segment op• 資料種類:voxel → labels
• 呼叫: import fullseye as fs; fs.ledger.vol_watershed(vol, markers, mask=None)(要直接呼叫實作,import volops; volops.vol_watershed(vol, markers, mask=None);從台帳取用則 ops3d.get("vol_watershed"))
標記控制的 3-D 分水嶺分割(可選 —— scikit-image)。
> 以下的詳細說明為原文 —— 摘要與標題已翻譯。
Floods the volume from the labelled *markers* (an int volume, `0` = unset),
optionally restricted to *mask*. Delegates to
`skimage.segmentation.watershed` — treat *vol* as a landscape (e.g. a
gradient magnitude, or the negated distance transform for splitting touching
blobs). Returns an `int32 (D, H, W)` label volume.
Raises `ImportError with a clear pip install scikit-image` message when
the optional dependency is absent; the rest of this module needs only
numpy + scipy.
• 範例資料目錄(下載 URL / 授權) —— 2-D 用 skimage.data(BSD/公有領域)加合成圖,3-D 給出真實資料源(Stanford/PDS 等)的下載 URL。
• 運算子來歷與參考文獻 —— 該運算子族所依據的研究/方法出處。
• molecule_atom_count — py -3.11 examples_3d/molecule_atom_count.py
labels 作為輸入)fuse_to_voxel · vol_region_props
segment)region_growing · euclidean_cluster · plane_segmentation
*Provenance: volops.py — 3D 運算子登記表。本條目由 tools/opdocs.py md 自動產生(請勿手動編輯)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.