rank_kendall_b — IMGMETRICS iqa op

• Data kinds: any × any → scalar

• Call: import fullseye as fs; fs.ledger.rank_kendall_b(x, y, chunk: 'int' = 512) -> 'float' (to call the implementation directly, import iqatid; iqatid.rank_kendall_b(x, y, chunk: 'int' = 512) -> 'float'; from the registry, opsimgmetrics.get("rank_kendall_b"))

Usage

> This operator's description has not been translated yet. The original text follows as it is.

Kendall の τ_b(同順位補正あり)= (C − D) / sqrt((n0 − n1)(n0 − n2))。

C / D = 一致 / 不一致の対の数、n0 = n(n−1)/2、n1 / n2 = x 側 / y 側で同順位の対の数。O(n²) を `chunk` 行ずつの

放送で数える(n = 3000 で 4.5e6 対、数秒)。単調増加で 1、反転で −1。どちらかが定数なら `nan`。TID2013 の公表 Kendall は

τ_b(14 本が 3 桁一致、τ_a だと 4 本外れる)。±inf は順序だけ使う。Raises `ValueError`: 長さ不一致・2 未満・nan。

Detailed usage guide

• image_difference_metrics family guide

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_iqa_fsim_gmsd_vif — py -3.11 examples/poc_iqa_fsim_gmsd_vif.py

• poc_iqa_tid2013 — py -3.11 examples/poc_iqa_tid2013.py

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

luma_limited_u8 · rank_data · tid2013_index

Same category (iqa)

luma_limited_u8 · rank_data · rank_spearman · tid2013_published · tid2013_root · tid2013_index · tid2013_metric_values · tid2013_evaluate

*Provenance: iqatid.py — IMGMETRICS 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.