GUIs and Visualization — MIDAS

Version: 11.0 Contact: hsharma@anl.gov

Note

v11 updates: FF viewer: independent colormap dropdowns per plot, all Grains.csv columns as color options, CBF file format support, zoom preservation on log scale toggle, and broader session restore coverage for data-source, display, and overlay state. NF viewer: right-click for spot selection (left-click = zoom), cyan crosshair markers, auto-compute distances on "Finished", visual results dialog with crop patches + ray triangulation diagram, box profile auto-refresh with pixel coordinates, orange-red edge lines, cursor tracking + colorbar on mic scatter map, defaults to highest-res map, zoom preserved on plot type switch.

Overview

MIDAS provides desktop GUIs for interactive data exploration, calibration, and real-time monitoring. Two generations of GUIs are available:

Generation	Framework	Benefits
Legacy (Tkinter)	Tkinter + Matplotlib	Mature, full feature set, familiar toolbar
Modern (PyQtGraph)	PyQt5 + PyQtGraph	Faster rendering, richer interaction, navigation toolbar

Tip

The PyQtGraph-based viewers (ff_asym_qt.py, nf_qt.py) are recommended for daily use. They offer faster image rendering, a built-in navigation toolbar, and a more responsive interface.

graph LR
    subgraph "Desktop GUIs (gui/)"
        FFQ["ff_asym_qt.py<br>FF Viewer (PyQtGraph)"]
        NFQ["nf_qt.py<br>NF Viewer (PyQtGraph)"]
        FF["ff_asym.py<br>FF Viewer (Tkinter)"]
        NF["nf.py<br>NF Viewer (Tkinter)"]
        DT["dt.py<br>DT Viewer"]
        IM["imageManipulation.py<br>Image Tools"]
    end
    subgraph "Web / Dashboard (utils/)"
        IFP["interactiveFFplotting.py<br>FF Interactive Browser"]
        PFV["pfIntensityViewer.py<br>PF Intensity Viewer"]
        LV["live_viewer.py<br>Live Integration Dashboard"]
    end
    subgraph "Plotting Utilities (utils/)"
        PIP["plot_integrator_peaks.py<br>Peak Analysis"]
        PCR["plot_calibrant_results.py<br>Calibrant QC"]
        SCT["sino_cleanup_tomo.py<br>Sinogram + Tomo"]
    end

1. FF-HEDM Viewer — PyQtGraph (`gui/ff_asym_qt.py`)

Launch:

cd <data_directory>
python ~/opt/MIDAS/gui/ff_asym_qt.py &

Features

Auto-detection of data files from current directory
Image display with PyQtGraph: fast pan, zoom, crosshair with pixel coordinates
Navigation toolbar below image: Home, Back/Forward (view history), Pan, Zoom-to-rect
Intensity control: P2–P98 percentile auto-scaling, editable Min I / Max I fields (in toolbar), log scale
Ring overlays from ring material database with live redraw on BC/Lsd change
Image transforms: HFlip, VFlip, Transpose checkboxes
Dark subtraction: load and apply dark field images
HDF5 support: browse H5 datasets with customizable data path
Frame navigation: spinner + Ctrl+mouse-wheel frame stepping
Max/Sum over frames: aggregate multiple frames for improved signal (mutually exclusive toggles)
ZIP auto-initialization: loading a MIDAS .zip automatically reads Lsd, BC, PixelSize, Wavelength, SpaceGroup, LatticeConstant, and ImTransOpt (HFlip/VFlip/Transpose)
Auto ring computation: when a ZIP is loaded, clicking Rings Material skips the parameter dialog and computes rings directly from ZIP metadata
Nearest ring in status bar: when rings are displayed, status bar shows the nearest ring's number and hkl at the cursor position
Colormap and theme (dark/light) selection
Font size control (8–24pt)
Export PNG of current view
🎬 Movie mode: Play/Pause/Stop + FPS slider (1–30) — animates through frames automatically
🖱️ Drag-and-drop: drop a file or folder onto the viewer to open it
📊 Intensity vs Frame: dock widget showing mean/max intensity across all frames, click to jump (View → Intensity vs Frame)
💾 Session save/restore: File → Save/Load Session as .session.json (Ctrl+S / Ctrl+Shift+S), restoring HDF5 paths, dark and mask settings, geometry, display limits, detector mode, and overlay state

Key Differences from Legacy (`ff_asym.py`)

Feature	Legacy (Tkinter)	Modern (PyQtGraph)
Rendering speed	Matplotlib (slower)	PyQtGraph (GPU-accelerated)
Navigation	Matplotlib toolbar	Custom Home/Back/Forward/Pan/Zoom
Mouse-wheel zoom	Enabled by default	Disabled (explicit Zoom button)
Intensity scaling	Manual min/max	P2–P98 auto-scaling + manual override (in toolbar)
Ring update	Requires reload	Live on BC/Lsd text field change
Calibrate dialog	Available	Removed (use `AutoCalibrateZarr.py`)

Keyboard Shortcuts

Key	Action
← / →	Previous / Next frame
L	Toggle log scale
R	Toggle ring overlay
Q	Quit
Ctrl+Q	Quit
Ctrl+S	Save session
Ctrl+Shift+S	Load session

1b. Caked Peak Viewer (`gui/viewers/plot_caked_peaks.py`)

Interactive viewer for _caked_peaks.h5 output from IntegratorZarrOMP or fit_caked_peaks.py.

Launch:

python ~/opt/MIDAS/gui/viewers/plot_caked_peaks.py /path/to/results/

# With lattice parameter analysis:
python ~/opt/MIDAS/gui/viewers/plot_caked_peaks.py /path/to/results/ \
    -paramFN refined_params.txt

Features

Caked heatmap: 2θ × η with colormap, interactive crosshair
1D profile: intensity vs 2θ at selected η bin, with fitted peak overlays
Peak data table: peak center, area, FWHM, sig, gam, η_mix, d-spacing, χ²
Lattice parameter mode (with --paramFN): adds Ring, a(Å), Δa/a₀ columns to table, and two bottom plots:
- Lattice parameter a vs η
- Relative lattice strain Δa/a₀ vs η
Ring selection checkboxes: toggle visibility of individual rings (only detected rings shown)
Font size and marker size spinners for the lattice plots
UI font spinner for all Qt widgets (labels, table, etc.)
Toggle overlays: Raw, BG, Corrected, Fit, Peak markers, Log scale

Caked Peak Fitting (`utils/fit_caked_peaks.py`)

Standalone Python peak fitter for _caked.hdf.zarr.zip output:

python ~/opt/MIDAS/utils/fit_caked_peaks.py \
    /path/to/scan.caked.hdf.zarr.zip \
    -paramFN refined_params.txt

Produces _caked_peaks.h5 with per-η-bin fitted peaks using GSAS-II-convention pseudo-Voigt profiles. The chi-squared rejection filter is normalized by peak area to prevent intensity-dependent rejection.

2. NF-HEDM Viewer — PyQtGraph (`gui/nf_qt.py`)

Launch:

cd <data_directory>
python ~/opt/MIDAS/gui/nf_qt.py &

Features

All features from the FF viewer plus:

Microstructure overlay: load .mic / .map files, color by Confidence, GrainID, Euler angles, KAM, GROD, Phase
Spot simulation: load grain parameters and simulate diffraction spots
LineoutH / LineoutV: horizontal and vertical intensity profiles
Box H / Box V ROI tools: rectangular region-of-interest with real-time Sum, Mean, Min, Max statistics
Beam Center determination: guided workflow for multi-distance calibration
Origin at bottom-right (matching physical NF detector convention)
🎬 Movie mode, 🖱️ drag-drop, 💾 session save — same as FF viewer

For detailed NF-specific calibration workflows, see NF_GUI.md.

3. Legacy Tkinter GUIs (`gui/`)

FF-HEDM Viewer (`gui/ff_asym.py`)

cd <data_directory> && python ~/opt/MIDAS/gui/ff_asym.py &

Full-featured FF viewer with Matplotlib rendering. See FF_Visualization.md.

NF-HEDM Viewer (`gui/nf.py`)

cd <data_directory> && python ~/opt/MIDAS/gui/nf.py &

Full-featured NF viewer with calibration workflow. See NF_GUI.md.

Diffraction Tomography Viewer (`gui/dt.py`)

python ~/opt/MIDAS/gui/dt.py &

2D detector image viewer with ring overlays and 1D intensity profiles.

Image Manipulation (`gui/imageManipulation.py`)

python ~/opt/MIDAS/gui/imageManipulation.py &

General-purpose detector image viewer: dark subtraction, flat-field, ROI selection, transformations, histograms.

4. Interactive Web Dashboards (`utils/`)

FF-HEDM Interactive Plotting (`gui/viewers/interactiveFFplotting.py`)

Browser-based interactive exploration of FF-HEDM results using Plotly Dash.

python ~/opt/MIDAS/gui/viewers/interactiveFFplotting.py <Grains.csv>

See FF_Interactive_Plotting.md.

PF-HEDM Intensity Viewer (`gui/viewers/pfIntensityViewer.py`)

Interactive sinogram, intensity patch, and tomographic reconstruction viewer for Point-Focus HEDM.

python ~/opt/MIDAS/gui/viewers/pfIntensityViewer.py

See PF_Interactive_Plotting.md.

Live Integration Dashboard (`gui/viewers/live_viewer.py`)

Real-time visualization of GPU integrator output. Tails lineout.bin and fit.bin as they are written by IntegratorFitPeaksGPUStream.

python ~/opt/MIDAS/gui/viewers/live_viewer.py \
    --lineout lineout.bin --nRBins 500 \
    --fit fit.bin --nPeaks 3 \
    --params setup_30keV.txt

Three-panel display:

1D Lineout — latest integrated I(R) profile with log scale option
Heatmap — time-evolving intensity waterfall (R × frame)
Peak evolution — fitted Imax, Rcen, Sigma, η, BG, GoF, Area vs. frame for each peak

Interactive Features:

Triple x-axes (R, 2θ, Q) on both lineout and heatmap — enabled via --params or --lsd/--px/--wavelength
🎯 Peak Pick — click on peaks to select R values, then 📤 Send Peaks to the running GPU integrator (Replace or Append mode)
Active fit overlays — red dashed lines at currently-fitted radii on both plots

See FF_Radial_Integration.md §4.2.1 for details on the interactive peak selection protocol.

5. Plotting Utilities (`utils/`)

Peak Analysis (`gui/viewers/plot_integrator_peaks.py`)

Post-hoc Pseudo-Voigt peak fitting on caked zarr.zip output. Produces 2D scatter plots of fitted 2θ vs η with ring assignment.

python ~/opt/MIDAS/gui/viewers/plot_integrator_peaks.py \
    --zarr scan_01.caked.hdf.zarr.zip \
    --peaks 245.3 347.1 425.8

Calibrant QC (`gui/viewers/plot_calibrant_results.py`)

Quick lattice-parameter-vs-η scatter plot from CalibrantPanelShiftsOMP _corr.csv output, with ideal lattice overlay.

Auto-discovers every *corr.csv in the current directory, so running AutoCalibrateZarr produces three selectable flavors in the file dropdown:

<rawFN>.corr.csv — calibrant per-(ring, η) residuals from the final M-step.
integrator_<stem>.corr.csv — the same format, but produced by the independent IntegratorZarrOMP validation pass (post-calibration).
compare_<stem>.corr.csv — bin-by-bin join of the two files keyed on (RingNr, η), carrying the original 18 calibrant columns plus StrainCalib, StrainInteg, StrainDiff, and StrainDiffMicroStrain (already in µε). Pick any Strain* column as the Y or Color axis to inspect calibrant-vs-integrator agreement.

The viewer also has a Caked checkbox (mutually exclusive with Polar). When enabled, it loads the matching <rawFN>.caked.hdf (copied back from the integrator work dir during validation) and renders OmegaSumFrame/LastFrameNumber_0 as a grayscale imshow with η on X and R (pixels) on Y, then overlays the currently-loaded corr.csv's (Eta, RadFit) points colored by the active Color column. The Log checkbox switches the imshow intensity to log-scale.

Sinogram Cleanup and Tomo Reconstruction (`utils/sino_cleanup_tomo.py`)

Processes PF-HEDM sinograms: column normalization, hole filling, despeckling, gap interpolation, 2D Gaussian smoothing, and MIDAS_TOMO reconstruction.

python ~/opt/MIDAS/utils/sino_cleanup_tomo.py \
    --topdir /analysis/pf_output \
    --sinoType raw --filterNr 2

See Tomography_Reconstruction.md for the reconstruction backend.

6. Shared Components (`gui/gui_common.py`)

The PyQtGraph-based viewers share a common library providing:

Component	Description
`MIDASImageView`	Image viewer with crosshair, auto-levels, navigation toolbar, axis origin control
`apply_theme()`	Dark/light palette for Qt + PyQtGraph
`AsyncWorker`	Background thread wrapper with progress/error signals
`LogPanel`	Redirects `print()` to a collapsible log dock widget
`get_colormap()`	Name-based colormap lookup with matplotlib fallback

Navigation Toolbar

Built into MIDASImageView, displayed below the image:

Button	Action
🏠 Home	Reset to full image extent
◀ Back	Previous view in history
▶ Forward	Next view in history
📋 Pan	Toggle: left-drag to move
🔍 Zoom	Toggle: left-drag rectangle to zoom
▶ Play	Animate frames at set FPS
⏸ Pause	Pause animation
⏹ Stop	Stop animation, reset mode
FPS	1–30 frames per second spinner

Mouse-wheel zoom is disabled by default (optimized for SSH/remote display where wheel events cause expensive redraws). Use the Pan and Zoom toolbar buttons instead. Double-click anywhere on the image resets to full view (same as Home).

Ctrl+scroll wheel scrolls through frames.

7. Requirements

PyQtGraph GUIs

PyQt5
pyqtgraph
numpy

Legacy Tkinter GUIs

tkinter (built-in)
matplotlib
numpy
Pillow

Web Dashboards

dash / plotly
flask
pandas
numpy

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

GUIs and Visualization — MIDAS

Overview

1. FF-HEDM Viewer — PyQtGraph (`gui/ff_asym_qt.py`)

Features

Key Differences from Legacy (`ff_asym.py`)

Keyboard Shortcuts

1b. Caked Peak Viewer (`gui/viewers/plot_caked_peaks.py`)

Features

Caked Peak Fitting (`utils/fit_caked_peaks.py`)

2. NF-HEDM Viewer — PyQtGraph (`gui/nf_qt.py`)

Features

3. Legacy Tkinter GUIs (`gui/`)

FF-HEDM Viewer (`gui/ff_asym.py`)

NF-HEDM Viewer (`gui/nf.py`)

Diffraction Tomography Viewer (`gui/dt.py`)

Image Manipulation (`gui/imageManipulation.py`)

4. Interactive Web Dashboards (`utils/`)

FF-HEDM Interactive Plotting (`gui/viewers/interactiveFFplotting.py`)

PF-HEDM Intensity Viewer (`gui/viewers/pfIntensityViewer.py`)

Live Integration Dashboard (`gui/viewers/live_viewer.py`)

5. Plotting Utilities (`utils/`)

Peak Analysis (`gui/viewers/plot_integrator_peaks.py`)

Calibrant QC (`gui/viewers/plot_calibrant_results.py`)

Sinogram Cleanup and Tomo Reconstruction (`utils/sino_cleanup_tomo.py`)

6. Shared Components (`gui/gui_common.py`)

Navigation Toolbar

7. Requirements

PyQtGraph GUIs

Legacy Tkinter GUIs

Web Dashboards

See Also

FilesExpand file tree

GUIs_and_Visualization.md

Latest commit

History

GUIs_and_Visualization.md

File metadata and controls

GUIs and Visualization — MIDAS

Overview

1. FF-HEDM Viewer — PyQtGraph (gui/ff_asym_qt.py)

Features

Key Differences from Legacy (ff_asym.py)

Keyboard Shortcuts

1b. Caked Peak Viewer (gui/viewers/plot_caked_peaks.py)

Features

Caked Peak Fitting (utils/fit_caked_peaks.py)

2. NF-HEDM Viewer — PyQtGraph (gui/nf_qt.py)

Features

3. Legacy Tkinter GUIs (gui/)

FF-HEDM Viewer (gui/ff_asym.py)

NF-HEDM Viewer (gui/nf.py)

Diffraction Tomography Viewer (gui/dt.py)

Image Manipulation (gui/imageManipulation.py)

4. Interactive Web Dashboards (utils/)

FF-HEDM Interactive Plotting (gui/viewers/interactiveFFplotting.py)

PF-HEDM Intensity Viewer (gui/viewers/pfIntensityViewer.py)

Live Integration Dashboard (gui/viewers/live_viewer.py)

5. Plotting Utilities (utils/)

Peak Analysis (gui/viewers/plot_integrator_peaks.py)

Calibrant QC (gui/viewers/plot_calibrant_results.py)

Sinogram Cleanup and Tomo Reconstruction (utils/sino_cleanup_tomo.py)

6. Shared Components (gui/gui_common.py)

Navigation Toolbar

7. Requirements

PyQtGraph GUIs

Legacy Tkinter GUIs

Web Dashboards

See Also

1. FF-HEDM Viewer — PyQtGraph (`gui/ff_asym_qt.py`)

Key Differences from Legacy (`ff_asym.py`)

1b. Caked Peak Viewer (`gui/viewers/plot_caked_peaks.py`)

Caked Peak Fitting (`utils/fit_caked_peaks.py`)

2. NF-HEDM Viewer — PyQtGraph (`gui/nf_qt.py`)

3. Legacy Tkinter GUIs (`gui/`)

FF-HEDM Viewer (`gui/ff_asym.py`)

NF-HEDM Viewer (`gui/nf.py`)

Diffraction Tomography Viewer (`gui/dt.py`)

Image Manipulation (`gui/imageManipulation.py`)

4. Interactive Web Dashboards (`utils/`)

FF-HEDM Interactive Plotting (`gui/viewers/interactiveFFplotting.py`)

PF-HEDM Intensity Viewer (`gui/viewers/pfIntensityViewer.py`)

Live Integration Dashboard (`gui/viewers/live_viewer.py`)

5. Plotting Utilities (`utils/`)

Peak Analysis (`gui/viewers/plot_integrator_peaks.py`)

Calibrant QC (`gui/viewers/plot_calibrant_results.py`)

Sinogram Cleanup and Tomo Reconstruction (`utils/sino_cleanup_tomo.py`)

6. Shared Components (`gui/gui_common.py`)