The problem it solves
Thresholded objects can merge, split, or include background artifacts, which makes automatic counts and size measurements unreliable unless masks are reviewed.
Who it is for
- Researchers quantifying particles, pores, droplets, or cells in images
- Microscopy laboratories reviewing segmentation before measurement
- Students learning how method choice affects object statistics
How it works
- Load a representative image and choose manual, watershed, or available Cellpose analysis.
- Tune the segmentation and inspect object masks and boundaries against the source.
- Correct or exclude unsuitable objects, then review size and shape measurements for export.
Output you can keep
A reviewed segmentation overlay or mask with per-object count, size, and shape results in the formats available in the active release and plan.
Limitations to check
- Segmentation quality depends on contrast, focus, object overlap, and parameter choice.
- Cellpose or other assisted methods may require an available processing service or local engine.
- Every mask should be reviewed; an automatic boundary is not ground truth.