smlmlp.bkgd_temporal_median module

smlmlp.bkgd_temporal_median(channels, /, median_window_fr=25, bkgds=None, noise_corrections=None, *, cuda=False, parallel=False)[source]

Compute a temporal local median background for each channel.

This function estimates the background of each input channel by applying a temporal median filter along the frame axis. It also propagates the noise_corrections values so the output format remains consistent with other background estimation functions.

Parameters:

  • channels (sequence of ndarray) – Sequence of channel stacks. Each element is expected to be a temporal image stack and is processed independently.

  • median_window_fr (int, optional) – Temporal median window size, in frames.

  • bkgds (sequence of ndarray or None, optional) – Optional preallocated output arrays for the computed backgrounds. If provided and longer than the corresponding acquisition, each background stack is truncated to the channel length.

  • noise_corrections (sequence of float or None, optional) – Optional per-channel noise correction factors. If None, they are initialized to 1.0 for each channel. In this function, the values are preserved because the original logic multiplies them by 1.0.

  • cuda (bool, optional) – Whether to enable CUDA processing in stacklp.temporal_median().

  • parallel (bool, optional) – Whether to enable parallel processing in stacklp.temporal_median().

Returns:

A tuple (new_bkgds, noise_corrections, info) where:

  • new_bkgds is the list of computed background stacks,

  • noise_corrections is the updated list of correction factors,

  • info is a dictionary containing reusable intermediate results.

The dictionary contains the following keys:

'median_window_fr'

Temporal window size (in frames) used for the median filter.

Return type:

tuple

Examples

Use the same temporal median window for all channels:

>>> import numpy as np
>>> channels = [
...     np.random.rand(10, 16, 16).astype(np.float32),
...     np.random.rand(10, 16, 16).astype(np.float32),
... ]
>>> bkgds, noise_corr, info = bkgd_temporal_median(channels, median_window_fr=25)
>>> len(bkgds)
2
>>> len(noise_corr)
2
>>> info["median_window_fr"]
25

Reuse preallocated output arrays:

>>> out = [np.empty_like(ch) for ch in channels]
>>> bkgds, noise_corr, info = bkgd_temporal_median(
...     channels,
...     median_window_fr=11,
...     bkgds=out,
... )
>>> len(bkgds) == len(channels)
True

Provide existing noise corrections:

>>> init_corr = [np.float32(1.0), np.float32(2.0)]
>>> bkgds, noise_corr, info = bkgd_temporal_median(
...     channels,
...     median_window_fr=9,
...     noise_corrections=init_corr,
... )
>>> noise_corr[1]
2.0