An Efficient Technique for Surface Strain Recovery from Photogrammetric Data using Meshless Interpolation

Abstract

a b s t r a c tIn the context of plane fracture problems, we introduce an algorithm based on our previously proposed rotation of edges but now including the injection of continuum softening elements directly in the process region. This is an extension of the classical smeared (or regularized) approach to fracture and can be seen as an intermediate proposition between purely cohesive formulations and the smeared modeling. Characteristic lengths in softening are explicitly included as width of injected elements. For materials with process regions with macroscopic width, the proposed method is less cumbersome than the cohesive zone model. This approach is combined with smoothing of the complementarity condition of the constitutive law and the consistent updated Lagrangian method recently proposed, which simplifies the internal variable transfer. Propagation-wise, we use edge rotation around crack front nodes in surface discretizations and each rotated edge is duplicated. Modified edge positions correspond to the crack path (predicted with the Ma-Sutton method). Regularized continuum softening elements are then introduced in the purposively widened gap. The proposed solution has algorithmic and generality benefits with respect to enrichment techniques such as XFEM. The propagation algorithm is simpler and the approach is independent of the underlying element used for discretization. To illustrate the advantages of our approach, yield functions providing particular cohesive behavior are used in testing. Traditional fracture benchmarks and newly proposed verification tests are solved. Results are found to be good in terms of load/deflection behavior.

Method for assessing beam column joints in RC structures using photogrammetric computer vision

“…Step 3 consists in detect the centre of the targets for all stages analyzed . The algorithm used is based on the Hough Transform for the detection of the centre of circular targets . A single image per stage is used, and the coordinates of the targets (in mm) are calculated (Figure ).…”

Assessing steel strains on reinforced concrete members from surface cracking patterns

“…Presently, there are several approaches available for tracking displacements using this technique [16][17][18][19][20]. In the present work, high contrast circular targets have been painted in the surface of the specimen and tracked for in-plane displacements during the test [16,21].…”

“…This matrix was used to scale and orientate all remaining images (see more details in [22]). The displacements measured during the test were then used to calculate the strain field [20,22,23] using standard finite element post-processing techniques. For this purpose, a strain-nodal displacement matrix was built for each target and used to calculate the strain [16].…”

Crack propagation monitoring using an image deformation approach

“…The displacement field is a function of the reference position and is known for a set of discrete points using photogrammetry or image correlation . A linear or nonlinear interpolation scheme is used to calculate the displacement values all over the surface . With the displacement field function defined, Equation can be used to compute a virtual image frame for any given time instant (assumed hereafter as “current”) that predicts how the structure should look like from the initial image of the surface (see Figure a,b).…”