scallops.visualize.utils.channel_thresholds

scallops.visualize.utils.channel_thresholds(image, percentile_min=0.0, percentile_max=99.0, pad_min=0, pad_max=0, thresholds=None)

Compute thresholds per channel for contrast adjustment in visualization.

Parameters:

  • image (DataArray) – XArray with dimensions (t,c,z,y,x) or (i,t,c,z,y,x)

  • percentile_min (Sequence[float] | float) – Determines the minimal threshold for contrast adjustment. If a list, apply different percentiles per channel. If a float, apply the same percentile for all channels.

  • percentile_max (Sequence[float] | float) – Determines the maximal threshold for contrast adjustment. If a list, apply different percentiles per channel. If a float, apply the same percentile for all channels.

  • pad_min (Sequence[float] | float) – Adjust the minimum value for visualization. If a list, apply different pads per channel. If a float, apply the same pad for all channels.

  • pad_max (Sequence[float] | float) – Adjust the maximum value for visualization. If a list, apply different pads per channel. If a float, apply the same pad for all channels.

  • thresholds (dict[int, tuple[float, float]]) – Optional precomputed dictionary mapping channel to tuple of (tmin, tmax).

Returns:

A dictionary that maps channel to tuple of (tmin, tmax) for contrast adjustment.

Example:
import numpy as np
import xarray as xr
from scallops.visualize.utils import channel_thresholds

# Create synthetic data
channels, z, y, x = 3, 10, 100, 100
synthetic_data = np.random.rand(channels, z, y, x)
synthetic_image = xr.DataArray(synthetic_data, dims=("c", "z", "y", "x"))

# Compute thresholds
thresholds = channel_thresholds(
    synthetic_image, percentile_min=1, percentile_max=99
)

print(thresholds)
Return type:

dict[int, tuple[float, float]]