Package sc.fiji.snt.tracing.auto

package sc.fiji.snt.tracing.auto

Automatic neuron tracing algorithms that reconstruct complete neuronal morphologies from images without user interaction.

This package provides implementations of whole-neuron auto-tracing algorithms, as opposed to the interactive point-to-point A*-based tracers in the parent sc.fiji.snt.tracing package.

Key Classes

• AutoTracer - Common interface for all auto-tracers, defining ROI strategy constants and the traceTrees() contract
• AbstractAutoTracer - Base class for grayscale-based tracers with shared soma ROI handling and graph utilities
• GWDTTracer - APP2-style tracer using Gray-Weighted Distance Transform and hierarchical pruning
• BinaryTracer - Skeleton-based tracer using topological thinning and AnalyzeSkeleton

Algorithm Categories

Grayscale-based (extends AbstractAutoTracer)

Operate directly on intensity images without binarization. Use geodesic distance transforms and intensity-weighted path finding.

Skeleton-based (BinaryTracer)

Require binarization and topological skeletonization. Convert skeleton graphs to neuronal trees. Suitable for high-contrast images with clear foreground/background separation.

Soma ROI Strategies

All auto-tracers support configurable soma handling via AutoTracer constants:

• ROI_UNSET - Ignore soma ROI, root at algorithm-specific point
• ROI_EDGE - Split into separate trees per neurite exiting soma
• ROI_CENTROID - Collapse soma nodes to ROI geometric centroid
• ROI_CENTROID_WEIGHTED - Collapse to weighted centroid of soma nodes
• ROI_CONTAINED - Root on nodes inside ROI (skeleton-based only)

Usage Example

 // Grayscale-based tracing
 GWDTTracer<?> tracer = GWDTTracer.create(imagePlus);
 tracer.setSeed(somaCenter);
 tracer.setSomaRoi(somaRoi, AutoTracer.ROI_CENTROID);
 List<Tree> trees = tracer.traceTrees();

 // Skeleton-based tracing
 BinaryTracer tracer = new BinaryTracer(binaryImage);
 tracer.setRootRoi(somaRoi, AutoTracer.ROI_EDGE);
 List<Tree> trees = tracer.traceTrees();
 

Author:

Tiago Ferreira

See Also:

• sc.fiji.snt.tracing
• Tree

Related Packages

Package	Description
sc.fiji.snt.tracing	Path-finding algorithms for interactive neuron tracing.
sc.fiji.snt.tracing.auto.gwdt	Storage backends for the Gray-Weighted Distance Transform (GWDT) tracing algorithm.
sc.fiji.snt.tracing.artist	Artist classes provide visual feedback for tracing algorithm execution.
sc.fiji.snt.tracing.cost	Cost function implementations for classes implementing SearchInterface
sc.fiji.snt.tracing.heuristic	Heuristic implementations for classes implementing SearchInterface
sc.fiji.snt.tracing.image	Data structures used to keep track of encountered nodes during image-based searches for classes implementing SearchInterface

Class	Description
AbstractAutoTracer	Abstract base class for grayscale-based automatic neuron tracers.
AbstractGWDTTracer<T extends net.imglib2.type.numeric.RealType<T>>	Abstract base class for GWDT-based tracers: APP2-style neuron tracer using Gray-Weighted Distance Transform and Fast Marching.
AbstractGWDTTracer.HierarchySegment	Represents a hierarchical segment as defined by APP2.
AbstractGWDTTracer.WaypointBiasSource	How the per-waypoint bias amount is derived.
AutoTraceConfig	Derives AbstractGWDTTracer configuration parameters from one or more example paths.
AutoTracer	Interface for automatic neuron tracers that reconstruct complete neuronal morphologies from images.
AutoTracer.SeedRole	Role a SeedPoint plays in a tracing run.
BinaryTracer	Class for generation of Trees from a skeletonized image.
ComponentReconnector<T extends net.imglib2.type.numeric.RealType<T>>	Bridges gaps between disconnected components of an auto-traced neuron tree.
DiskBackedGWDTTracer<T extends net.imglib2.type.numeric.RealType<T>>	Disk-backed GWDT tracer for very large images.
GWDTTracer<T extends net.imglib2.type.numeric.RealType<T>>	In-memory GWDT tracer using array storage.
GWDTTracerFactory	Factory for creating optimal GWDT tracer implementations based on image size.
SomaUtils	Utilities for automatic soma detection and ROI generation.
SomaUtils.SomaResult	Container for soma detection results.
SparseGWDTTracer<T extends net.imglib2.type.numeric.RealType<T>>	Sparse GWDT tracer using hash map storage for memory efficiency.