The second edition of iDICs’ “Good Practices Guide for DIC” is now available: Stop relying on guesswork at the test site for uncertainty assessment in measurement protocols—start by running virtual DIC simulations first.
Release time:
2026-10-28 13:24
Source:
The second edition of iDICs’ “Good Practices Guide for Digital Image Correlation” was released in October 2025, introducing, for the first time across all measurement fields, a systematic framework for assessing reliability scores. For experimental engineers, this means one thing: whether a measurement plan is sound is shifting from “leaving it to chance on site” to “verifying it before running the experiment.” The tool for this verification is virtual DIC pre‑simulation.
01 Current Status of Uncertainty Assessment in Measurement Schemes
Status Quo
The current situation is that most laboratories still rely on experience for scheme reviews. Taking residual stress measurement as an example, the incremental hole‑drilling method continues to use strain rosettes as the reference configuration (per the ASTM E837 framework). Digital image correlation (DIC) has been added to capture non‑uniform stress gradients; however, the validity of the DIC‑based approach still depends on validation against the strain rosette configuration. [1] 。
Algorithmic progress has been rapid: using deep learning, DIC achieves an average endpoint error of 0.003 pixels and a throughput of 448 frames per second on 36,000 reference-deformed image pairs, far surpassing traditional methods—yet real-world challenges such as texture scarcity, uneven illumination, scratches, and hotspots remain a common bottleneck for all algorithms. [2] No matter how powerful the algorithm is, it can’t compensate for a fundamentally flawed solution.
02 Standardization Requirements Introduced by the Second Edition of iDICs
What the 2nd Edition Asks
The second edition of the guideline is very clear in its orientation: uncertainty must be quantified, and sources of error must be attributed item by item. Research has already broken down the accounting in great detail—in complex deformation measurements, undermatched shape functions are the primary source of systematic error in DIC systems; zeroth-, first-, and second-order shape functions each have their own applicable limits, and selecting the wrong option introduces direct systematic bias into the strain results. [3] 。

Figure 1: Speckle pattern and region of interest (ROI): The choice of subregion size, shape, and function order directly determines the error level (adapted from reference [3]).
The impact of subregion size can also be quantified: in the same measurement, increasing the subregion size from 10 pixels to 70 pixels reduces the standard deviation error from 0.010 to below 0.001 pixels. The appropriate parameter settings should be calculated during the design phase rather than determined through on-site experimentation. [3] 。

Figure 2: Curve showing the relationship between subdomain size and standard deviation error: Under first- and second-order shape functions, the error converges as the subdomain size increases (reproduced from reference [3]).
When the object being measured is still in motion, the error is further amplified: studies on the full-field uncertainty of fluorescence thermometry show that conventional methods introduce significant biases during material motion, whereas geometric compensation techniques can reduce such biases by a factor of three—uncertainty assessments must account for real-world operating conditions, rather than idealized static states. [4] 。

Figure 3: Dual-camera measurement setup for fluorescence thermometry combined with DIC: Filters and camera geometry are both included in the uncertainty model (reproduced from reference [4]).
03 Practices and Benefits of Virtual DIC Rehearsals
Rehearse, Then Test
The virtual rehearsal process involves three steps: importing the simulation model into a virtual testing environment and virtually placing measurement points according to real camera parameters; generating virtual images with realistic rendering and processing them using the same DIC algorithm; and calculating measurement uncertainty by comparing with the simulation truth—revising the point layout, speckle pattern, or frame rate if the results fail to meet the criteria. The practical value for large components is thus demonstrated: in boundary load tests on porcelain insulators, the minimum relative error between DIC monitoring and finite element analysis was only 0.7%, and high-stress regions were reliably identified—provided that the experimental setup has first been validated. [5] 。

Figure 4: Component design for the 4.8-m large beam test: Measurement schemes for such large-scale structural tests, in particular, require prior validation (reproduced from reference [6]).
At the macro scale, comparative tests between DIC and distributed fiber optics yielded complementary datasets: for a 4.8‑m CFRP‑prestressed concrete beam under three‑point bending, DIC captured the full-field crack evolution while the fiber optic sensor provided strain distribution along the length; the two data streams cross‑validated each other. Crucially, the integration of multiple measurement techniques still hinges on first quantifying the uncertainty inherent in each individual method. [6] 。

Figure 5: DIC grid coverage of a large beam test, showing the actual photograph: full-field data directly corresponds to the in-situ conditions (reproduced from reference [6]).
04 Recommended Implementation Plans
Implementation Recipe
EikoTwin Virtual has formalized this rehearsal process into a standard workflow (as documented in the official product materials): ModelConverter transforms Abaqus/HyperWorks simulation models into virtual test scenarios; BlenDIC sets up virtual cameras within Blender—confirming fields of view, depth of field, and framing parameters. It then generates photorealistic virtual images, estimates measurement uncertainty, quantifies speckle sizes, and even outputs scaffolding and on-site implementation drawings. Measurement feasibility is thus established in writing well before testing begins.
ArianeGroup large-scale structure case study (official data): For a major test involving eight cameras, virtual sensor placement was first used to achieve millimeter-level accuracy. The estimated strain uncertainty was about one-fifth of the expected maximum strain, which guided the determination of scaffolding dimensions and speckle pattern design, saving four days of on-site preparation per test—the benefits of virtual pre‑simulation can be directly converted into equivalent rig‑days.
Advice for the trial leader: Translate the requirements of iDICs Version 2 into an internal audit checklist—uncertainty assessment, attribution of error sources, and rationale for parameter selection—and populate each item with evidence from virtual simulations. Demonstrating the approach first, then implementing it on the machine, will be a fundamental skill in the era of the new guidelines.
The guidelines have already laid out the rules; the dry run is your answer sheet.
References
[1] Measurement of Non-uniform Residual Stresses Using the Incremental Hole Drilling Method and Digital Image Correlation: A Feasibility Study on Deep Rolled Specimens. Journal of Failure Analysis and Prevention, 2026. DOI:10.1007/s11668-026-02532-4
[2] Multi-resolution speckle priors for scale-aware and simultaneous probabilistic digital image correlation. Neural Computing and Applications, 2026. DOI:10.1007/s00521-026-12631-w
[3] Advanced Algorithms of Mitigating Undermatched Systematic Error in DIC. Experimental Mechanics, 2025. DOI:10.1007/s11340-025-01238-2
[4] Full-Field Measurement Uncertainty Quantification for Luminescence Phosphor Thermography on Surfaces in 3D Space. Experimental Mechanics, 2026. DOI:10.1007/s11340-026-01364-5
[5] Analysis of the boundary load impact on porcelain insulators and DIC test. Scientific Reports, 2026. DOI:10.1038/s41598-026-59525-4
[6] Comparison of Distributed Fiber Optic Sensing and Digital Image Correlation Measurement Techniques for Evaluation of Flexural Behavior of CFRP-Prestressed Concrete Beams. Sensors, 2025. DOI:10.3390/s25237357
Full-field Strain Measurement,DIC Test Plan Verification,Large-Scale Structural Testing,Virtual DIC Simulation,Measurement uncertainty of DIC,IDICs Guidelines, Second Edition