squidpy.experimental.tl.StitchParams

class squidpy.experimental.tl.StitchParams(distance_tol=0.75, min_edge_length=5.0, min_edge_length_ratio=0.4, min_edge_coverage=0.5, candidate_min_iou=0.2, close_radius=3)[source]

Advanced tuning knobs for assign_stitch_groups().

Defaults work for typical 2D segmentation tiles produced by cellpose-like pipelines. Pass an instance (or a Mapping of field names to values) as stitch_params to override. These are advanced knobs – the defaults rarely need changing.

Methods

Attributes

`distance_tol`	Sub-pixel tolerance for "lies on a bbox edge".
`min_edge_length`	Absolute floor on cut-edge length (pixels).
`min_edge_length_ratio`	Minimum cut-edge length relative to the cell's equivalent diameter.
`min_edge_coverage`	Minimum fraction of parallel-axis positions covered by near-edge contour points.
`candidate_min_iou`	Loose 1-D IoU floor at candidate enumeration.
`close_radius`	Morphological closing disk radius for the union mask.