Image Metadata Viewer (IMetVi)

IMetVi is a cross-format desktop application for extracting, standardizing, displaying, exporting, and curating scientific image metadata. It targets researchers, librarians, and research technicians working with microscopy, remote sensing, medical, astronomical, and general-purpose image files in Canadian academic institutions.

See ARCHITECTURE.md for how the codebase is organized and UPDATES.md for a chronological changelog.

Features

• 🖼️ Supported formats, each mapped to a discipline-aligned standard:

  Format	Extensions	Standard
  TIFF	.tif .tiff	REMBI (microscopy) or EXIF (general photo/scan/illustration)
  CZI (Zeiss)	.czi	REMBI (microscopy)
  OME-TIFF	.tif .tiff	OME-XML
  GeoTIFF	.tif .tiff	ISO 19115 (geospatial)
  JPG / JPEG	.jpg .jpeg	EXIF + IPTC + XMP
  PNG	.png	EXIF-style general fields
  DICOM	.dcm	DICOM PS3.3 (patient-identifying fields excluded by design)
  FITS	.fits .fit	FITS / WCS (astronomy)
  HDF5	.h5 .hdf5 .nc4	Generic container inspection
  LIF (Leica)	.lif	REMBI (microscopy)
  NetCDF	.nc .nc4	ISO 19115 via CF Conventions
  LAS/LAZ (LiDAR)	.las .laz	ISO 19115 / ASPRS LAS

  Formats sharing an extension (TIFF / OME-TIFF / GeoTIFF all use .tif/.tiff; HDF5 / NetCDF both cover .nc4) are selected explicitly via the format dropdown, which auto-suggests a match on load. TIFF is a general-purpose container, not just a microscopy format — the Select Application dropdown offers both Microscopy and General / EXIF for it, so a scanned figure or illustration gets EXIF-style fields (camera, date, resolution) instead of REMBI microscopy fields.

• 🛰️ LAS/LAZ point classification analysis (on demand) — beyond the header fields every LAS/LAZ file shows on load, a LiDAR sidebar action reads the full point records to report point classification breakdown (ground/vegetation/building/water/etc.), scan angle range, GPS time range, and flight-line count. Kept separate from the normal load path since it's slower (reads every point); sensor model and flying height are not stored in LAS/LAZ files at all, so the app says so rather than guessing. The sidebar section only appears when the loaded file's format is LAS — it stays out of the way for every other format instead of sitting there greyed out.

• 📁 Load a single file or an entire folder (batch loading runs off the UI thread with a progress bar).

• 🇨🇦 Bilingual EN/FR interface — a language selector switches every button, label, tooltip, dialog, and error/success message between English and French, including the dynamically rendered Curation summary. (Extracted metadata field labels and standards documentation text are not yet translated — see ARCHITECTURE.md.)

• 💬 Tooltips on every sidebar action — hover any button to see what it does before clicking it.

• 🌳 Readable CRS WKT — a CRS's WKT definition (often 1000+ characters on one line) is shown indented by nesting depth in the Recommended Fields tab instead of one unbroken wall of text. The value exported to JSON/CSV/sidecar files is unaffected — this only changes the on-screen display.

• 🖼️ Collapsible thumbnail preview — confirms visually which file is loaded (JPG/PNG/TIFF, first page for multi-page TIFF); other formats show a generic extension placeholder.

• 🧾 Three-tab metadata view:

  • Raw Metadata — everything the parser extracted, unfiltered.
  • Recommended Fields — standardized, discipline-aligned fields with human-readable labels and units.
  • Curation — automated flags and integrity info for this file.

• ℹ️ Metadata Standard Info dialog — for the active context, shows what the target standard requires and honestly reports what is (and isn't) captured from the file itself.

• 🚩 Curation flags, computed per batch:

  • DUPLICATE — identical MD5 checksum to another file in the batch
  • CORRUPT — parser reported a read failure
  • HAS_GPS_DATA — GPS metadata present (privacy/consent flag)
  • DIMENSION_OUTLIER — dimensions deviate from the batch's most common size
  • LOSSY_TIFF — TIFF using internal JPEG compression
  • NO_CRS_FOUND — a georeferenced format (GeoTIFF/NetCDF/LAS) with no embedded coordinate system

• ⚠️ Missing required fields — each format declares the standardized fields a well-formed capture is expected to have; gaps are shown in red at the top of the Recommended Fields tab and exported as a MissingFields column in the curation report

• ✍️ Write metadata back into the file (JPEG: EXIF + IPTC + XMP; TIFF: ImageDescription), with an editable form and an overwrite confirmation. Disabled (with an explanatory tooltip) for GeoTIFF and OME-TIFF, where writing would destroy georeferencing or the OME-XML structure — see ARCHITECTURE.md.

• 📎 Save Sidecar JSON — writes <basename>.json beside the source image without touching the original (QGIS/ArcGIS/repository convention).

• 🔒 Cross-session integrity verification — Save Checksums persists a checksums.json manifest per folder; Verify Integrity re-scans and reports OK / MODIFIED / MISSING / NEW per file.

• 📋 Batch Compliance Summary — aggregates missing-required-field and curation-flag counts across a whole loaded batch into one dataset-wide readiness snapshot ("N / M files fully compliant", ranked by which fields/flags are most common), so you can judge deposit-readiness without reading every file's Curation tab individually.

• 💾 Export:

  • Standardized metadata as JSON or CSV (single file or full batch)
  • Dedicated Curation Report CSV (one row per file, curation-focused columns, compatible with CUR_Res_CurationTools report shape)

Screenshot

Recommended Fields view for a loaded LAS file — note the CRS (WKT) field indented for readability, and the LiDAR sidebar section that only appears for this format

---

Installation

Requirements

• Python 3.8+
• Tested on Windows 10/11
• Recommended: run within a conda or venv environment

Dependencies

pip install -r requirements.txt

Launch the App

python main.py

Run the tests

python -m pytest -q

---

License

This project is licensed under the MIT License. See LICENSE for details.

---

Acknowledgements

• TIFF handling: tifffile
• Zeiss CZI reader: czifile
• Leica LIF reader: readlif
• EXIF read/write: Pillow + piexif
• IPTC read/write: iptcinfo3
• GeoTIFF / geospatial: rasterio
• HDF5: h5py
• NetCDF: netCDF4
• LiDAR / LAS-LAZ: laspy + pyproj
• DICOM: pydicom
• FITS: astropy
• REMBI Guidelines: https://doi.org/10.1038/s41592-021-01166-8
• OME-XML spec: https://www.openmicroscopy.org/ome-files/
• ISO 19115 (geospatial metadata): https://www.iso.org/standard/53798.html
• IPTC standard: https://www.iptc.org/std/photometadata/specification/
• ASPRS LAS specification: https://www.asprs.org/divisions-committees/lidar-division/laser-las-file-format-exchange-activities