envelope op• 数据种类:sweep → signal
• 调用:import interferometry; interferometry.csi_envelope(signal, remove_bias=True)(或 opsinterferometry.get("csi_envelope"))
z 扫描干涉图的相干包络(解析信号的模)。
> 以下的详细说明为原文 —— 摘要与标题已翻译。
`dsp.envelope computes |hilbert(x)|` for any 1-D signal and is called
here verbatim, so the two can never drift apart. The one thing added is the
one thing an interferogram needs and a generic signal does not: **removal of
the intensity pedestal**. An interferogram is `a + b*V(z)*cos(...)` with
`a > 0, and the analytic-signal magnitude of that is not b*V` — the DC
term passes through the Hilbert transform untouched and dominates.
Measured on the module's reference scan (`a = 0.5, b = 0.4`, envelope
sigma 1.2 um, surface centred): with the bias removed the envelope matches the
analytic `b*V(z)` to 1.83e-07; without it, the error is 0.5 — the
entire pedestal — and the recovered "envelope" never goes near zero.
`dsp.envelope` called directly on the same raw interferogram gives that
identical 0.5, which is the honest statement of what this operator adds. That
is why `remove_bias defaults to True; False` is available for a scan
whose pedestal you have already removed some other way, and it is your
statement that you did.
signal: 1-D scan intensities (`(n,)`, n >= 3).
remove_bias: subtract the scan mean before the transform.
Returns a 1-D float64 envelope of the same length.
Raises `ValueError`: a non-1-D, empty, too-short (< 3), non-finite,
complex or masked *signal*, a *signal* over :data:MAX_SCAN_POINTS elements
(checked before the float64 promotion), or a non-bool *remove_bias*.
• 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。
• coherence_scanning — py -3.11 examples/coherence_scanning.py
signal 作为输入)—
envelope)—
*Provenance: interferometry.py — INTERFEROMETRY 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.