tb_statistical_outlier_removal — 2D typed op

Data kinds: pointspoints

Call: fullseye.apply(img, "tb_statistical_outlier_removal", a=0.5, b=0.5) (the 2-D model is one image plus two scalar knobs a,b∈[0,1])

Usage

Removes points whose k-nearest-neighbor mean distance is a global outlier (statistical outlier removal).

> The detailed description below is the original text — the summary and the headings are translated.

点ごとに「最も近い k 個(自分自身は除く)までの平均距離」を測り、その全点分布の

`mean + std_ratio*std` を超える点を飛び点とみなして落とす。まばらな飛び点の掃除に

有効で、密な面上の点は残る。

Parameters

----------

points : array_like, shape (N, 3)

入力点群。

k : int

近傍数(既定 16)。点数が少なければ内部で `n-1` に丸める。

std_ratio : float

しきい値の緩さ。大きいほど残りやすい(除去が緩い)。

Returns

-------

filtered : ndarray, shape (M, 3)

生き残った点(元の順序を保持)。

keep_mask : ndarray of bool, shape (N,)

各入力点を残すか(True=残す)。`points[keep_mask]filtered` に等しい。

Notes

-----

点数 < 3 では統計が立たないため、全点を残す(graceful)。

2-D 進化レジストリへ橋渡しした 3d の op `statistical_outlier_removal。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。ak(既定 16)、bstd_ratio`(既定 2)を振る。

References (sample data, literature)

• 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.

Runnable examples (verified samples that actually call this op)

• (none yet)

Ops the type connects to (they accept points as input)

identity · tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_angle_3points · tb_project_points · tb_render_point_depth · tb_radius_outlier_removal

Same category (typed)

tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_angle_3points · tb_project_points · tb_render_point_depth · tb_radius_outlier_removal · tb_voxel_grid_downsample


*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.