ilastik
Interactive machine-learning image segmentation tool that lets non-programmers perform pixel classification.
- Main category
- Bioimage analysis
- Cost
- See the official site for cost
- Licence type
- Open source
- Platforms
- Windows / macOS / Linux
Version-watched: add it to favourites and confirmed new releases will show on your favourites page (stored in this browser only).
Compare with similar tools
Practical notes
Typical uses
- Pixel classification: label a few example pixels of each class and train a segmentation.
- Object classification, Autocontext, tracking, density counting and interactive carving.
Good for
- Microscopy images that simple thresholds cannot segment.
- Users willing to draw a few annotations but not to write code.
Less suited / consider instead
- Fully automatic analysis with no annotation: the main workflows rely on example labels from the user.
Technical limits and cautions
- Pixel classification needs sparse example labels for each class; object classification needs example objects assigned to classes (official docs).
- An official Fiji plugin converts anything Fiji can read into the HDF5 format ilastik prefers.
- On Windows the automatic tracking workflow needs the external CPLEX library (official docs).
Official getting-started docs: ilastik documentation
Compiled from the official pages below (read 2026-11-05); not run by us. The “less suited” items are our suggestions.
Details
ilastik is an interactive machine-learning image analysis tool where users train a model with a small number of annotations to perform pixel classification, object segmentation, object classification and tracking. It suits researchers who need image segmentation but lack deep learning programming experience.
Use cases: microscopy image segmentation, cell/particle recognition, agricultural image classification, semi-automated image analysis
