Method for defect indication detection

Patent metadata

Field	Value
Publication number	US-10203290-B2
Application number	US-201415100567-A
Country	US
Kind code	B2
Filing date	Dec 1, 2014
Priority date	Dec 12, 2013
Publication date	Feb 12, 2019
Grant date	Feb 12, 2019

Abstract

Methods, apparatus and computer-readable media for detecting potential defects in a part are disclosed. A potential defect may be automatically detected in a part, and may be reported to an operator in various ways so that the operator may review the defect and take appropriate action.

First claim

The invention claimed is:

1. A volumetric computed tomography (VCT)-based method for providing notification of a potential defect in a composite part, comprising:
obtaining ( 102 ) VCT scan data representing a three-dimensional (3D) volume of the composite part;
normalizing ( 220 ) the 3D volume against a generated background of the composite part that includes beam hardening and/or scattering artifacts;
detecting ( 222 ) an indication of the potential defect in the normalized 3D volume; and
outputting ( 112 ) information about the indication.

2. The VCT-based method of claim 1 , wherein the outputting comprises rendering, on a display ( 806 ), a 3D image of the composite part that includes the indication.

3. The VCT-based method of claim 1 , further comprising determining ( 224 ) a boundary of the detected indication based at least in part on a computer-aided design (CAD) model, wherein the information about the indication includes information about the boundary.

4. The VCT-based method of claim 3 , further comprising:
identifying ( 222 ) a seed voxel of the indication; and
performing ( 620 , 720 ) region growing around the seed voxel.

5. The VCT-based method of claim 4 , wherein the CAD model is a ply model ( 400 ) of the composite part, and growth of the region around the seed voxel is constrained by the ply model.

6. The VCT-based method of claim 1 , further comprising denoising ( 610 ) the normalized 3D volume using an adaptive bilateral filter.

7. The VCT-based method of claim 1 , further comprising denoising ( 606 ) the normalized 3D volume using a non-local means filter.

8. The VCT-based method of claim 1 , further comprising:
identifying ( 224 ) a signal area of the indication; and
identifying ( 220 ) a background of the indication based on a computer-aided design (CAD) model.

9. The VCT-based method of claim 8 , further comprising calculating ( 226 ) a signal strength of the indication based on the signal area and the background of the indication.

10. The VCT-based method of claim 9 , wherein the normalizing includes generating ( 602 ) a part mask of the composite part.

11. The VCT-based method of claim 10 , further comprising subtracting ( 220 ) the generated background of the composite part from the 3D volume of the composite part.

12. The VCT-based method of claim 1 , wherein the detecting comprises calculating ( 222 ) global and local porosities of the 3D volume.

Assignees

• Gen Electric

Inventors

• Ferro Andrew Frank
• Yang Xingwei
• Mendonca Paulo Ricardo Dos Santos
• Nafis Christopher Allen
• Howard Patrick Joseph

Classifications

• G01N21/00

  Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light (G01N3/00 - G01N19/00 take precedence) · CPC title

• G06T7/001

  using an image reference approach · CPC title

• G06T2207/30108

  Industrial image inspection · CPC title

• G06T7/00

  Image analysis · CPC title

• G01N23/046

  using tomography, e.g. computed tomography [CT] · CPC title