Method and system for direct strain imaging

• US 9,311,566 B2
• Filed: 08/02/2013
• Issued: 04/12/2016
• Est. Priority Date: 08/03/2012
• Status: Active Grant

• Current Assignee: George Mason Research Foundation Inc., The Government Of The United States Of America, As Represented By The Secretary Of The Navy
• Original Assignee: George Mason Research Foundation Inc., the united states of america as represented by the secretary of the navy
• Inventors: Michopoulos, John G., Iliopoulos, Athanasios
• Primary Examiner(s): Koziol, Stephen R
• Assistant Examiner(s): Gilliard, Delomia L
• Application Number:
  US13/958,247
• Publication Number:
  US 20140037217A1
• Time in Patent Office: 984 Days
• Field of Search: 382/201
• US Class Current: 1/1
• CPC Class Codes: G01B 11/16 for measuring the deformation in a solid, e.g. optical strain gauge
• CPC Class Codes: G01N 2203/0212 Theories, calculations
• CPC Class Codes: G01N 2203/0647 Image analysis
• CPC Class Codes: G01N 3/068 with optical indicating or recording means
• CPC Class Codes: G06F 18/213 Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
• CPC Class Codes: G06T 2207/10021 Stereoscopic video; Stereoscopic image sequence
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30108 Industrial image inspection
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06T 7/285 using a sequence of stereo image pairs

Abstract

A method and system for measuring and determining the full-field spatial distributions of strain tensor field components in a two or three-dimensional space, as a consequence of deformation under generalized loading conditions. One or more digital cameras may be used to acquire successive images of a deforming body with optically distinctive features on its surface. A method for determining the location of characteristic points of the surface features and another one for tracking these points as deformation occurs. Elongations between neighboring points on the vicinity of a location of interest are computed. The elongation between points is calculated even though discontinuities may exist between them. Strain tensor fields are directly calculated as a tensor approximation from these elongations without determining or using the displacement vector distributions.

Claims

1. A computer-implemented method for determining strain characteristics in a vicinity of a point of interest w={x_w, y_w}^T in two parametric dimensions embedded in a two or three dimensional space, wherein the point of interest w is part of a deformable domain or body having a plurality of features which may be represented as a node pattern of n nodes, the method comprising:

receiving a first set of captured data of a plurality of features and representing the plurality of features by nodes indexed by i or j in the domain in an un-deformed configuration;

receiving a second set of captured data of the plurality of features represented by nodes i,j in the domain in a deformed configuration;

measuring engineering strains e_i,j between a plurality of node combinations in the vicinity of w to produce e_w={e₂₁, e₃₁, . . . e₄₂ . . . , e_n(n-1)}^T; and

applying an approximation or interpolation to the engineering strains e_i,j to approximate strain ε

^h in the vicinity of w.

2. The method of claim 1, wherein the approximation defines an overdetermined system of equations.

3. The method of claim 1, wherein the moving least squares approximation expresses strain ε

^h in two parametric dimensions embedded in a two or three dimensional space by

4. The method of claim 3, wherein the first and second sets of captured data each being a digital images, and further comprising identifying real mean strain e_i,j between a number of node combinations by

5. The method of claim 1, wherein the first and second sets of captured data comprise a first and second sequence of images from at least two digital imaging devices positioned facing the node pattern.

6. The method of claim 1, wherein the first and second sets of captured data are digital images and the plurality of features are visible.

7. The method of claim 1, wherein each of said n nodes comprises a centroid of a visible dot or feature.

8. The method of claim 1, wherein the first and second sets of captured data are the first and second digital images respectively, the method further comprising identifying the node pattern in the first and second digital images via one or more image processing steps.

9. The method of claim 8, wherein said one or more image processing steps comprise eliminating irregular objects from said first and second digital images based on the objects having an intensity below a threshold value of intensity, a pixel area outside a predetermined range, an out-of-range aspect ratio, an out-of-range moment of inertia, an out-of-range major axes direction, or an out-of-range compactness ratio.

10. The method of claim 1, wherein the first and second sets of captured data are the first and second images respectively, wherein said receiving said first and second images comprises digitally photographing a deformable body during deformation with two or more cameras for generating image data and communicating the image data to a computer processor having a memory.

21. The method of claim 1, wherein the applying comprises applying a spatial moving least squares approximation to the engineering strains.

23. The method of claim 1, wherein the first set of captured data is a first image in the second set of captured data is a second image.

11. A system for determining strain characteristics in a vicinity of a point of interest w={x_w, y_w}^T in two parametric dimensions embedded in a two or three dimensional space for a deformable domain having a plurality of features which may be represented as a node pattern of n nodes, the system comprising:

at least two imaging devices arranged facing the plurality of features at different angles to the surface of the deformable domain for acquiring a first image of the plurality of features in the domain in an un-deformed configuration and a second image of the plurality of features in the domain in a deformed configuration; and

a computer processor having programmed instructions thereon for

(a) representing the plurality of features by nodes i,j;

(b) measuring engineering strains e_i,j between a plurality of node combinations in the vicinity of w to produce e_w={e₂₁, e₃₁, . . . e₄₂ . . . , e_n(n-1)}^T; and

(c) applying an approximation or interpolation to the engineering strains e_i,j to approximate strain ε

^h in the vicinity of w.

12. The system of claim 11, wherein the approximation defines an overdetermined system.

13. The system of claim 11, wherein the approximation expresses strain by

14. The system as in claim 11, the processor having further instructions for corresponding nodes in an un-deformed configuration with nodes in a deformed configuration to form the plurality of node combinations, such that where x_i, x_j are vectors representing a position of nodes i,j at the un-deformed configuration and x′

_i, x′

_j are the same vectors in the deformed configuration.

15. The system as in claim 14, the processor having further instructions for identifying real mean strain e_i,j between a number of node combinations by:

16. The system as in claim 11, further comprising:

a display connected to the processor, and the processor having further instructions for displaying and storing the first and second images and a graphical representation of strain in the domain.

17. The system as in claim 11, wherein the processor comprises instructions for determining the node pattern stereoscopically.

22. The system of claim 11, wherein the applying comprises applying a spatial moving least squares approximation to the engineering strains.

18. A computer-implemented method for measuring strain characteristics in a vicinity of a point of interest w={x_w, y_w}^T in two parametric dimensions embedded in a two or three dimensional space for a deformable domain having a plurality of features which may be represented as a node pattern of n nodes, the method comprising:

receiving separation-data for a plurality of nodes i,j in the domain in an un-deformed configuration;

receiving separation data for the plurality of nodes i,j in the domain in a deformed configuration;

measuring engineering strains e_i,j between a number of node combinations in the vicinity of w to produce e_w={e₂₁, e₃₁, . . . e₄₂ . . . , e_n(n-1)}^T; and

approximating strain ε

^h in the vicinity of w by

19. The method of claim 18, further comprising identifying real mean strain e_i,j between a number of node combinations by:

20. The method of claim 18, wherein the separation data is derived from neighboring markers using one of ultrasonography or X-ray tomography imaging or digital imaging of any kind.