Representation of a component using cross-sectional images

Patent metadata

Field	Value
Publication number	US-10957075-B2
Application number	US-201816191322-A
Country	US
Kind code	B2
Filing date	Nov 14, 2018
Priority date	Dec 15, 2017
Publication date	Mar 23, 2021
Grant date	Mar 23, 2021

Abstract

A method includes receiving a cross-sectional image of a component, including a plurality of pixels representing the component, at a perspective, determining a threshold color value based on color values associated with the plurality of pixels, and setting an updated color value for each pixel of the plurality of pixels based on the threshold color value. The method includes analyzing a set of adjacent pixels of the cross-sectional image that is selected based on a geometric parameter based on an expected geometry of a physical domain of the component, identifying a plurality of adjacent pixels from the set of adjacent pixels that is likely to be associated with the physical domain of the component based on the updated color values associated with the plurality of adjacent pixels, and outputting a representation of the component including the plurality of adjacent pixels that are likely to be associated with the physical domain.

First claim

What is claimed is:

1. A method, comprising:
receiving, by a controller, a cross-sectional image of a component at a perspective, wherein the cross-sectional image comprises a plurality of pixels representing the component;
determining, by the controller, a threshold color value;
setting, by the controller, an updated color value for each pixel of the plurality of pixels based on the threshold color value;
analyzing, by the controller, a set of adjacent pixels of the cross-sectional image, wherein the set of adjacent pixels is selected based on a geometric parameter, and wherein the geometric parameter is based on an expected geometry of a physical domain of the component;
identifying, by the controller, a plurality of adjacent pixels from the set of adjacent pixels that is likely to be associated with the physical domain of the component based on the updated color values associated with the plurality of adjacent pixels, wherein the component comprises at least two physical domains; and
outputting, by the controller, a representation of the component including the plurality of adjacent pixels that are likely to be associated with the physical domain.

2. The method of claim 1 , wherein the geometric parameter defines a size or shape of a pixel region that includes the set of adjacent pixels and corresponds to a size or shape of an instance of the physical domain, and wherein identifying the plurality of adjacent pixels from the set of adjacent pixels that is likely to be associated with the physical domain of the component based on the updated color values associated with the plurality of adjacent pixels further comprises identifying, by the controller, the plurality of adjacent pixels that include an updated color value associated with the physical domain and fit into the pixel region.

3. The method of claim 2 , wherein the component comprises a woven material, the physical domain comprises at least one tow, and the geometric parameter is an aspect ratio associated with an average cross-section of the at least one tow.

4. The method of claim 1 , further comprising determining, by the controller, whether the plurality of adjacent pixels is likely to be associated with the component based on a geometric parameter associated with a geometry of the physical domain of the component at the perspective.

5. The method of claim 4 , wherein determining whether the plurality of adjacent pixels is likely to be associated with the physical domain based on the geometric parameter further comprises:
identifying, by the controller, a reference point of the plurality of adjacent pixels; and
determining, by the controller, whether the reference point of the plurality of adjacent pixels is within a value of the geometric parameter.

6. The method of claim 5 , wherein the reference point of the plurality of pixels is a center of the plurality of pixels and wherein identifying the reference point of the plurality of adjacent pixels comprises identifying a center of an area defined by the plurality of adjacent pixels.

7. The method of claim 5 , wherein the cross-sectional image is from a series of sequential cross-sectional images captured at various depth increments along the perspective, wherein the geometric parameter represents an anticipated change in a position of the component in the cross-sectional image between sequential cross-sectional images of the series of sequential cross-sectional images, and wherein determining whether the reference point of the plurality of adjacent pixels is within a value of the orientation parameter further comprises determining, by the controller, whether the reference point of the plurality of adjacent pixels is within the value of the geometric parameter from an adjacent reference point of an adjacent sequential cross-sectional image of the series of sequential cross-sectional images.

8. The method of claim 7 , further comprising storing, by the controller and in response to determining that the reference point is within the value of the geometric parameter, the reference point of the plurality of pixels.

9. The method of claim 7 , further comprising:
determining, by the controller, that the cross-sectional image does not include a reference point associated with the physical domain of the component; and
determining, by the controller and in response to determining that the cross-sectional image does not include the reference point associated with the physical domain of the component, a projected reference point for the plurality of adjacent pixels based on at least the adjacent reference point of the adjacent sequential cross-sectional image.

10. The method of claim 9 , wherein the projected reference point is determined by interpolation or extrapolation.

11. The method of claim 1 , further comprising determining, by the controller and in response to identifying the plurality of adjacent pixels as likely to be associated with the physical domain, a color of the plurality of adjacent pixels.

12. The method of claim 11 , wherein the color is determined based on a distribution parameter associated with a distribution of the component in the material.

13. The method of claim 12 , wherein the distribution parameter is a percentage of pluralities of adjacent pixels that are likely to be associated with the component that are identified in the cross-sectional image, and wherein determining the color further comprises:
determining, by the controller, a range of values of the geometric parameter associated with the distribution parameter; and
determining, by the controller, that the dimension of the plurality of adjacent pixels is within the range of values of the geometric parameter associated with the distribution parameter.

14. A system, comprising:
a controller configured to:
receive a cross-sectional image of a component at a perspective, wherein the cross-sectional image comprises a plurality of pixels representing the component;
determine a threshold color value;
set an updated color value for each pixel of the plurality of pixels based on the threshold color value;
analyze a set of adjacent pixels of the cross-sectional image, wherein the set of adjacent pixels is selected based on a geometric parameter, and wherein the geometric parameter is based on an expected geometry of a physical domain of the component;
identify a plurality of adjacent pixels from the set of adjacent pixels that is likely to be associated with the physical domain of the component based on the updated color values associated with the plurality of adjacent pixels, wherein the component comprises at least two physical domains; and
output a representation of the component including the plurality of adjacent pixels that are likely to be associated with the physical domain.

15. The system of claim 14 , wherein the geometric parameter defines a size or shape of a pixel region that includes the set of adjacent pixels and corresponds to a size or shape of an instance of the physical domain, and wherein the controller is further configured to identify the plurality of adjacent pixels that include an updated color value associated with the physical domain and fit into the pixel region.

16. The system of claim 14 , wherein the controller is further configured to determine whether the plurality of adjacent pixels is likely to be associated with the component based on a geometric parameter associated with a geometry of the physical domain of the component at the perspective.

17. The system of claim 14 , wherein the controller is further configured to:
identify a reference point of the

Assignees

• Rolls Royce Corp

Inventors

• Ritchey Andrew James
• Mason Tyler

Classifications

• G06T7/0006 Primary

  using a design-rule based approach · CPC title

• G06T2207/30164

  Workpiece; Machine component · CPC title

• G06T2207/30124

  Fabrics; Textile; Paper · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G06T15/00

  Three-dimensional [3D] image rendering · CPC title