Enhanced analysis for image-based serpentine belt wear evaluation

• US 9,098,914 B2
• Filed: 10/16/2013
• Issued: 08/04/2015
• Est. Priority Date: 03/11/2013
• Status: Expired due to Fees

• Current Assignee: Gates Corporation (Gates Industrial Corp. Plc)
• Original Assignee: Gates Corporation (Gates Industrial Corp. Plc)
• Inventors: Sobczak, Flloyd M., Sealey, James H., Sedlacek, Douglas R., Stuemky, Mark E., Aschenbrenner, Justin, Bourque, James Christian
• Primary Examiner(s): Harandi, Siamak
• Application Number:
  US14/055,156
• Publication Number:
  US 20140254862A1
• Time in Patent Office: 657 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/20132 Image cropping
• CPC Class Codes: G06T 2207/30164 Workpiece; Machine component
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: G06T 7/0004 Industrial image inspection
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/73 using feature-based methods

Abstract

Systems and methods are provided for the improvement of an image of a device under test, such as a belt. The image of device under test is made more optimal by determining if the object is rotated away from a preferred axis of the image frame. If so, the image is rotated an opposing angle such that the object is parallel to the preferred axis of the image frame. The rotated image is then made available for analysis of the object. Rib width analysis is performed along the entire length of the detected rib by either de-rotating the image or not.

Claims

1. A method of processing an image of an object under test to be analyzed, the method comprising:

acquiring an image of the object under test, the image having a frame with a preferred axis and comprising a number of pixels;

identifying, by a processor, an indicia of the angle of the object under test within the image relative to the preferred axis, comprising;

determining a number of edge pixels forming a number of sets of the edge pixels, the number of edge pixels being determined from a number of pixels of the image;

determining a number of indicia candidate regions, each candidate region forming a polygon bounded by at least one of the number of sets of the edge pixels; and

determining the indicia from at least two of the number of indicia candidate regions; and

determining the angle between the preferred axis and the indicia;

rotating the image by the negative value of the angle;

performing the analysis on the object under test without rotating the image; and

presenting a result of the analysis on a user device.

2. The method of claim 1, wherein determining the indicia, further comprises determining the indicia from at least one of an average, mean, and mode of the number of indicia candidate regions.

3. The method of claim 1, wherein determining the indicia from the at least one of the number of indicia candidate regions, further comprises determining the indicia from the number of indicia candidate regions indicating at least one edge of the object under test.

4. The method of claim 1, further comprising:

upon determining, at least one of, the quantity of the number of edge pixels is outside of a previously determined first range and the quantity of the determined number of indicia candidate regions is out of a previously determined second range;

altering the image to improve the quantity of the number of edge pixels, when the quantity of the number of edge pixels is outside of the previously determined first range, and improve the quantity of edge candidates, when the quantity of the number of indicia candidate regions is out of a previously determined second range; and

determining which, of the number of pixels of the altered image, are the number of edge pixels.

5. The method of claim 4, wherein altering the image comprises altering the image by changing at least one of resolution, contrast, brightness, at least one color value, gamma, or sharpness.

6. The method of claim 4:

wherein the step of acquiring an image comprises acquiring an image at a first resolution; and

wherein altering the image comprises altering the image by changing the resolution to substantially negate the first resolution.

7. The method of claim 1, wherein determining which, of the number of pixels, are the number of edge pixels, further comprises, performing a pixel neighborhood analysis on the number of pixels to identify the number of edge pixels.

8. The method of claim 1, wherein the object under test is a belt.

9. The method of claim 1, wherein at least one of the number of sets of edge pixels comprises a delineation of a rib of a belt.

10. An image-based evaluation system, comprising:

an image processor, configured to create a more optimal image from a captured image of an object under test by rotating an image of the object under test to substantially align a preferred axis of image of the object under test to align with a preferred axis of the image wherein the angle of rotation is determined by the processor determining a number of edge pixels forming a number of sets of the edge pixels, the number of edge pixels being determined from a number of pixels of the image, determining a number of indicia candidate regions, each candidate region forming a polygon bounded by at least one of the number of sets of the edge pixels, and determining the indicia from at least two of the number of indicia candidate regions;

an analysis module, configured to evaluate the more optimal image and output the results of the evaluation; and

a storage medium, configured to store the more optimal image.

11. The system of claim 10, further comprising, an image-capture device configured to capture the image of the object under test.

12. The system of claim 10, further comprising a user interface configured to receive input commands to the system from a human user, display the more optimal image, and provide the input commands to the image processor.

13. The system of claim 12, whereby the image processor further comprises:

an image rotation detection processor, configured to determine the angle of the object under test to a frame of the captured image; and

an image rotation processor, configured to rotate the captured image by the negative value of the determined angle of the object under test for display as the more optimal image.

14. The system of claim 13, whereby the image rotation detection processor further comprises identifying an indicia of the angle of the object under test to a preferred axis of the frame of the captured image.

15. The system of claim 10, further comprising, an image enhancement processor, configured to increase the perceived detail of the object under test in the image.

16. The system of claim 10, wherein the image analysis module, is further configured to analyze the object under test from the captured image to determine the remaining service life for the object under test.

17. A non-transitory computer-readable medium storing a program that, when executed, causes a computing device to execute the process comprising:

acquiring an image of an object under test;

identifying an indicia of the angle of the object under test relative to a preferred axis of the image by determining a number of edge pixels forming a number of sets of the edge pixels, the number of edge pixels being determined from a number of pixels of the image, determining a number of indicia candidate regions, each candidate region forming a polygon bounded by at least one of the number of sets of the edge pixels, and determining the indicia from at least two of the number of indicia candidate regions;

determining the angle between the preferred axis and the indicia;

rotating the image by the negative value of the angle;

analyzing the image of the object under test; and

saving the rotated image.

18. The non-transitory computer-readable medium of claim 17, whereby identifying an indicia, further comprises:

identifying pixel neighborhoods that indicate a number of edge pixels;

identifying a number of candidate lines from a number of sets of the number of edge pixels; and

identifying the indicia from at least one of the number of indicia candidate lines.

19. The non-transitory computer-readable medium of claim 17, whereby the identifying the indicia further comprises, identifying a first and second edge of the object under test.

20. The non-transitory computer-readable medium of claim 19, further comprising cropping the rotated image to substantially exclude portions of the image not between the first and second edge.

21. The non-transitory computer-readable medium of claim 19, further comprising cropping the rotated image to include a crop buffer beyond the first and second edge.

22. The non-transitory computer-readable medium of claim 21, wherein the crop buffer is substantially ten percent larger than the cropped rotated image.