Shape Measurement Using a New Multi-step Stereo-DIC Algorithm That Preserves Sharp Edges

Abstract : Digital Image Correlation is widely used for shape, motion and deformation measurements. Basically, the main steps of 3D-DIC for shape measurement applications are: off-line camera calibration, image matching and triangulation. The matching of each pixel of an image to a pixel in another image uses a so-called subset (correlation window). Subset size selection is a tricky issue and is a trade-off between a good spatial resolution, achieved with small subsets that preserve image details, and a low displacement uncertainty achieved with large subsets that can smooth image details. In this paper, we present a new multi-step DIC algorithm specially designed for measuring the 3D shape of objects with sharp edges. With this new algorithm an accurate 3D reconstruction of the whole object, including sharp edges that are preserved, can be achieved.
Document type :
Journal articles
Complete list of metadatas

Cited literature [2 references]  Display  Hide  Download

https://hal-mines-albi.archives-ouvertes.fr/hal-01169750
Contributor : Jean-José Orteu <>
Submitted on : Tuesday, June 30, 2015 - 11:31:55 AM
Last modification on : Tuesday, October 22, 2019 - 5:20:43 PM
Long-term archiving on : Tuesday, April 25, 2017 - 8:17:33 PM

File

EM_2015.pdf
Files produced by the author(s)

Identifiers

Citation

J. Harvent, Benjamin Coudrin, L. Brèthes, Jean-José Orteu, Michel Devy. Shape Measurement Using a New Multi-step Stereo-DIC Algorithm That Preserves Sharp Edges. Experimental Mechanics, Society for Experimental Mechanics, 2015, 55 (1), pp.167-176. ⟨10.1007/s11340-014-9905-z⟩. ⟨hal-01169750⟩

Share

Metrics

Record views

369

Files downloads

498