Labkit

Labkit is a local image segmentation plugin for Fiji, aimed at researchers working with microscopy datasets that can be too large for conventional desktop workflows. It supports manual labeling and automated pixel classification, with a design that lets consumer laptops work with images spanning multiple terabytes.

It handles single-channel and multi-channel images, 2D images, 3D volumes and timelapse movies. Researchers can create ground-truth annotations by hand and curate segmentation results within the same interface. Labkit comes bundled with Fiji.

Automated segmentation uses a random forest classifier built on WEKA. Optional GPU acceleration can speed up classification, while a separate command-line tool can run the classification algorithm on HPC clusters for distributed processing of large datasets. Desktop work runs in Fiji; cluster processing runs on the computing infrastructure you use.

Its approach to large images rests on ImgLib2 and BigDataViewer, with memory-efficient pixel classification. That makes dataset size a central reason to consider it alongside its annotation tools, especially for microscopy work that combines large volumes with manual review.

Labkit is open source under the BSD-2-Clause license. Developers with their own segmentation algorithms can also use its Java interface as a front end rather than build a separate annotation and segmentation UI.