SWIR or multispectral reflection example

This example will show decomposition of multispectral reflection data, such as SWIR imaging.

Planned workflow

  1. Load a multispectral TIFF stack.
  2. Load wavelength values or channel labels from wavelength.json.
  3. Define spectral seeds or import reference spectra.
  4. Run fixed-H NNLS or seeded NNMF.
  5. Inspect component maps.
  6. Export Fiji-compatible TIFF and spectral CSV files.

Points to highlight

  • Wavelength units in nm.
  • Custom channel labels.
  • Fixed-H NNLS with known spectra.
  • Reuse of presets across fields of view.
  • Application to mouse or cholesteatoma data.

Placeholder: insert representative SWIR or reflection dataset.

GIF placeholder: load wavelength metadata and run fixed-H NNLS.