polarization op• データ種: polsweep → stokes
• 呼び出し: import specularity; specularity.polarization_stokes(images, angles_deg=(0.0, 45.0, 90.0, 135.0), max_violation_frac=0.0) (または opsspecular.get("polarization_stokes"))
偏光板回転から、シーン全体を積分した Stokes ベクトル。→ (4,)。:func:optics.stokes_analyze にそのまま渡せる。
> 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。
Stokes parameters are linear in radiance, so the spatial mean of the frames
has a Stokes vector that is the mean of the per-pixel ones — the vector a
non-imaging polarimeter looking at the whole field would report. Fitting
`I(t) = 0.5 (S0 + S1 cos 2t + S2 sin 2t) gives (S0, S1, S2)` directly.
S3 is always exactly 0, and that is a limitation, not a measurement. A
set of linear analysers cannot see circular polarisation; a quarter-wave
plate is needed. The returned vector is therefore the linear part of the
truth, and :func:optics.stokes_analyze will report `handedness="linear"`
for it no matter what the scene actually did. Returning an invented `S3`
would be worse, and returning a 3-vector would break the Stokes contract the
optics family is built on.
The result satisfies that contract by construction — ``S0 >= sqrt(S1^2 +
S2^2 + S3^2)``, i.e. degree of polarisation at most 1 — because the same
non-negativity check that guards :func:polarization_separate is exactly
the condition for it. That is why `optics.stokes_analyze` accepts the
output without a further test.
Raises `ValueError: as :func:polarization_separate`.
• specular_photometric 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。
• specular_photometric — py -3.11 examples/specular_photometric.py
stokes を入力に取れる)—
polarization)polarization_render · polarization_separate · polarization_dolp_map
*Provenance: specularity.py — SPECULAR 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.