illumination op• 数据种类:无 → table(仅由参数决定的算子 —— 不接受图像或数据输入)
• 调用: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"))
标准机器视觉照明族的发光点集合(`table`)。
> 以下的详细说明为原文 —— 摘要与标题已翻译。
*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 的每个算子都先校验输入再计算(不让任何东西无声通过):
• 单位写进参数名 —— _mm / _um / _deg / _mrad。把 mm 和 µm 弄混不会崩溃,而是给出「看着合理却是错的答案」,所以用命名来防。这里绝不从数值大小去猜单位。
• **字符串一律 ValueError** —— float('50') 会成功,于是未解析的配置值会被当成长度混进来(实测:thin_lens('50', '200') 曾返回看着合理的 66.667 mm)。bool 也按 True == 1 的隐式提升拒绝。
• **complex / masked array 一律 ValueError(仅接受实数槽位;拒绝无声丢弃虚部或剥掉掩码)。所有输入中的 NaN/Inf 一律 ValueError**。
• 逐项点名拒绝除零及其近亲:焦距 0、曲率半径 0、折射率 <= 0、全不透明光阑(全为 0,归一化变成 0/0)、总和 <= 0 的 PSF、S0 = 0 的 Stokes 矢量、物体位于前焦点(像在无穷远)。
• 只有两个算子会返回非有限值,而且都写进了契约:depth_of_field 在超焦距以外返回 far_mm = inf(这正是超焦距的定义),gaussian_beam 在束腰处返回 wavefront_radius_mm = inf(平面波前的曲率半径)。两者都同时返回一个有限的搭档(far_is_infinite / curvature_per_mm)。**除此之外的无声 NaN/Inf 都在内部检出并 ValueError** ——「float64 溢出了」和「答案是无穷大」是两种不同的主张,不能拿后者的脸去交付前者。
• 尺寸上限:生成网格受 optics.MAX_GRID(4096)限制,传入的场/PSF/光阑受 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「小参数引发巨大内部分配」的路径(实测:n_max=40 × 4096² 需要 108 GB)。
• 物理上不可能的状态同样拒绝:偏振度 > 1 的 Stokes 矢量、负透过率、负强度、n-|m| 为奇数等非法 Zernike 指标。
• mv_illumination_practice — 照明の実務知識 — 波長・偏光・点灯方式・外光・安全
• 示例数据目录(下载 URL / 许可证) —— 2-D 用 skimage.data(BSD/公有领域)加合成图,3-D 给出真实数据源(Stanford/PDS 等)的下载 URL。
• 算子来历与参考文献 —— 该算子族所依据的研究/方法出处。
• 算法的正典(作者・年份)与用途见上面的族使用指南。
• illumination_design_demo — py -3.11 examples/illumination_design_demo.py
• virtual_machine_vision — py -3.11 examples/virtual_machine_vision.py
table 作为输入)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 算子登记表。本条目由 tools/opdocs.py md 自动生成(请勿手工编辑)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.