Package sc.fiji.snt.seed
package sc.fiji.snt.seed
Classes for Seed Tracing.
- Author:
- Tiago Ferreira
-
ClassDescriptionHelper that turns a labels image (e.g. cellpose, Labkit, StarDist segmentation output) into a list of
SeedPoints, one per non-zero label.Shared helper for turning a detector's raw per-candidate scores (integrated density, local contrast, cluster density, ...) intoSeedPoint.confidencevalues in[0, 1], orNaNwhen there is no basis to judge one candidate against another.Transient, in-memory store ofSeedPoints used to render candidate tracing seeds (e.g., the output of a deep-learning point detector) on SNT's canvas and to feed them to downstream consumers (autotracers, click-to-trace anchors).HowSeedOverlayRenderershould pick each seed's LUT slot.Thrown bySeedOverlay.parseCsv(File)when the CSV header is missing a required columnResult ofSeedOverlay.parseCsv(File): valid seeds plus a count of rows skipped due to parse errors (missing/non-finite coordinates, etc)Listener fired on any change to the overlay (add/clear/threshold/visibility).Bdv/Bvv counterpart ofSeedOverlayCanvasHandler.Attaches a lightweightMouseListenerto aTracerCanvasthat lets the user click on a rendered seed to select it (mirroringSeedOverlayBigViewerHandler's plain-click behavior for Bvv/Bdv.Selection flows throughSeedOverlay.setSelectedSeeds(java.util.Collection<? extends sc.fiji.snt.seed.SeedPoint>)so the Seed Manager table stays in sync), or Alt+Click it to open the per-seed edit dialog.Immutable candidate seed point used to bootstrap auto-tracing or as a visualization aid for the output of upstream point detectors (e.g., deep-learning probability maps + local-max suppression).Modal seed editor with two modes: Single: One seed, all nine fields (X, Y, Z, confidence, radius, channel, frame, type, source) are shown and editable; Save rebuilds an immutableSeedPointand applies it viaSeedOverlay.replaceAt(int, SeedPoint); Delete removes the seed.BidirectionalSeedPoint-Roiconverter.TableModelbacked by aSeedOverlay.