Algorithm for measuring wear pin length using an input image

Patent metadata

Field	Value
Publication number	US-9786042-B2
Application number	US-201514609368-A
Country	US
Kind code	B2
Filing date	Jan 29, 2015
Priority date	Jan 29, 2015
Publication date	Oct 10, 2017
Grant date	Oct 10, 2017

Abstract

In one example, the disclosure is directed to a method of determining a length of a wear pin in a brake assembly. The method includes obtaining an input image of a portion of the brake assembly, such as with a camera. The input image includes the wear pin and a reference object, and the reference object has a known dimension. A processor may determine, based on the input image, an image dimension of the reference object. The processor may determine, based on the input image, an image dimension of the wear pin. The processor may further determine, based on the image dimension of the reference object, the image dimension of the wear pin, and the known dimension of the reference object, an estimated measurement of the dimension of the wear pin.

First claim

The invention claimed is:

1. A method of determining a length of a wear pin in a brake assembly, the method comprising:
obtaining an input image of a portion of the brake assembly, wherein the input image comprises an image of the wear pin and a reference object, and wherein the reference object has a known dimension;
determining, based on the input image, an image dimension of the reference object;
determining, based on the input image, an image dimension of the wear pin; and
determining, based on the image dimension of the reference object, the image dimension of the wear pin, and the known dimension of the reference object, an estimated measurement of the dimension of the wear pin.

2. The method of claim 1 , wherein the reference object comprises one of a bracket configured to hold the wear pin, a bolt in the brake assembly, a distance between two or more bolts near the bracket, or a piston shaft.

3. The method of claim 1 , wherein obtaining the input image comprises:
receiving, by a server, the input image from a remote computing device.

4. The method of claim 1 , wherein obtaining the input image comprises:
capturing the input image using a camera.

5. The method of claim 4 , further comprising:
outputting, on a display operatively connected to the camera, a prompt to capture the input image;
outputting, on a preview display operatively connected to the camera, a graphical outline overlaid on a preview of the input image, wherein the graphical outline is an indication of an optimal area of the input image for the wear pin to fall within when the input image is captured; and
outputting, on a display operatively connected to the camera and based on the image dimension of the wear pin and the determined estimated measurement of the dimension of the wear pin, a graphical representation of the determined estimated measurement of the dimension of the wear pin, wherein the graphical representation of the determined estimated measurement of the dimension of the wear pin is displayed overlaid on top of an approximate location of the wear pin in the input image.

6. The method of claim 5 , further comprising:
outputting, on the display, a confirmation prompt requesting input indicating whether the graphical representation is a suitable representation of the determined estimated measurement of the dimension of the wear pin or an inadequate representation of the determined estimated measurement of the dimension.

7. The method of claim 1 , further comprising:
rotating the input image such that the wear pin is approximately horizontal in the rotated input Image.

8. The method of claim 7 , wherein rotating the input image comprises:
performing edge detection analysis on the input image to detect one or more edges of the brake assembly; and
rotating the input image such that the detected edges of the brake assembly are approximately axis-aligned.

9. The method of claim 1 , further comprising:
comparing one or more portions of the input image to one or more templates of the reference object;
determining a correlation to each template of the one or more templates for each portion of the input image; and
identifying the reference object in the input image by determining the portion of the one or more portions of the input image which has the highest correlation to a template of the one or more templates.

10. The method of claim 9 , wherein determining the estimated remaining life expectancy for the wear pin is further based on a vehicle type, and wherein the estimated remaining life expectancy comprises one of an estimated time until replacement for the wear pin or an estimated number of uses until replacement.

11. The method of claim 1 , wherein the input image is a first input image, the method further comprising:
obtaining a one or more subsequent input images, wherein the one or more subsequent input images identifies a vehicle associated with the brake assembly;
retrieving one or more stored dimensions of the wear pin; and
saving the estimated measurement of the dimension as a most recent stored dimension for the vehicle associated with the brake assembly.

12. The method of claim 11 , further comprising:
determining, based on the determined estimated measurement of the dimension of the wear pin and the one or more stored dimensions of the wear pin, an estimated remaining life expectancy for the wear pin based on a rate of change of the wear pin.

13. The method of claim 1 , wherein determining the image dimension of the reference object comprises:
detecting a first point on the reference object and a second point on the reference object in the input image, wherein the first point and the second point are on a perimeter of the reference object and are on opposite extremes of the reference object;
determining a number of pixels that separates the first point and the second point; and
determining a length per pixel ratio for the input image based on the number of pixels that separates the first point and the second point and the known dimension of the reference object.

14. The method of claim 1 , wherein determining the image dimension of the wear pin comprises:
detecting a location of the wear pin in the input image;
detecting a first point of the wear pin, wherein the first edge represents a fixed end of a bracket assembly;
identifying one or more line segments near the wear pin starting at the first edge;
identifying one or more clusters of the one or more line segments;
identifying a cluster of the one or more clusters that is coincident with the wear pin;
merging each line segment in the cluster coincident with the wear pin;
determining a number of pixels corresponding to the length of the merged line segment.

15. The method of claim 14 , wherein identifying the one or more line segments comprises:
applying an edge detection kernel to detect approximately horizontal edges;
determining pixels with responses above a threshold as horizontal edges; and
applying a Hough line transform to the each horizontal edge.

16. The method of claim 1 , further comprising:
converting the input image to a greyscale image.

17. The method of claim 1 , further comprising:
applying a bilateral filter to the input image.

18. The method of claim 1 , further comprising:
applying an equalization algorithm to the input image.

19. The method of claim 1 , further comprising:
determining a quality estimation parameter for the input image;
determining whether the quality estimation parameter is within an acceptable range;
in response to the quality estimation parameter being not within the acceptable range, sending an indication to a user; and
in response to the quality estimation parameter being within the acceptable range, determining the estimated measurement of the dimension of the wear pin based on the input image.

20. The method of claim 19 , wherein the quality estimation parameter is determined based on a blurriness of the input image, and wherein determining whether the quality estimation parameter is within the acceptable range comprises:
dividing the input image into a plurality of windows, wherein each window overlaps a portion of each window horizontally adjacent to the respective window;
determining a blurriness parameter for each window of the plurality of windows based on a magnitude of a Fast Fourier transform algorithm applied over a percentile of available spatial frequencies;
determining the quality estimation parameter for the input image as a weighted function of

Assignees

• Honeywell Int Inc

Inventors

• Venkatesha Sharath
• Lloyd Ryan Andrew
• Mylaraswamy Dinkar
• Behnke Mark E

Classifications

• B60T17/221 Primary

  Procedure or apparatus for checking or keeping in a correct functioning condition of brake systems (hydraulic pressure systems in general F15B19/00, F15B21/04; testing structures or apparatus G01M) · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title

• G06V10/42

  Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation · CPC title

• G01B11/00

  Measuring arrangements characterised by the use of optical techniques · CPC title

• G06K9/4604

  Physics · mapped topic