typed op• Data kinds: points → volume
• Call: fullseye.apply(img, "tb_region_growing", a=0.5, b=0.5) (the 2-D model is one image plus two scalar knobs a,b∈[0,1])
*No figure: this op takes points as input. A Studio program starting from an image cannot reach that type — see the runnable examples below for how it is used.*
Grows regions by normal similarity, assigning the same label to connected smooth regions (a variant without a curvature gate).
> The detailed description below is the original text — the summary and the headings are translated.
各点を k 近傍グラフ上で BFS 成長させ、隣接点 q を「法線 n_p と n_q の成す角が
`angle_thresh_deg` 未満」のときだけ同領域に加える。平面内の法線はほぼ平行なので
同一領域に連結し、向きの違う面の境界では角度が開いて連結が切れる → 面ごとに別領域。
法線は符号不定(PCA 由来)なので `|n_p·n_q|` で判定(表裏を同一視)。
Args:
points: (N,3) 点群。
normals: (N,3) 単位法線。None なら :func:pointcloud.estimate_normals で PCA 推定。
angle_thresh_deg: 隣接法線角度の許容上限[度]。(0,180) の範囲。
k: 近傍数(kNN グラフの次数)。
Returns:
labels: (N,) int。連結平滑領域ごとに 0,1,2,... を付与。**min_region_size 未満の
小領域(孤立点・向き不一致のゴミ)は -1(ノイズ/未割当)** = 統一契約(-1=ノイズ)に従う。
空入力は shape (0,) を返す。
2-D 進化レジストリへ橋渡しした 3d の op `region_growing。実装は同じで、呼び出し規約だけ op(v, a, b) に合わせてある。a が angle_thresh_deg(既定 15)、b が k`(既定 20)を振る。
• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).
• Operator provenance and references — the sources of the research/methods this op family came from.
• (none yet)
volume as input)identity · vol_gaussian · vol_median · vol_erode · vol_dilate · vol_threshold · vol_reg_dilate · vol_reg_erode
typed)tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_angle_3points · tb_project_points · tb_render_point_depth · tb_statistical_outlier_removal · tb_radius_outlier_removal
*Provenance: ops.py — 2D operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.