illumination op• 資料種類:無 → table(僅由參數決定的運算子 —— 不接受影像或資料輸入)
• 呼叫:import illumdesign; illumdesign.light_source(kind='ring', radius_mm=60.0, height_mm=100.0, n=24, tilt_deg=None, length_mm=100.0, intensity=1.0, cos_exponent=1.0, position=None, emitters=None, directions=None)(或 opsoptics.get("light_source"))
標準機器視覺照明族的發光點集合(`table`)。
> 以下的詳細說明為原文 —— 摘要與標題已翻譯。
*kind*:
• `"point" — one emitter at *position* (default (0, 0, height_mm)`)
pointing straight down.
• `"ring"` — *n* emitters on a circle of *radius_mm* at *height_mm*,
each tilted toward the axis by *tilt_deg* (default: aimed at the origin,
so the ring's elevation angle is `atan(height/radius)`; low
height/radius = dark field, high = bright field).
• `"bar"` — *n* emitters along a line of *length_mm* parallel to x at
`y = radius_mm, z = height_mm`, aimed at the origin line.
• `"dome"` — *n* emitters spread over a hemisphere of *radius_mm*
(Fibonacci lattice, zenith angles 20°–85°), each pointing at the centre:
the diffuse, shadow-free illumination of a dome light.
• `"coaxial"` — a disc of *n* emitters of radius *radius_mm* at
*height_mm* pointing down: the diffuse area source a beam splitter folds
onto the viewing axis. Put it at the camera height with a radius of at
least twice the part size, so that every point of a flat glossy part
sees the source in its mirror direction (a *small* source at the camera
gives one bright spot, not a field-wide glare). The dict carries an
`area record so :func:defect_contrast` takes the specular glare from
the disc's uniform radiance `L = n I0/(π r²)` by a mirror-hit test
instead of a point sum that cannot resolve a narrow lobe.
• `"backlight" — *n* × *n* emitters on a square of side 2·radius_mm`
at `z = −height_mm` pointing up: the part is seen in silhouette.
• `"custom"` — *emitters* (N,3) and *directions* (N,3) given explicitly.
*intensity* is the radiant intensity of one emitter on its axis (arbitrary
units), *cos_exponent* the Lambertian order of its angular distribution
(1 = ideal Lambertian LED; ~2–4 for a lensed LED with a narrower beam).
optics 的每個運算子都先檢驗輸入再計算(不讓任何東西無聲通過):
• 單位寫進參數名 —— _mm / _um / _deg / _mrad。把 mm 和 µm 弄混不會當掉,而是給出「看似合理卻是錯的答案」,所以用命名來防。這裡絕不從數值大小去猜單位。
• **字串一律 ValueError** —— float('50') 會成功,於是未解析的設定值會被當成長度混進來(實測:thin_lens('50', '200') 曾回傳看似合理的 66.667 mm)。bool 也按 True == 1 的隱式提升拒絕。
• **complex / masked array 一律 ValueError(僅接受實數槽位;拒絕無聲丟棄虛部或剝掉遮罩)。所有輸入中的 NaN/Inf 一律 ValueError**。
• 逐項點名拒絕除零及其近親:焦距 0、曲率半徑 0、折射率 <= 0、全不透明光闌(全為 0,正規化變成 0/0)、總和 <= 0 的 PSF、S0 = 0 的 Stokes 向量、物體位於前焦點(像在無窮遠)。
• 只有兩個運算子會回傳非有限值,而且都寫進了契約:depth_of_field 在超焦距以外回傳 far_mm = inf(這正是超焦距的定義),gaussian_beam 在束腰處回傳 wavefront_radius_mm = inf(平面波前的曲率半徑)。兩者都同時回傳一個有限的夥伴(far_is_infinite / curvature_per_mm)。**除此之外的無聲 NaN/Inf 都在內部檢出並 ValueError** ——「float64 溢位了」和「答案是無窮大」是兩種不同的主張,不能拿後者的臉去交付前者。
• 尺寸上限:生成網格受 optics.MAX_GRID(4096)限制,傳入的場/PSF/光闌受 optics.MAX_FIELD_ELEMENTS(2^24),ABCD 元件序列受 optics.MAX_SYSTEM_ELEMENTS(1024),Zernike 受 MAX_ZERNIKE_TERMS(512)/ MAX_ZERNIKE_ORDER(40)/ MAX_ZERNIKE_BASIS(2^25)。以 fail-closed 堵住「小參數引發巨大內部配置」的路徑(實測:n_max=40 × 4096² 需要 108 GB)。
• 物理上不可能的狀態同樣拒絕:偏振度 > 1 的 Stokes 向量、負穿透率、負強度、n-|m| 為奇數等非法 Zernike 指標。
• mv_illumination_practice — 照明の実務知識 — 波長・偏光・点灯方式・外光・安全
• 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。
• illumination_design_demo — py -3.11 examples/illumination_design_demo.py
• virtual_machine_vision — py -3.11 examples/virtual_machine_vision.py
table 作為輸入)abcd_matrix · wavefront_stats · paraxial_trace · seidel_coefficients · spot_stats · tolerance_analysis · wavefront_from_opd · spot_diagram
illumination)irradiance_map · illumination_uniformity · defect_contrast · lighting_sweep · illumination_design
*Provenance: illumdesign.py — OPTICS 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.