# Image segmentation#

Image analysist refer to image segmentation when subdividing an image into multiple groups of pixels having different characteristics. In this chapter we will learn basic algorithms for binarizing images, for labeling objects in images.

## Installation of requirements#

As in the chapters before, we will use scikit-image, pyclesperanto-prototype and napari-simpleitk-image-processing for segmenting the images. Some visualization will again be done using matplotlib.

## Installation of optional dependencies#

For some short-cuts to watershed-based image segmentation algorithms, installation of the scriptable napari plugin napari-segment-blobs-and-things-with-membranes is recommended. You can install it using pip:

pip install napari-segment-blobs-and-things-with-membranes