seg_object_counts_match — SEGMENTATION `score` op

• Data kinds: labels2d × labels2d → table

• Call: import fullseye as fs; fs.ledger.seg_object_counts_match(labels_pred, labels_true, *, background: 'int' = 0, min_overlap: 'float' = 0.5) -> 'Dict[str, object]' (to call the implementation directly, import segeval; segeval.seg_object_counts_match(labels_pred, labels_true, *, background: 'int' = 0, min_overlap: 'float' = 0.5) -> 'Dict[str, object]'; from the registry, opssegmentation.get("seg_object_counts_match"))

Usage

Object-count agreement: counts split / merged / missed / false / matched from the degrees of a 'mainly overlapping' bipartite graph (seg_object_counts_match).

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

真 i と予測 j(どちらも背景でない)の間に辺を置く条件 = 重なり n_ij ≥ `min_overlap` × min(|i|, |j|)

(小さい方がもう一方に主に入っている)。真の次数 d_i、予測の次数 e_j:

`missed = #{d_i = 0}、false = #{e_j = 0}、split` = #{d_i ≥ 2}(1 つの真が複数の予測に割れた)、

`merged = #{e_j ≥ 2}(1 つの予測が複数の真を抱えた)、matched` = #{d_i = 1 かつその相手の e_j = 1}。

`counts_match` = 真と予測の個数が同じで全部が一致。min_overlap > 0.5 なら各予測は高々 1 つの真に「主に入る」が、

真が小さく予測が大きい辺もあるので融合は検出できる。

返り値: 上の数 + `n_true / n_pred、edges(真ラベル, 予測ラベル)の (m, 2)、split_true / merged_pred` /

`missed_true / false_pred(ラベルの配列)、min_overlap`。

Detailed usage guide

• halcon_segmentation family guide

Background guides (the physics and conventions behind this op)

• segmentation_scoring_and_worlds — 分割の採点と真値つき合成世界 — どの物差しがどの壊れ方に盲目か

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.

• The canonical algorithm (author, year) and its uses are named in the family usage guide above.

Runnable examples (verified samples that actually call this op)

• poc_graph_hierarchy_segmentation — py -3.11 examples/poc_graph_hierarchy_segmentation.py

Ops the type connects to (they accept `table` as input)

class_ndim_norm

Same category (`score`)

seg_confusion_table · seg_dice_jaccard · seg_boundary_f · seg_hausdorff · seg_mean_surface_distance · seg_under_over_segmentation · seg_score_card

*Provenance: segeval.py — SEGMENTATION operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*

seg_object_counts_match — SEGMENTATION score op