LabelsToObjects¶
LabelsToObjects converts a 2D label image or a TYX label stack into object dataframe rows.
Input is label_image.
Outputs are frame, label, y, x, area, and object_count.
No objects CSV artifact is written.
Use it as the bridge from segmentation to tracking.
Label 0 is ignored.
Dependencies and Core Libraries¶
BioImageFlow core APIs, imageio, and NumPy.
Minimal Example¶
from bioimageflow_core import Arguments
from bioimageflow_tracking_tools import LabelsToObjects
LabelsToObjects().process_row(Arguments(label_image="labels_tyx.tif"))
Expected Results¶
The workflow dataframe contains one row per non-zero label per frame with centroid and area columns. All-background inputs produce an empty object table with the declared columns; no sentinel object row is created.
Failure Modes¶
Non-2D/non-3D inputs raise ValueError; missing or unsupported image files fail through imageio.