DistanceWatershedSegment

DistanceWatershedSegment thresholds an image into foreground, computes a distance transform, places markers on local distance peaks, and runs skimage.segmentation.watershed.

Inputs are input_image, threshold, and min_distance. Outputs are labels and object_count.

Minimal Example

from bioimageflow_core import Arguments
from bioimageflow_segmentation_tools import DistanceWatershedSegment

result = DistanceWatershedSegment().process_row(
    Arguments(input_image="mask.tif", labels="labels.tif", threshold=0.5)
)

Inputs

• input_image: binary-like or intensity image.

• threshold: foreground cutoff.

• min_distance: minimum marker spacing in pixels.

Outputs

• labels: watershed label image.

• object_count: number of watershed regions.

Dependencies and Core Libraries

imageio, NumPy, SciPy distance transforms, and scikit-image watershed helpers.

Assumptions

Foreground objects are brighter/non-zero and have separable distance peaks.

Expected Results

Synthetic touching disks split into distinct labels when marker spacing is appropriate.

Failure Modes

Unreadable images, invalid marker settings, unsupported dimensions, and write failures raise errors.