Share Email Print

Optical Engineering

Superfast robust digital image correlation analysis with parallel computing
Author(s): Bing Pan; Long Tian
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

Existing digital image correlation (DIC) using the robust reliability-guided displacement tracking (RGDT) strategy for full-field displacement measurement is a path-dependent process that can only be executed sequentially. This path-dependent tracking strategy not only limits the potential of DIC for further improvement of its computational efficiency but also wastes the parallel computing power of modern computers with multicore processors. To maintain the robustness of the existing RGDT strategy and to overcome its deficiency, an improved RGDT strategy using a two-section tracking scheme is proposed. In the improved RGDT strategy, the calculated points with correlation coefficients higher than a preset threshold are all taken as reliably computed points and given the same priority to extend the correlation analysis to their neighbors. Thus, DIC calculation is first executed in parallel at multiple points by separate independent threads. Then for the few calculated points with correlation coefficients smaller than the threshold, DIC analysis using existing RGDT strategy is adopted. Benefiting from the improved RGDT strategy and the multithread computing, superfast DIC analysis can be accomplished without sacrificing its robustness and accuracy. Experimental results show that the presented parallel DIC method performed on a common eight-core laptop can achieve about a 7 times speedup.

Paper Details

Date Published: 10 March 2015
PDF: 8 pages
Opt. Eng. 54(3) 034106 doi: 10.1117/1.OE.54.3.034106
Published in: Optical Engineering Volume 54, Issue 3
Show Author Affiliations
Bing Pan, BeiHang Univ. (China)
Long Tian, BeiHang Univ. (China)

© SPIE. Terms of Use
Back to Top