Issues

• Public pixel-free check of to_scalar options (preset, white, stain vectors, polarizer)

  CellMigrationLab/morphobricks-normalization#97 · Filter by author guijacquemet opened on Nov 14, 2026

• Public check-only helper for A_SCOPE_INCOMPLETE (channel presence in a group), without fitting

  CellMigrationLab/morphobricks-normalization#96 · Filter by author guijacquemet opened on Nov 13, 2026

• RGB-to-scalar stain presets (H&E first) and later RGB colour normalization: normalization owns the conversion

  CellMigrationLab/morphobricks-normalization#94 · Filter by author guijacquemet opened on Nov 12, 2026
  2

• Design: streamed (block-by-block) intensity statistics for very large images

  CellMigrationLab/morphobricks-normalization#73 · Filter by author guijacquemet opened on Nov 6, 2026

• [Sad Catapella N8][Performance infrastructure] Establish reproducible kernel, pooling and end-to-end benchmarks before optional Numba backend

  CellMigrationLab/morphobricks-normalization#60 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella N5][Performance] CSV reader and writer materialize every plane and duplicate wide histogram arrays

  CellMigrationLab/morphobricks-normalization#49 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella N3][Performance] Pixel-array fitting copies each complete channel with np.take before histogram computation

  CellMigrationLab/morphobricks-normalization#40 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella N2][Performance/Numba] Pooling concatenates and sorts every occupied bin from every plane

  CellMigrationLab/morphobricks-normalization#34 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella N1][Performance/Numba] Optimize multi-quantile radix selection without repeated whole-block scans

  CellMigrationLab/morphobricks-normalization#30 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella N1][Performance/Numba] Sparse wide-integer images allocate a large per-value table solely based on intensity span

  CellMigrationLab/morphobricks-normalization#29 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella B7][Performance] Leave-one-out outlier and repeat-shift checks recompute medians at quadratic cost

  CellMigrationLab/morphobricks-normalization#24 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella B7][Scientific QC] Zero MAD silently suppresses large image outliers and repeat shifts

  CellMigrationLab/morphobricks-normalization#22 · Filter by author guijacquemet opened on Nov 1, 2026

• [Sad Catapella B7][Scientific QC] Repeat-shift detection pseudoreplicates FOVs and weights biological repeats by image count

  CellMigrationLab/morphobricks-normalization#21 · Filter by author guijacquemet opened on Nov 1, 2026