Segmentation with Fiji

What is segmentation?

Partitioning a digital image into multiple segments!

Segmentation

Three common assumptions/requirements:

1. Do you have overlapping objects? And do you care?

2. What are the shapes of your objects?

Tubes? Blobs? Clouds?

3. What modality was used? And how does the signal look?

Machine learning

Analysis of Aggregates with TWS

How to segment?

1. Preprocess

Prepare your image for thresholding

Find Edges | Kuwahara | Gaussian Blur | Median | Background Correction | Deconvolution

2. Threshold

Separate foreground from background

Global | Local | Available methods

3. Cleanup

Improve mask with morphological operations

Erode | Dilate | Open | Close | Skeletonize | Outline | Fill Holes | Watershed
MorphoLibJ | Morphological Segmentation | Distance Transform Watershed

4. Identify

Split foreground into objects

Analyze Particles | Extended Particle Analyzer

5. Analyze

Measure the original data

Set Measurements

Other techniques:

Real World Examples

Separating Objects
Skeletonization
Spatial Resolution
Uneven Illumination

Further Reading

B