wherein the background moves relative to the machine over time, and wherein the captured time-separated images are comprised of pixels;

(b) initial processing one or more pairs of the captured time-separated images to select time-separated images from the captured time-separated images;

(c) generating a displacement image from the selected tune-separated images, wherein the displacement image provides an indication of movement of the pixels between the selected time-separated images;

(d) locating a first boundary of the machine in the displacement image;

(e) locating a second boundary of the machine in the displacement image;

(f) creating a model machine using the first boundary and the second boundary;

(g) comparing the machine from the displacement image with the model machine to provide a comparison; and

(h) identifying from the comparison whether the machine part is damaged or missing.

22. The method as claimed in claim 21, further comprising providing a sensible output which indicates the damaged or missing machine part.

23. The method as claimed in claim 22 wherein the machine is a bucket and wherein the machine part is a tooth on the bucket.