lensrender optable → tableimport fullseye as fs; fs.ledger.render_ghost_overlay(system, scene=None, light_xy_mm=None, distance_mm=1000.0, intensity_W_sr_nm=1.0, reflectance=0.04, sensor=None, wavelengths=None, exposure_s=0.0001, rings=10, max_ghosts=6, defocus_mm=0.0, focus_mm=None, base_electrons=None, band_nm=None, illuminant=None) -> 'dict' (実装を直接呼ぶなら import lensrender; lensrender.render_ghost_overlay(system, scene=None, light_xy_mm=None, distance_mm=1000.0, intensity_W_sr_nm=1.0, reflectance=0.04, sensor=None, wavelengths=None, exposure_s=0.0001, rings=10, max_ghosts=6, defocus_mm=0.0, focus_mm=None, base_electrons=None, band_nm=None, illuminant=None) -> 'dict'、台帳から引くなら opsoptics.get("render_ghost_overlay"))The two-reflection ghosts of a bright point, as electrons on the sensor -> dict.
The point is the scene’s first source or, with light_xy_mm (x, y), a source at distance_mm in front of the lens with
spectral intensity intensity_W_sr_nm. raytrace3d.ghost_analysis gives, for every pair of refracting surfaces, the ghost’s
energy relative to the image (R_i R_j times the transmittances along its path, reflectance being one value for all
surfaces, or a raytrace3d coating spec) and the share of rays that reach the sensor; ghost_path gives the real rays of the strongest
max_ghosts ghosts, which are spread over the pixels with the power I Omega T energy_ratio of the point. A reflectance of
0 gives an overlay that is exactly zero. The table has overlay_electrons (H, W), electrons_with_ghosts (when
base_electrons, e.g. a render’s electrons_mean, is given), image_electrons (the point’s own image, for the ratio), ghosts (one
row per ghost: surfaces i, j, energy_ratio, reach_fraction, power_ratio = their product, the footprint centroid in
pixels (row, col) and its radius, and the offset from the main image in pixels), main_image_px and wavelengths_um. The point
must be on or off the axis of an axisymmetric system; the sensor plane follows defocus_mm / focus_mm as in the render.
optics の全 op は入力を検証してから計算する(黙って通さない):
_mm / _um / _deg / _mrad。mm と µm の取り違えは crash ではなく「もっともらしく間違った答え」なので、名前で防ぐ。大きさから単位を推測する処理は一切しない。ValueError — float('50') は成功してしまうため、未パースの設定値が長さとして通り抜ける(実測: thin_lens('50', '200') がもっともらしい 66.667 mm を返していた)。bool も True == 1 の暗黙昇格として拒否。ValueError(実数枠のみ。虚部の無言切り捨て・マスク剥がしを拒否)。NaN/Inf は全入力で ValueError。depth_of_field の過焦点距離以遠の far_mm = inf(それが過焦点距離の定義)と gaussian_beam のウエストでの wavefront_radius_mm = inf(平面波面の曲率半径)。どちらも有限の相棒(far_is_infinite / curvature_per_mm)を併せて返す。それ以外の無言 NaN/Inf は内部で検出して ValueError —「float64 が溢れた」と「答えが無限大」は別の主張なので、後者の顔で前者を返さない。optics.MAX_GRID(4096)、供給された場/PSF/開口は optics.MAX_FIELD_ELEMENTS(2^24)、ABCD 素子列は optics.MAX_SYSTEM_ELEMENTS(1024)、Zernike は MAX_ZERNIKE_TERMS(512)/ MAX_ZERNIKE_ORDER(40)/ MAX_ZERNIKE_BASIS(2^25)。小さな引数から巨大な内部確保が起きる経路(実測: n_max=40 × 4096² で 108 GB)を fail-closed で塞ぐ。| 物理的に不可能な状態も拒否: 偏光度 > 1 の Stokes ベクトル、負の透過率、負の強度、n- | m | が奇数などの不正な Zernike 添字。 |
py -3.11 examples/poc_photograph_through_prescription.pytable を入力に取れる)abcd_matrix · wavefront_stats · sfr_from_edge · paraxial_trace · seidel_coefficients · spot_stats · tolerance_analysis · wavefront_from_opd
lensrender)lens_scene · exit_pupil_samples · trace_sensor_rays · scene_radiance · render_scene_through_lens · bokeh_shape · render_compare_prescriptions
Provenance: lensrender.py — OPTICS 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.