patchdist_score — SPC anomaly_patch op

• 資料種類:image2d × table → table

• 呼叫: import fullseye as fs; fs.ledger.patchdist_score(image, model, smooth=4.0)(要直接呼叫實作,import anomaly_patch; anomaly_patch.patchdist_score(image, model, smooth=4.0);從台帳取用則 opsspc.get("patchdist_score"))

用法

將一張影像與逐位置的常態分佈比較:由各圖塊的馬氏距離產生異常圖(table)。

> 以下的詳細說明為原文 —— 摘要與標題已翻譯。

model comes from patchdist_fit; image must have the training images' shape. The grid of distances sqrt((x - mu)' P (x - mu)) is interpolated (bilinear, between patch centres) to the image size and smoothed by a Gaussian of smooth px (0 = no smoothing). Returns map (H, W; higher = more anomalous), grid (the raw distances), image_score (the maximum of the smoothed map) and smooth.

Raises ValueError for a model that is not a patchdist_fit result, a different image shape, or a negative / non-finite smooth.

參考(範例資料、文獻)

• 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。

• 演算法的正典(作者、年份)與用途見上面的族使用指南。

可執行的範例(實際呼叫該運算子並已驗證的樣例)

• (尚無)

型別可銜接的下一個運算子(可接受 table 作為輸入)

msa_anova_table · msa_gauge_rr · msa_bias_linearity · msa_attribute_agreement · gum_standard_uncertainty · gum_propagate · gum_expanded · gum_monte_carlo

同類別(anomaly_patch)

patch_features · patchdist_fit · patchbank_fit · patchbank_score

*Provenance: anomaly_patch.py — SPC 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.