Welcome to Nornir’s documentation!

Nornir’s takes large sets of overlapping images in 2D and produces registered (a.k.a. aligned) 2D and 3D volumes of any size and scale.

The github repository for Nornir lives at http://github.com/nornir/


Nornir evolved from a collaboration between the Marc Lab and the Scientific Computing Institute’s Tasdizen and Whitaker groups at the University of Utah. The original tools, known as the NCR Toolset were used to construct RC1, a 250um diameter, 33um tall cylinder of rabbit retina at a resoultion of 2.18 nm/pixel.

Nornir is a work in progress. It currently supports importing images from transmission electron microscopes running SerialEM (.idoc files), light microscopes running Surveyor (.pmg files), and other volumes represented with a single image per section (.png files).

System requirements

  • 64-bit operating system, only tested on Windows at this time
  • 2GB of RAM per CPU core.

Indices and tables