Basic image processing

!pip install git+https://github.com/guiwitz/napari-skimage.git
!pip install napari-skimage-regionprops

import napari
from napari.utils import nbscreenshot
from napari_skimage.skimage_filter_widget import median_filter_widget
from napari_skimage.skimage_threshold_widget import threshold_widget
from skimage import data

# Load a sample 3D grayscale image (like cells)
viewer = napari.Viewer()
sample_image = data.cells3d()[30, 1]  # This will load a sample grayscale image from skimage
image_layer = viewer.add_image(sample_image, name='sample_cells')

# Create the widgets without parameters
laplace_widget = median_filter_widget()
threshold_widget_instance = threshold_widget()

# Add the widgets to the viewer's dock
viewer.window.add_dock_widget(laplace_widget, area='right')
viewer.window.add_dock_widget(threshold_widget_instance, area='right')

# Capture the screenshot for the notebook
screenshot = nbscreenshot(viewer)
screenshot

Downloading file 'data/cells3d.tif' from 'https://gitlab.com/scikit-image/data/-/raw/2cdc5ce89b334d28f06a58c9f0ca21aa6992a5ba/cells3d.tif' to '/home/runner/.cache/scikit-image/0.24.0'.

What other tools does napari-skimage expose?

• filters

• thresholding

• binary morphology (erosion, dilation, opening, closing)

• image math

• rolling ball operations

• find maxima

• marching cubes

Activity

Use napari-skimage to do a vanilla segmentation of the cells image and display their statistics:

• Use your favorite filter (median filter maybe?)

• Use your favorite thresholding method (perhaps Otsu?)

• Get the connected components

• Use napari-skimage-regionprops (Note that this is under the Tools menu as Measurement Tables>Regionprops)

• Turn on the Show selected checkbox from the Labels layer controls