ezDIC

A lightweight virtual extensometer for extracting linear strain from image sequences.

ezDIC is designed for researchers who need fast, practical strain extraction without running a full-field Digital image correlation workflow. It tracks two user-defined ROI markers across an image sequence and exports engineering strain, true strain, quality-control information, and Origin-compatible TXT files for plotting and reporting.

Developed by Dr. Delun Gong.

DOI: 10.5281/zenodo.20222465

Why ezDIC?

Many materials experiments only require a reliable 1D strain history rather than full-field displacement maps. ezDIC focuses on that narrower problem:

• Simple workflow: load images, draw two ROIs, run tracking, export strain.
• Virtual extensometer model: strain is computed from the changing distance between two tracked ROI centers.
• Origin-compatible TXT output: the default export contains Frame, EngineeringStrain, and TrueStrain.
• Origin OPJU export: optionally writes core result tables directly into an OriginPro project file.
• Mean-strain export: multiple extensometers with the same role and direction are averaged frame by frame with standard deviation, SEM, and valid-count columns.
• Poisson-ratio export: mark axial and transverse ROI groups to export transverse strain and PoissonRatio; multiple groups are averaged before the ratio is computed.
• Quality control built in: rejected frames, adaptive accepts, correlation scores, and QC summaries are reported.
• No Python required for users: Windows releases are distributed as a green, portable folder with ezDIC.exe.
• Research-oriented defaults: failed tracking frames remain NaN instead of being silently interpolated.

Typical Use Cases

• Tensile, compression, bending, or thermal-deformation image sequences where the primary target is a linear strain curve.
• Quick validation of extensometer or DIC measurements.
• Teaching image-based strain extraction without requiring a full commercial DIC package.
• Lightweight exploratory analysis before committing to full-field DIC.

Outputs

By default, ezDIC writes a compact core/ result folder:

core/
  strain_G01.txt
  strain_all_groups.txt
  strain_mean_groups.txt
  ezDIC_results.opju       # optional, requires OriginPro 2021+ and originpro
  poisson_ratio.txt        # when axial/transverse ROI roles are set
  engineering_strain_G01.png
  engineering_strain_all_groups.png
  poisson_ratio.png        # when axial/transverse ROI roles are set
qc/
  qc_summary.txt

The primary TXT format is intentionally simple:

Frame	EngineeringStrain	TrueStrain
1	0.00000000	0.00000000
2	-0.00000254	-0.00000254
3	0.00000580	0.00000580

Optional exports include an Origin OPJU project, full CSV tables, correlation plots, tracking overlays, and parameter summaries. The OPJU export requires Windows, OriginPro 2021+, a valid local OriginPro license, and the originpro Python package. It writes worksheet data only; publication figures still come from the existing PNG exports or from manual plotting in OriginPro.

For repeated virtual extensometers, strain_mean_groups.txt averages groups with the same role and actual_mode frame by frame. Rejected frames and NaN strain values are excluded from the mean. The mean table includes:

MeanEngineeringStrain_<role>_<mode>	MeanTrueStrain_<role>_<mode>	StdEngineeringStrain_<role>_<mode>	SemEngineeringStrain_<role>_<mode>	ValidGroupCount_<role>_<mode>

For Poisson-ratio analysis, add at least one ROI group with role axial and at least one with role transverse. strain_all_groups.txt keeps the original per-group columns, appends mean-strain columns, and appends:

AxialEngineeringStrain	TransverseEngineeringStrain	PoissonRatio

Windows Quick Start

1. Download ezDIC_Windows_x64_v0.1.3.zip from the release package.
2. Extract the full ezDIC_Windows_x64 folder.
3. Double-click ezDIC.exe.
4. Do not copy ezDIC.exe alone; keep _internal/ in the same folder.

Target platform: Windows 10/11 x64.

Running From Source

py -m venv .venv
.\.venv\Scripts\python.exe -m pip install -r requirements.txt
.\.venv\Scripts\python.exe dic_virtual_extensometer_gui_v7_multi_roi_range.py

Building The Windows Release

powershell -NoProfile -ExecutionPolicy Bypass -File .\build_release.ps1

The script creates:

release/
  ezDIC_Windows_x64/
  ezDIC_Windows_x64_v0.1.3.zip

Validation

The current automated checks cover:

• Origin-compatible TXT export.
• Origin OPJU table construction and failure handling with a fake OriginPro API.
• true strain recomputation from engineering strain.
• QC summary generation.
• GUI title and developer attribution.
• release metadata and PyInstaller packaging files.

Run:

py -m pytest -q

How to Cite

This repository includes CITATION.cff, so GitHub will show a Cite this repository link on the project page. For papers, theses, reports, and presentations, please cite the Zenodo DOI:

Gong, D. (2026). ezDIC: A lightweight virtual extensometer for extracting linear strain from image sequences (Version 0.1.3) [Computer software]. Zenodo. https://doi.org/10.5281/zenodo.20222465

DOI: 10.5281/zenodo.20222465

GitHub repository: https://github.com/D-sudoasd/ezDIC

Scientific Notes

ezDIC computes engineering strain using:

engineering strain = (L - L0) / L0

and true strain using:

true strain = ln(L / L0) = ln(1 + engineering strain)

where L0 is the initial ROI-center separation and L is the current separation. If tracking fails, the frame is exported as NaN to preserve the experimental record.

Poisson ratio is computed from engineering strain. If more than one axial or transverse ROI group is defined, ezDIC first computes the frame-by-frame mean strain within each role:

PoissonRatio = - TransverseEngineeringStrain / AxialEngineeringStrain

If either role has no valid strain for a frame, either mean strain is NaN, or abs(AxialEngineeringStrain) < 1e-6, the Poisson-ratio value is exported as NaN.

Limitations

ezDIC is not a replacement for full-field DIC. It does not compute strain maps, displacement fields, or local strain heterogeneity. It is intended for fast extraction of a representative linear strain curve from image sequences where a virtual extensometer is scientifically appropriate.

Attribution And Usage

This software was developed by Dr. Delun Gong for lightweight extraction of linear strain from image sequences.

DOI: 10.5281/zenodo.20222465

Users are not permitted to:

1. claim that they developed this software;
2. remove or alter the developer attribution;
3. redistribute, copy, forward, or share this software with unauthorized users;
4. use this software outside the authorized research or teaching context.

If you need to share or reuse this software, please obtain permission from Dr. Delun Gong first.

Keywords

Digital image correlation, virtual extensometer, strain extraction, engineering strain, true strain, materials testing, tensile testing, image sequence analysis, Origin-compatible TXT.