A Comparative Guide to Optical Strain Measurement Methods: Local DIC vs. Global DIC – Which Method Is Best Suited for Your Application?
Release time:
2026-03-12 10:16
Source:
1. Introduction
Digital Image Correlation (DIC) technology employs two primary algorithmic approaches: Localized DIC (Based on subsets) and Global DIC (Based on the finite element method.) Both can generate displacement and strain fields, but their underlying mechanisms are entirely different.
Key points: In 90% of cases, the results from the two methods are consistent, with only differences attributable to measurement error. The DIC Challenge organized by the iDICs Society has also demonstrated “significant similarity between the two methods.”
2. Local DIC: Subset Principle
Operating Principle
Local DIC divides the image into small square regions called “subsets” (typically 15–50 pixels in width). The algorithm searches for matching locations in the deformed image using a correlation criterion, usually the sum of squared differences with zero normalization (ZNSSD). The displacement of each subset is computed independently, and strain is obtained through numerical differentiation of the displacement field.
Advantages
Robustness
Independent processing means that poor image quality in one region will not affect adjacent regions.
Ease of Use
No need to prepare a mesh or numerical model in advance to use it.
Versatility
Applicable to almost all visible geometric shapes.
Limitations
Point cloud output
The results are expressed as a regular grid of image point clouds, rather than as finite element analysis (FEA) mesh nodes.
Projection Requirements
DIC data must be projected onto the FEA mesh, requiring reference frame transformation, 3D interpolation, and the incorporation of error sources.
No guarantee of continuity.
Independent processing may introduce artificial discontinuities at the edges.
Parameter Sensitivity
The size of the subset and the grid step significantly affect the resolution and noise level.
Representative Software
3. Global DIC: Based on the Finite Element Method
Operating Principle
Global DIC uses a finite element mesh as the measurement carrier. It does not rely on isolated subsets; instead, it parameterizes the displacement field using finite element shape functions—these shape functions are identical to those used in the simulation solver. The correlation problem is solved as a global optimization problem, minimizing the correlation residuals across the entire surface while simultaneously enforcing continuity of the displacement field at the mesh nodes.
Advantages
Direct FEA mesh output
The measured displacement is directly represented at the simulation mesh nodes, enabling real‑time test–simulation comparison.
Ensure continuity.
The displacement field is structurally continuous, with no artificial discontinuities.
Edge measurement
The regularity of shape functions can typically extend the measurement capability to near the edges of the specimen.
The ideal verification tool
Data and FEA models exist in the same space, making them an ideal choice for simulation verification.
Limitations
FEA mesh required
It cannot be run without a numerical model (but EikoTwin DIC can generate a mesh for simple cases).
Not robust enough against discontinuities.
Forced continuity may introduce measurement bias near the crack tip (but it can provide information on crack location).
Higher computational costs
Solving a global problem may consume more computational resources than solving N independent problems (though modern implementations are now extremely fast).
Representative Software
4. Comparative Summary
| Comparison Item | Localized DIC (Subset) | Global DIC (Finite Element-Based) |
|---|---|---|
| Principle | Conduct correlation analysis through an independent window. | Perform correlation analysis on the global finite element mesh. |
| Data format | Point Cloud (Image Grid) | FEA mesh nodes |
| Continuity | Not guaranteed | Ensured through shape functions |
| FEA Connection | Post‑processing projection (sources of error) | Direct connection, no interpolation required. |
| Do you need a grid? | No | Yes (can be generated in EikoTwin DIC) |
| Discontinuity | Good robustness | Not very suitable for strong discontinuities. |
| Ease of Use | High | High (using EikoTwin) |
| Calculation time | Quick | Quite capable on modern hardware. |
| Main uses | Materials Research and Development | FEA Simulation Verification and Model Update |
5. How do you choose a method for your application?
Direct Strain Measurement
When there is no grid and direct comparison with the simulation is not required, both methods are applicable.
Finite Element Model Validation
Projecting local DIC data onto the FEA mesh introduces systematic errors and requires extensive manual work. Global DIC integrates this step, directly providing data that can be compared with simulations.
Tests involving discontinuities (cracks, slip)
Either use global DIC with an adaptive mesh, or employ enhanced techniques such as X-FEM or element deletion. If left unaddressed, the enforced continuity of global DIC may introduce measurement biases near the crack tip.
Model Update for Parameter Identification
The parameter identification loop requires that experimental data and the FEA model reside in the same space—this is precisely what global DIC natively provides.
6. The Best DIC Software for FEA Validation
The most widely used local DIC software.
GOM Correlate / ATOS
Localized DIC
A widely used software in the industry, featuring excellent human–machine ergonomics and tight integration with the Zeiss ecosystem. It is primarily geared toward metrology and quality control rather than closed-loop FEA validation. Exporting to FEA solvers requires manual projection.
VIC-3D
Localized DIC
The academic and industrial reference standard for 3D local DIC is widely used in research laboratories. It shares the same limitations as GOM—data are represented in the form of point clouds rather than directly on the FEA mesh.
MatchID
Localized DIC
The European solution is rapidly growing and renowned for its flexibility and open APIs. It features seamless integration with research workflows. Software capabilities are being developed to overcome the constraints associated with FEA comparisons.
Istra4D
Localized DIC
Typically bundled with Dantec camera systems, it employs the classic local method.
The only industrial-grade, finite-element–based global DIC software.
EikoTwin DIC
Global Finite Element DIC
The only commercial software that implements a finite-element–based, global DIC system tailored for industrial applications. Measurement data is directly mapped onto the simulation mesh nodes (Abaqus, Ansys, HyperWorks, Nastran) without the need for projection, enabling real‑time, interpolation‑free comparison between experimental and simulation results.
Software Comparison Summary
| Software | Method | FEA Connection | Main uses |
|---|---|---|---|
| GOM Correlate (Zeiss) | Local |
Post‑processing | Measurement and Quality Control |
| VIC-3D (Correlated Solutions) | Local |
Post‑processing | Scientific Research, Materials Characterization |
| MatchID | Local |
Post‑processing | Industrial research and academic testing |
| Istra4D (Dantec) | Local |
✗ | Industrial Metrology |
Global Finite Element |
✓ Native support |
FEA Simulation Verification and Model Update |
7. Conclusion
The key to choosing which method to use lies not in algorithm performance, but in the intended use of the data. For testing laboratories that need to input measurement results into simulation models, Global DIC based on the finite element method is the optimal solution. It generates data that is ready for immediate use, requiring no post‑processing, no interpolation bias, and no spatial transformation.
References
- International Digital Image Correlation Society, Jones, E.M.C. and Iadicola, M.A. (Eds.) (2025). A Good Practices Guide for Digital Image Correlation, 2nd Edition . https://doi.org/10.32720/idics/gpg.ed2
- Hild, F. & Roux, S. (2012). Comparison of Local and Global Approaches to Digital Image Correlation . Experimental Mechanics, 52(9), 1503–1519.
- Leclerc, H., Périé, J.-N., Roux, S. & Hild, F. (2009). Integrated digital image correlation for the identification of mechanical properties . MICCAI.
- Pan, B. et al. (2009). Two-Dimensional Digital Image Correlation for In-Plane Displacement and Strain Measurement: A Review . Measurement Science and Technology, 20(6).
- Lava, P., Jones, E., Wittevrongel, L., Pierron, F. (2020). Validation of Finite-Element Models Using Full-Field Experimental Data: Aligning Finite-Element Analysis Data via a Digital Image Correlation Engine . Strain, 56(4).
Localized DIC,Global DIC,Optical Strain Measurement,Finite Element Simulation Comparison and Validation,Strain Measurement
Previous Page