Published September 29, 2021 | Version 3.0.0
Software Open

seung-lab/kimimaro: Zenodo Release v1

  • 1. Princeton University
  • 2. Google BRAIN

Description

Rapidly skeletonize all non-zero labels in 2D and 3D numpy arrays using a TEASAR derived method. A skeleton is a stick figure 1D representation of a 2D or 3D object that consists of a graph of vertices linked by edges. A skeleton where the vertices also carry a distance to the nearest boundary they were extracted from is called a "Medial Axis Transform", which Kimimaro provides.

Designed for skeletonizing neurons at high resolution, Kimimaro has special support for centering somata and fixing split nuclei from the cell body.

Kimimaro also provides a facility for piecing together skeletons from neighboring chunks with single voxel overlap to enable extremely efficient skeletonization of large objects.

Files

seung-lab/kimimaro-3.0.0.zip

Files (9.0 MB)

Name Size Download all
md5:d57f88a20a97c8118c5f60f73e252f09
9.0 MB Preview Download

Additional details

Related works