The Napari Assistant

2 minute read

If you haven’t installed Napari, please do so before proceeding with this tutorial. If you just want to practice this tutorial, install the following minimal installation:

conda create -n napari_bioimage_analysis -y python=3.9 mamba
conda activate napari_bioimage_analysis
mamba install -y napari napari-assistant

Introduction

The Napari Assistant pluginꜛ is a collection of tools for many common image processing tasks. It comes with a graphical user interface and is a very handy tool for exploring and applying the image processing methods, that we have discussed so far, are yet to discuss and beyond.

The Assistant widget can be opened via the Tools->Utilities->Assistant menu and will look as follows:

The Napari Assistant widget.

The Assistant offers a wide range of image processing methods, covering

noise removal,
background subtraction,
image filters,
image maths (combine, transform, project),
image segmentation and feature extraction, and
mesh transformation.

When you click on the desired image processing function, a new widget will open together with a new layer in the layer viewer. As usual, any image processing task in Napari is performed in a non-disruptive way, i.e., the original image is not modified. Instead, the results of the processing is added as a new layer. The new widget allows you to adjust the corresponding parameters of the selected function:

Example of a widget for adjusting the parameters of a selected image processing function.

Creating image processing workflows

The Assistant is in fact designed for building image processing workflows. I.e., you can set-up your own pipeline of image processing steps simply by selecting and adjusting the corresponding functions. And when your pipeline is ready, you can save the workflow as a YML file and re-use it as often as you like:

The Assistant let’s you save and reload entire image processing workflows as YML files…

You can export your workflow as a Python script or Jupyter notebook as well:

…as well as Python scripts or Jupyter notebooks.

Acknowledgements

The Napari Assistant was developed by by Robert Haase (Websiteꜛ, Mastodonꜛ, Twitterꜛ, GitHubꜛ) and Ryan Savill (Blogꜛ, Twitterꜛ). Check out their websites as they contain many very useful and illustrative examples of image processing methods. And by the way, many of the plugins used in this course have been developed by Robert Haase.

Introduction

Creating image processing workflows

Further readings

Acknowledgements