spad_deadtime_correct — PHOTON `spad` op

• データ種: countrate → countrate

• 呼び出し: import photoncount; photoncount.spad_deadtime_correct(measured_hz, dead_time_ns=50.0) (または opsphoton.get("spad_deadtime_correct"))

不感時間で歪んだ測定レートから真の光子レートを復元する。

> 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。

The exact inverse of the non-paralysable law of

:func:spad_deadtime_apply::

n = m / (1 - m*tau)

A round trip `apply -> correct` is exact to machine precision (measured max

elementwise relative error 6.0e-16 over 2000 rates spanning 1e3 to 5e7 Hz at

`tau = 50 ns`, where the measured rate reaches 71.4% of the 20 MHz

saturation rate).

There is deliberately no paralysable inverse. `m = n*exp(-n*tau)` is not

injective — every measured rate below the maximum `1/(e*tau)` corresponds to

*two* true rates, one below and one above `1/tau` — so returning one of them

would be a fabrication dressed as a correction. Resolve the branch with an

independent measurement (e.g. an attenuator step) and invert it yourself.

*measured_hz* is a 1-D array of measured rates in counts per second;

*dead_time_ns* the dead time in nanoseconds (default 50, the same

placeholder :func:spad_deadtime_apply uses — replace it with the

datasheet value). Returns the corrected true rates as a float64 1-D array.

Raises `ValueError`: negative, non-finite or non-1-D *measured_hz*, a

non-positive *dead_time_ns*, and — instead of returning `inf` or a negative

rate — any measured rate at or above the saturation rate `1/tau`, which no

non-paralysable detector can ever produce.

• 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。

• photon_timeresolved — py -3.11 examples/photon_timeresolved.py

*Provenance: photoncount.py — PHOTON 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)。*

spad_deadtime_correct — PHOTON spad op