model op• データ種: simmodel × images → simmodel
• 呼び出し: import fullseye as fs; fs.ledger.simcal_observe(model=None, images=None, roi=None, align=False, exposure=None, max_shift=8, split='train') -> 'dict' (実装を直接呼ぶなら import simcal; simcal.simcal_observe(model=None, images=None, roi=None, align=False, exposure=None, max_shift=8, split='train') -> 'dict'、台帳から引くなら opssimcal.get("simcal_observe"))
実測の画像を simmodel に付ける: 位置合わせ・関心領域・露出の正規化。
> 以下の詳細説明は原文のままです —— 要約と見出しは訳出済み。
images — one 2-D array per condition (shape = what the model outputs). align=True registers each image to the simulation at the initial knobs by phase correlation (filters_freq.phase_correlation_fft, integer pixels), shifts it back and masks the wrapped border; a shift larger than max_shift is refused. roi = boolean mask or (row0, row1, col0, col1). exposure = a factor per image to divide by (the noise variance is divided by its square) — refused when the model has a free gain knob: the two scale the same signal. split labels the data train / holdout / locked (the TRAIN / OBSERVED / LOCKED discipline of robust.py). Returns a new simmodel with observed filled (the input is not modified), including data_range.
• sim_to_real_calibration ファミリ ガイド
• サンプルデータ カタログ(DL URL / ライセンス) — 2-D は skimage.data(BSD/public)+ 合成、3-D は実データ源(Stanford/PDS 等)の DL URL。
• 演算子の来歴・参考文献 — この op 族の元になった研究/手法の出典。
• アルゴリズムの正典(著者・年)と用途は上記ファミリ使い方ガイドに記載。
• poc_pocket_interreflection — py -3.11 examples/poc_pocket_interreflection.py
• poc_simcal_thin_film — py -3.11 examples/poc_simcal_thin_film.py
• simcal_checks_tour — py -3.11 examples/simcal_checks_tour.py
• simcal_tour — py -3.11 examples/simcal_tour.py
simmodel を入力に取れる)simcal_simulate · simcal_screen · simcal_fit · simcal_validate
model)simcal_model · simcal_simulate
*Provenance: simcal.py — SIMCAL operator registry. この per-op ノートは tools/opdocs.py md が自動生成(手編集しない)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.