typed opcounts → countsfullseye.apply(img, "tb_spad_deadtime_apply", a=0.5, b=0.5) (2-D は 1 画像 + 2 スカラつまみ a,b∈[0,1] のモデル)
図は合成の入力 128×128 で実際に走らせた出力。左が入力、右が出力。点群は上から見た散布(明るさ = z)、1-D 列は折れ線、体積は z 方向の最大値投影、動画は中央フレーム、複素画像は振幅、絵にならない返り値は値そのもの。
つまみ a を振る(0.1 / 0.5 / 0.9、もう一方は既定):

つまみ b は出力を変えない(実測: 0.1 / 0.5 / 0.9 で同一)。
段階(前置きの op → この op。左から順):

別の画像でも(合成シーン / 写真 / 硬貨。上段が入力、下段がその出力。つまみは既定):

Distort a true photon rate by the detector’s dead time (counts lost).
After every detection a SPAD is blind for a recharge (dead) time tau, so
the measured rate m is always below the true incident rate n.
Two classical laws, and this op implements both:
m = n / (1 + n*tau). Monotonic, saturating at 1/tau.paralyzable=True) — an arriving photon restarts the
dead time: m = n * exp(-n*tau). This law peaks at n = 1/tau
(where m = 1/(e*tau)) and then falls, so a bright scene can read
darker than a dim one. That is why no inverse op exists for it (see
:func:spad_deadtime_correct).rate_hz is a 1-D array of true rates in counts per second; dead_time_ns is the dead time in nanoseconds, defaulting to 50 — the middle of the 10-100 ns range a passively quenched SPAD occupies, and a placeholder to be replaced by the datasheet value, never a measurement of your detector. Returns the measured rates as a float64 1-D array of the same length.
Ground truth (pinned in the tests): at n = 1/tau the non-paralysable law
gives exactly n/2; the paralysable law’s maximum is exactly
1/(e*tau) at n = 1/tau; both reduce to m = n as n*tau -> 0.
Raises ValueError: negative, non-finite or non-1-D rate_hz, a
non-positive dead_time_ns, and a non-bool paralyzable.
Typed bridge of the photon op spad_deadtime_apply into the 2-D evolution registry: the same implementation, called under the op(v, a, b) convention. a drives dead_time_ns (default 50); b is unused.
下のプログラムは実際に走ることを確かめてある(図と同じ入力)。Studio のヘルプではこのブロックがボタンになり、その場で読み込んで実行できる。
img_to_counts 0.50 0.50
tb_spad_deadtime_apply 0.50 0.50
次の例は元の台帳 op spad_deadtime_apply を呼ぶもの。この橋渡し op は同じ実装を fn(v, a, b) 規約に合わせただけなので、挙動はそのまま当てはまる(呼び出し形だけ違う)。
py -3.11 examples/photon_timeresolved.pypy -3.11 examples/poc_dtof_ranging.pycounts を入力に取れる)identity · tb_spad_deadtime_correct · tb_tcspc_coates_correct · tb_tcspc_irf_convolve · tb_tcspc_background_subtract · tb_dtof_depth · tb_countrate_to_counts · tb_counts_to_countrate
typed)tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_project_points · tb_render_point_depth · tb_points_render · tb_statistical_outlier_removal · tb_radius_outlier_removal
Provenance: ops.py — 2D 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.