Skip to content

NF_GUI.md

Latest commit

 

History

History
223 lines (169 loc) · 10.3 KB

NF_GUI.md

File metadata and controls

223 lines (169 loc) · 10.3 KB

Near-Field HEDM GUI

Version: 11.0
Contact: hsharma@anl.gov

This manual covers the PyQtGraph-based NF viewer (nf_qt.py).

DETAILED INSTRUCTIONS FOR CALIBRATION ARE AT: NF_Calibration.md

NF Viewer: nf_qt.py (PyQtGraph)

Launching

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

Features

graph LR
    NF["nf_qt.py"]
    NF --> Viz["Image Viewer<br/>(PyQtGraph)"]
    NF --> Nav["Navigation Toolbar<br/>Home/Back/Fwd/Pan/Zoom"]
    NF --> ROI["Box H / Box V<br/>ROI Stats"]
    NF --> Mic["Microstructure<br/>.mic / .map"]
    NF --> Sim["Spot Simulation<br/>GrainDialog"]
    NF --> Cal["Calibration<br/>Beam Center / Select Spots<br/>/ Compute Distances"]
    
    Viz --> Cross["Crosshair +<br/>pixel coords"]
    Viz --> Levels["P2-P98 Auto<br/>+ ManualI"]
    Viz --> Log["Log Scale"]
    Viz --> CM["Colormaps +<br/>Dark/Light Theme"]
Loading
Feature Description
Navigation toolbar Home, Back/Forward, Pan, Zoom-to-rect below the image. Mouse-wheel zoom disabled
Auto-detection Scans current directory for TIFF files, sets folder/stem/frame automatically
Intensity control P2–P98 auto-scaling with editable Min I / Max I fields and Apply button (in toolbar)
Log scale Checkbox for log₁₀ display
Median subtraction Toggle per-frame background subtraction
Max/Sum over frames Max/Frames and Sum/Frames checkboxes (mutually exclusive)
Image transforms HFlip, VFlip, Transpose checkboxes
Box H / Box V ROI Rectangular region with live Sum, Mean, Min, Max statistics. Auto-refreshes on frame/distance change
Box profile edge lines Half-maximum edges shown in orange-red with center/width in image pixel coordinates
Microstructure Load .mic/.map, color by Confidence, GrainID, Euler, KAM, GROD, Phase, GrainMap
Select Point Click on mic map to auto-populate grain parameters for simulation
Spot simulation GrainDialog: enter grain parameters, simulate expected spots
Select Spots Interactive workflow to pick diffraction spots at multiple distances. Right-click to select spots while maintaining zoom
Compute Distances Automatic ray triangulation on Finished — visual dialog with spot crop patches, triangulation ray diagram, per-pair Lsd/grainY output
Calc Median Compute median background from all frames per distance. Auto-enables Subt. Median and reloads on completion. Output stored in temp directory
Partial spot data Skip distances during Select Spots — only confirmed (≥2) spots are used for triangulation
Beam Center Dialog to enter beam center values for each distance
Colormap viridis, inferno, plasma, magma, turbo, gray, hot, cool, bone
Theme Dark or light
Font size Adjustable 8–24pt
Export PNG Save current view
Movie mode Play/Pause/Stop with configurable FPS
Drag-and-drop Drop TIFF files or folders to load
Session save/restore Ctrl+S / Ctrl+Shift+S

Keyboard Shortcuts

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

Prerequisites

PyQt5
pyqtgraph
numpy
tifffile

User Guide

1. Loading Data

  1. Launch the Application: Run python ~/opt/MIDAS/gui/nf_qt.py & from your data directory.
  2. Auto-Detection: When launched from a data directory, the GUI automatically sets Folder, File Stem, and Start Frame by scanning for .tif files. The image loads automatically.
  3. Load Initial File (alternative): Click the First File button to open a file dialog and manually select your first .tif image.

BeamPos / DetZBeamPos Folder Mode

When the GUI is launched from a folder whose name contains BeamPos or DetZBeamPos, it activates a special navigation mode:

  • All .tif files in the folder are collected and sorted by their numeric suffix.
  • The Frame spinner becomes an index (0, 1, 2, …) into this sorted list.
  • The spinner steps through files sequentially, regardless of file stem.
  • Median background subtraction is not available in this mode.

2. Image Display and Navigation

The left panel shows the detector image.

  • Toolbar: Use the navigation buttons (Home, Back/Fwd, Pan, Zoom) above the image.
  • Intensity Control:
    • Auto-scaling: Images are auto-scaled to the P2–P98 percentile range.
    • MinI / MaxI: Override intensity range manually, then click Apply.
    • Log: Check this box for log₁₀ display.
  • Frame stepping: Use the Frame spinner, ← / → keyboard keys, or Ctrl+mouse-wheel.

3. Part I: Determining the Beam Center

This procedure finds the (x, y) pixel coordinates of the beam center at each detector distance.

  1. Generate Horizontal Profile:
    • Click the Box H button.
    • A rectangular ROI appears on the image. Drag it over a horizontal diffraction line.
    • The right panel shows an integrated intensity profile summed along the vertical axis.
  2. Generate Vertical Profile:
    • Click the Box V button.
    • Drag the ROI over a vertical feature.
    • The right panel shows an integrated intensity profile summed along the horizontal axis.
  3. Find the Center:
    • Hover your mouse over the profile plot. The coordinates are shown in the status bar.
    • Identify the center of the slope/peak for both horizontal and vertical profiles.
    • Average the left and right edges to get the beam center (x, y).
  4. Enter Beam Center Values:
    • Click the Beam Center button.
    • Enter the (x, y) beam center for each distance.
    • Enter the known distance difference between detectors (μm).
    • Click Confirm.
  5. Repeat for other detector distances using the Distance spinner.

4. Part II: Determining Detector Position

  1. Enable Background Correction (Recommended):
    • Check Subt. Median. To compute a new median, click Calc Median.
  2. Select Spots:
    • Click the Select Spots button. A guide appears. Click Ready!
    • Navigate to a clear diffraction spot using Frame spinner and Dist spinner.
    • Use left-click-drag to zoom into the area of interest (rectangle zoom stays active).
    • Right-click on the spot center. A cyan crosshair appears showing where you clicked.
    • Click Confirm Selection in the status bar.
    • A dialog asks for the next distance. Enter the number and click Load, or click Finished if done.
    • Find the same spot on the next distance. Right-click to select and confirm.
    • You may skip distances — at least 2 confirmed spots are needed.
    • After clicking Finished, distances are auto-computed and a visual results dialog appears.
  3. Results Dialog:
    • Shows crop patches of each confirmed spot (100×100 px window from the image at confirmation time).
    • Shows a ray triangulation diagram with rays from the computed sample position through each spot.
    • Displays per-distance spot positions (Y, Z, R), per-pair Lsd/grainY values, and mean Lsd.
    • The mean Lsd is auto-filled in the main window.

5. Microstructure Analysis and Simulation

5.1 Loading Reconstruction Results

  1. Load Mic File: Click Load Mic and select your reconstruction output file.

    Format Extension Rendering Notes
    Binary map .map imshow (fast) Preferred format.
    Text mic .mic scatter (slower) Sparse point-based.

  2. Visualize Data: Use the radio buttons (Conf, GrainID, Eu0/1/2, KAM, GROD, GrMap, Phase) to change coloring.

5.2 Select Point: Auto-Populate Grain Parameters

  1. Click the Select Point button.
  2. Click on a grain of interest in the microstructure map.
  3. The grain's orientation and position are auto-populated in a Load Grain dialog. Click Confirm.
  4. Click Make Spots to simulate diffraction spots for that grain.

5.3 Diffraction Spot Simulation

  1. Manual Entry: Click Load Grain and enter grain parameters manually.
  2. Click Make Spots. A red circle shows the predicted spot position. Frame navigation auto-jumps to the correct rotation angle. A blue star marks the beam center.

6. Technical Implementation Details

6.1. Software Architecture

  • Framework: PyQt5 for window management and controls; PyQtGraph for high-performance image rendering.
  • Image rendering: Uses ImageItem with set_data() for efficient frame-by-frame updates.
  • Shared components: Uses gui_common.py for MIDASImageView, LogPanel, AsyncWorker, colormap utilities, and theme management shared with the FF viewer.

6.2. Simulation Backend

The Make Spots simulation uses C binaries:

  • GetHKLList — Generates reciprocal lattice vectors.
  • GenSeedOrientationsFF2NFHEDM — Transforms orientation matrix.
  • SimulateDiffractionSpots — Computes expected spot positions.

6.3. Distance Calibration Algorithm

The ComputeDistances function uses ray triangulation based on the intercept theorem. For each pair of detector positions (i, j), the Lsd is computed as:

$$LSD_0 = \frac{R_i \cdot (j-i) \cdot \Delta d}{R_j - R_i} - (i \cdot \Delta d)$$

The final distance is the arithmetic mean across all pairs.

7. Legacy Viewer

The legacy Tkinter-based viewer (nf.py) has been archived to gui/archive/nf.py. All functionality is available in the current Qt-based viewer. The legacy viewer is preserved for reference only.

8. See Also

If you encounter any issues or have questions, please open an issue on this repository.