SAMJ-IJ

SAMJ-IJ brings local AI segmentation into ImageJ and Fiji for researchers annotating microscopy images. It turns point or rectangle prompts into object regions, so users can work with Segment Anything models without writing code or leaving their image analysis environment.

The plugin runs models on your workstation and doesn't require a GPU. Supported choices include SAM2 Tiny, SAM2 Small, SAM2 Large, EfficientSAM and EfficientViTSAM-L2. Model downloads need an internet connection. CPU performance affects responsiveness, and EfficientSAM can be demanding on low-resource computers; EfficientViTSAM-L2 is the lightest and fastest listed choice.

Annotations feed into Fiji's ROI Manager, where they remain available for further analysis. The plugin can return all detected regions or retain only the largest, and it exports label masks with a distinct value for each marked region.

For repeated analysis, SAMJ-IJ accepts prompts produced by other Fiji plugins and routines. ImageJ's Find Maxima can supply object points automatically, while macro support allows batch annotation across folders of images. That makes it useful for both hands-on annotation and existing automated workflows.

Applications include outlining astrocytes, measuring bacterial spread on agar plates, and segmenting organoids for counting and shape measurements. SAMJ-IJ is written in Java and is open source under the Apache License 2.0.